Explore Your Local Site

Looks like you've landed on our   site. Let's take you home:    

Please note that the content and products on the    site might not be available in your region.

 

Choose the language:

  Homepage
Continue on the current website:  

 

La guida completa all'intelligenza artificiale in radiologia

L’intelligenza artificiale (IA) sta svolgendo un ruolo crescente nella vita di tutti noi e si è dimostrata promettente nell’affrontare alcune delle più grandi sfide sociali attuali e future. Il settore sanitario, sebbene notoriamente complesso e resistente ai cambiamenti, ha potenzialmente molto da guadagnare dall’uso dell’IA. Avendo già rivestito un ruolo di guida della trasformazione digitale nel settore sanitario, e in virtù dell’urgente necessità di migliorare l’efficienza, la radiologia è stata in prima linea nello sfruttare le potenzialità dell’IA.

Nella presente pubblicazione si spiega come e perché l'IA potrebbe servire ad affrontare molte delle principali problematiche dei reparti di radiologia, si fornisce una descrizione generale dei concetti di base dell'IA e si descrivono alcuni degli usi dell'IA in radiologia ritenuti più promettenti. Inoltre, si analizzano le principali difficoltà associate all’adozione dell’IA stessa nelle prassi radiologiche di routine. La pubblicazione tratta anche alcuni aspetti cruciali di cui i reparti di radiologia dovrebbero tenere conto nel decidere quali soluzioni basate sull’IA acquistare. Infine, si fornisce una panoramica sulle novità e sugli aspetti che si prevede evolveranno nell’IA per la radiologia nel prossimo futuro.

Perché l’intelligenza artificiale in radiologia

Negli ultimi decenni, nel settore sanitario sono emerse varie tendenze che hanno mostrato l’esigenza di un cambiamento nel modo di fare determinate cose. Queste tendenze sono particolarmente rilevanti in radiologia, dove la qualità diagnostica delle immagini è migliorata notevolmente e i tempi di esecuzione degli esami sono diminuiti. Di conseguenza, la quantità e la complessità dei dati di imaging medico acquisiti sono aumentate in misura sostanziale negli ultimi decenni (Smith-Bindman et al., 2019; Winder et al., 2021) e si prevede che continueranno ad aumentare (Tsao, 2020). La questione è complicata da una diffusa carenza generale di radiologi (AAMC Report Reinforces Mounting Physician Shortage, 2021, Clinical Radiology UK Workforce Census 2019 Report, 2019). Gli operatori sanitari, compresi i radiologi, hanno un carico di lavoro crescente (Bruls & Kwee, 2020; Levin et al., 2017), fatto che contribuisce al burnout e all’errore medico (Harry et al., 2021). Poiché la radiologia fornisce servizi essenziali praticamente a tutti gli altri reparti ospedalieri, la carenza di personale in tale disciplina ha un impatto importante su tutto l’ospedale e sulla società nel suo complesso (England & Improvement, 2019; Sutherland et al., senza data).

Con l’invecchiamento della popolazione globale e il crescente carico di malattie croniche, si prevede che in futuro questi problemi rappresenteranno una sfida ancora maggiore per il settore sanitario.

Le soluzioni di imaging medico basate sull’IA hanno le potenzialità per migliorare queste situazioni di impasse per varie ragioni. Innanzitutto, sono particolarmente adatte alla gestione di set di dati vasti e complessi (Alzubaidi et al., 2021). Inoltre, sono molto idonee per automatizzare alcune delle attività che sono tradizionalmente eseguite dai medici radiologi e dai tecnici di radiologia, potendo così liberare del tempo e rendere più efficienti i flussi di lavoro nei reparti di radiologia (Allen et al., 2021; Baltruschat et al., 2021; Kalra et al., 2020; O’Neill et al., 2021; van Leeuwen et al., 2021; Wong et al., 2019). L’IA è anche in grado di rilevare nei dati schemi complessi che l’uomo non è necessariamente in grado di identificare o quantificare (Dance, 2021; Korteling et al., 2021; Kühl et al., 2020).

Principi di base dell’intelligenza artificiale

Il termine “intelligenza artificiale” si riferisce all’uso di sistemi informatici per risolvere problemi specifici in un modo che simula il ragionamento umano. Una caratteristica fondamentale dell’IA è che, come l’essere umano, può adattare le proprie soluzioni al mutare delle circostanze. Va però notato che i sistemi di IA, pur avendo la capacità di imitare il pensiero umano a livello basilare, hanno capacità che spesso superano quelle degli esseri umani, ad esempio in termini di quantità di dati che possono gestire contemporaneamente, di natura e quantità di schemi che sono in grado di rilevare nel dati e di velocità con cui eseguono tali attività.

Le soluzioni di IA si presentano sotto forma di algoritmi informatici, ovvero pezzi di codice informatico che rappresentano istruzioni da seguire per risolvere un problema specifico. Nella sua forma più basilare, l'algoritmo prende i dati come un input, esegue alcuni calcoli su tali dati e restituisce un output.

Un algoritmo di AI può essere programmato esplicitamente per svolgere un compito specifico, proprio come una ricetta passo passo per cuocere una torta. D'altra parte, l'algoritmo può essere programmato per cercare schemi all'interno dei dati per risolvere un problema. Questi tipi di algoritmi sono chiamati algoritmi di apprendimento automatico (machine learning). Dunque, tutti gli algoritmi di apprendimento automatico sono IA, ma non tutta l’IA è apprendimento automatico. Gli schemi presenti nei dati che l'algoritmo può essere programmato esplicitamente per cercare o che può "scoprire" da solo vengono chiamati caratteristiche. Un aspetto importante dell’apprendimento automatico è che tali algoritmi apprendono dai dati stessi e le loro prestazioni migliorano quanti più dati vengono loro forniti.

Uno degli usi più diffusi dell'apprendimento automatico è nella classificazione, ovvero nell'assegnazione di una particolare etichetta a un dato. Ad esempio, un algoritmo di apprendimento automatico può essere utilizzato per stabilire se in una foto (l'input) è raffigurato un cane o un gatto (l'etichetta). L'algoritmo può imparare a svolgere questo compito in modo supervisionato o non supervisionato.

Apprendimento supervisionato

Nell'apprendimento supervisionato, all'algoritmo di apprendimento automatico vengono forniti dati che sono stati etichettati con verità di base, in questo esempio foto di cani e gatti etichettati come tali. Il processo attraversa poi le seguenti fasi:

1.Fase di addestramento (training): l'algoritmo apprende le caratteristiche associate ai cani e ai gatti utilizzando i suddetti dati (dati di addestramento).
2.Fase di test: viene fornita una nuova serie di foto (i dati di test) all’algoritmo, che le etichetta, e poi vengono valutate le prestazione dell’algoritmo su tali dati.

In alcuni casi esiste una fase intermedia tra quelle di addestramento e test, chiamata fase di convalida. In questa fase, all'algoritmo viene fornita una nuova serie di foto (che non erano state incluse né nei dati di addestramento né in quelli di test), vengono valutate le sue prestazioni su questi dati e poi il modello viene ottimizzato e riaddrestrato sui dati di addestramento. Questa fase viene ripetuta fino al raggiungimento di un criterio predefinito basato sulle prestazioni, dopodiché l'algoritmo entra in fase di test.

Apprendimento non supervisionato

Nell'apprendimento non supervisionato l'algoritmo identifica alcune caratteristiche all'interno dei dati immessi che gli consentono di assegnare delle classi ai singoli punti di dati, senza che gli venga detto esplicitamente quali sono o dovrebbero essere tali classi. Tali algoritmi possono identificare schemi o raggruppare punti di dati senza l'intervento umano e includere algoritmi di clustering e riduzione della dimensionalità. Non tutti gli algoritmi di apprendimento automatico eseguono la classificazione; alcuni vengono utilizzati per prevedere una metrica continua (ad es., la temperatura tra quattro settimane) o un'etichetta discreta (ad es., gatti vs. cani) e sono chiamati algoritmi di regressione.

Reti neurali e deep learning

Una rete neurale è costituita da uno strato di input e uno strato di output, a loro volta composti da nodi. Nelle reti neurali semplici, le caratteristiche che sono derivate manualmente da un set di dati vengono inserite nello strato di input, che esegue alcuni calcoli, i cui risultati vengono trasmessi allo strato di output. Nel deep learning esistono diversi strati “nascosti” tra gli strati di input e di output. Ciascun nodo degli strati nascosti esegue dei calcoli utilizzando determinati pesi e trasmette l'output allo strato nascosto successivo, fino al raggiungimento dello strato di output.

All'inizio ai pesi vengono assegnati valori casuali e viene calcolata l’accuratezza dell'algoritmo. I valori dei pesi vengono quindi regolati in modo iterativo finché non viene trovato un set di valori di peso che danno la massima accuratezza. Questa regolazione iterativa dei valori di peso viene solitamente eseguita spostandosi all'indietro dallo strato di output allo strato di input, una tecnica chiamata retropropagazione (backpropagation). L'intero processo viene eseguito sui dati di addestramento.

Valutazione delle prestazioni

Capire come vengono valutate le prestazioni degli algoritmi di IA è fondamentale per interpretare la letteratura sulla stessa. Esistono diverse metriche di prestazione per valutare l’efficacia di un modello nell'esecuzione di determinate attività. Nessuna singola metrica è perfetta, pertanto una combinazione di diverse metriche fornisce un quadro più completo delle prestazioni del modello.

Nella regressione, le metriche più comunemente utilizzate sono:

  • L’errore medio assoluto (Mean Absolute Error, MAE), ossia la differenza media tra i valori previsti e la verità di base.
  • L’errore quadratico medio (Root Mean Square Error, RMSE), dove le differenze tra i valori previsti e la verità di base vengono elevate al quadrato e quindi viene calcolata la media sul campione. Successivamente si prende la radice quadrata della media. A differenza del MAE, l’RMSE attribuisce quindi un peso maggiore alle differenze più ampie.
  • L’R2, ossia la percentuale della varianza totale della verità di base spiegata dalla varianza dei valori previsti. Varia da 0 a 1.

Nelle attività di classificazione si utilizzano per lo più le seguenti metriche:

  • Accuratezza: la percentuale di tutte le previsioni che sono state previste correttamente. Varia da 0 a 1.
  • Sensibilità: detta anche tasso di veri positivi (True Positive Rate, TPR) o richiamo, è la percentuale di veri positivi che sono stati previsti correttamente. Varia da 0 a 1.
  • Specificità: detta anche tasso di veri negativi (True Negative Rate, TNR), è la percentuale di veri negativi previsti correttamente. Varia da 0 a 1.
  • Precisione: detta anche valore predittivo positivo (Positive Predictive Value, PPV), è la percentuale di classificazioni positive previste correttamente. Varia da 0 a 1.

Esiste un compromesso intrinseco tra sensibilità e specificità. L’importanza relativa di ciascuna, così come la loro interpretazione, dipende in larga misura dalla specifica domanda di ricerca e dall’attività di classificazione.

È importante sottolineare che i modelli di classificazione, pur essendo destinati a raggiungere una conclusione binaria, si basano intrinsecamente sulla probabilità. Ciò significa che questi modelli daranno come output una probabilità che un punto di dati appartenga a una classe o a un'altra. Per giungere ad una conclusione sulla classe più probabile, si usa una soglia. Metriche come l’accuratezza, la sensibilità, la specificità e la precisione si riferiscono alle prestazioni dell'algoritmo in base a una determinata soglia. L'area sotto la curva (Area Under the Curve, AUC) della caratteristica di funzionamento del ricevitore è una metrica di prestazione indipendente dalla soglia. Essa può essere interpretata come la probabilità che un esempio positivo casuale venga classificato dall'algoritmo in una posizione più alta rispetto a un esempio negativo casuale.

Nelle attività di segmentazione delle immagini, un tipo di attività di classificazione, vengono comunemente utilizzate le seguenti metriche:

  • Indice di similarità: una misura di sovrapposizione tra due set (ad es., due immagini) che viene calcolata come il doppio del numero di elementi comuni ai set, diviso per la somma del numero di elementi presenti in ciascun set. Va da 0 (nessuna sovrapposizione) a 1 (sovrapposizione perfetta).
  • Distanza di Hausdorff: una misura della distanza tra due set (ad es., due immagini) all'interno di uno spazio. Fondamentalmente è la distanza maggiore tra un punto in un set e il punto più vicino dell'altro set.

Validità interna ed esterna

I modelli validi internamente svolgono bene le loro attività sui dati che si utilizzano per addestrarli e convalidarli. Il grado di validità interna viene valutato utilizzando le metriche di prestazione sopra descritte e dipende dalle caratteristiche del modello stesso e dalla qualità dei dati su cui il modello è stato addestrato e convalidato.

I modelli validi esternamente svolgono bene le loro attività sui dati nuovi (Ramspek et al., 2021). Migliori sono le prestazioni del modello su dati diversi dai dati su cui i modelli sono stati addestrati e convalidati, maggiore è la validità esterna. In pratica, ciò spesso richiede che le prestazioni dei modelli siano testate su dati provenienti da ospedali o aree geografiche che non facevano parte dei set di dati di addestramento e convalida del modello.

Linee guida per la valutazione della ricerca sull’intelligenza artificiale

Sono state sviluppate diverse linee guida per valutare l’evidenza a sostegno degli interventi basati sull’IA nel settore sanitario (X. Liu et al., 2020; Mongan et al., 2020; Shelmerdine et al., 2021; Weikert et al., 2021). Tali linee guida forniscono un modello per chi fa ricerca sull’IA nel settore sanitario e mirano a garantire che le informazioni rilevanti siano riportate in modo trasparente e completo, ma possono anche essere utilizzate da altri soggetti per valutare la qualità delle ricerche pubblicate. Questo serve per garantire che le soluzioni basate sull’IA che hanno dei limiti sostanziali potenziali o reali, e soprattutto quelle con presentazioni di scarso valore (Bozkurt et al., 2020; D.W. Kim et al., 2019; X. Liu et al., 2019; Nagendran et al., 2020; Yusuf et al., 2020), non siano adottate prematuramente (CONSORT-AI e SPIRIT-AI Steering Group, 2019). Sono state inoltre proposte delle linee guida per valutare l'affidabilità delle soluzioni basate sull'IA in termini di trasparenza, riservatezza, sicurezza e responsabilità (Buruk et al., 2020; Lekadir et al., 2021; Zicari et al., 2021).

Usi clinici

Negli ultimi anni, l’IA ha mostrato grosse potenzialità nell’eseguire una vasta gamma di attività all’interno di un reparto di imaging medico, comprese molte fasi precedenti al momento in cui si esegue l’esame sul paziente. Le implementazioni dell'IA per migliorare l'efficienza dei flussi di lavoro radiologici prima dell’esame sul paziente vengono talvolta definite "IA a monte" (upstream AI) (Kapoor et al., 2020; M.L. Richardson et al., 2021).

Programmazione

Un’applicazione promettente dell’IA a monte è la possibilità di prevedere quali pazienti potrebbero non presentarsi agli appuntamenti per gli esami. Gli appuntamenti saltati sono associati a un aumento significativo del carico di lavoro e dei costi (Dantas et al., 2018). Utilizzando un approccio Gradient Boosting, Nelson et al. hanno previsto con elevata accuratezza il numero di appuntamenti saltati per risonanze magnetiche (RM) ospedaliere nel servizio sanitario nazionale (National Health Service, NHS) del Regno Unito (Nelson et al., 2019). Le loro simulazioni hanno anche suggerito che agire in base alle previsioni di questo modello prendendo di mira i pazienti che probabilmente salteranno i loro appuntamenti produrrebbe potenzialmente un beneficio netto di diverse sterline per appuntamento su una gamma di soglie di modello e tassi di appuntamenti saltati (Nelson et al., 2019). Risultati simili sono stati recentemente ottenuti in uno studio condotto su un singolo ospedale di Singapore, dove nei 6 mesi successivi all'implementazione dello strumento predittivo sono riusciti a ridurre significativamente il tasso di mancate presentazioni agli esami dal 19,3% al 15,9%, con un potenziale risparmio di $180.000 (Chong et. al., 2020).

La programmazione degli esami in un reparto di radiologia è un’attività impegnativa perché, pur essendo per lo più un compito amministrativo, dipende molto dalle informazioni mediche. Per assegnare specifici appuntamenti ai pazienti, dunque, spesso ci vuole l’aiuto di qualcuno che conosca bene il campo, il che prevede che la persona che fissa l’appuntamento sia un medico radiologo o un tecnico di radiologia oppure che tali soggetti intervengano regolarmente nel processo. In entrambi gli scenari, il processo è alquanto inefficiente e potrebbe essere semplificato mediante algoritmi di IA che controllano le indicazioni e le controindicazioni degli esami e forniscono agli incaricati della programmazione degli esami le necessarie informazioni sull’urgenza o meno degli stessi (Letourneau-Guillon et al., 2020).

Determinazione del protocollo

A seconda delle politiche di ciascun ospedale o clinica, la decisione su quale esatto protocollo di scansione un paziente debba seguire viene solitamente presa in base alle informazioni contenute nella richiesta dell’esame emessa dal medico curante e a giudizio del radiologo. Questo è spesso integrato dalla comunicazione diretta tra il medico curante e il radiologo e dal controllo delle informazioni mediche relative al paziente da parte del radiologo. Questo processo migliora l’assistenza del paziente (Boland et al., 2014), ma può essere dispendioso in termini di tempo e inefficiente, soprattutto nel caso di esami quali la RM, in cui esiste un gran numero di protocolli possibili. In uno studio, la sola determinazione del protocollo da seguire ha rappresentato circa il 6% del tempo di lavoro del radiologo (Schemmel et al., 2016). I radiologi vengono spesso interrotti per attività come la determinazione del protocollo da seguire mentre stanno interpretando le immagini, nonostante quest'ultima attività dovrebbe essere il loro compito principale (Balint et al., 2014; J.-PJ Yu et al., 2014).

Utilizzando dei classificatori del linguaggio naturale, la stessa tecnologia utilizzata per i chatbot e gli assistenti virtuali, si è tentato di interpretare il testo della richiesta dell’esame fatta dal medico curante. I classificatori del linguaggio naturale basati sul deep learning (apprendimento profondo) si sono mostrati promettenti nell’assegnare ai pazienti un protocollo di RM con o senza mezzo di contrasto per la RM muscoloscheletrica, con un’accuratezza dell'83% (Trivedi et al., 2018) e del 94% (Y. H. Lee, 2018). Algoritmi simili hanno mostrato un'accuratezza del 95% nel prevedere il protocollo di RM cerebrale appropriato utilizzando una combinazione di un massimo di 41 diverse sequenze di RM (Brown & Marotta, 2018). In un'ampia gamma di regioni del corpo, un classificatore del linguaggio naturale basato sull'apprendimento profondo ha deciso, in base al testo della richiesta dell’esame, se assegnare automaticamente uno specifico protocollo di tomografia computerizzata (TC) o RM (operazione svolta con un’accuratezza del 95%) oppure, nei casi più difficili, se consigliare al radiologo un elenco dei tre protocolli più appropriati (operazione svolta con un’accuratezza del 92%) (Kalra et al., 2020).

L’IA è stata utilizzata anche per decidere se gli esami già inclusi nel protocollo dovessero essere ampliati, una decisione che deve essere presa in tempo reale mentre il paziente è all’interno del macchinario. Un esempio di questo tipo di situazione è la RM della prostata, in cui la decisione se somministrare un mezzo di contrasto o meno viene spesso presa una volta acquisite le sequenze senza mezzo di contrasto. Hötker et al. hanno scoperto che una rete neurale convoluzionale (Convolutional Neural Network, CNN) ha assegnato il 78% dei pazienti al protocollo di RM della prostata giusto (Hötker et al., 2021). La sensibilità della CNN per la necessità dell’uso del mezzo di contrasto è stata del 94,4%, con una specificità del 68,8%, e si è dovuto richiamare per ripetere l’esame con il mezzo di contrasto solo il 2% dei pazienti di tale studio (Hötker et al., 2021).

Miglioramento e monitoraggio della qualità delle immagini

Recentemente sono state implementate molte soluzioni basate sull’IA che funzionano in background nei flussi di lavoro radiologici per migliorare la qualità delle immagini. Tra queste vi sono soluzioni per il monitoraggio della qualità delle immagini, per la riduzione degli artefatti sulle immagini, per il miglioramento della risoluzione spaziale e per l’accelerazione dell’esecuzione degli esami stessi.

Tali soluzioni stanno entrando nel mainstream della radiologia, soprattutto nella tomografia computerizzata, che per decenni ha utilizzato metodi consolidati ma soggetti ad artefatti per ricostruire immagini interpretabili dai dati grezzi dei sensori (Deák et al., 2013; Singh et al., 2010). Oggi si stanno gradualmente sostituendo queste soluzioni con metodi di ricostruzione basati sul deep learning, che migliorano la qualità delle immagini mantenendo basse le dosi di radiazioni (Akagi et al., 2019; H. Chen et al., 2017; Choe et al., 2019; Shan et al., 2019), e la ricostruzione ora viene eseguita su supercomputer integrati direttamente nel tomografo o nel cloud. L'equilibrio tra la dose di radiazioni e la qualità delle immagini può essere regolato in base al protocollo specifico per adattare le scansioni ai singoli pazienti e scenari clinici (McLeavy et al., 2021; Willemink & Noël, 2019). Tali approcci sono risultati particolarmente utili in radiologia pediatrica, nelle donne in gravidanza e nei pazienti obesi, nonché nelle TC dell’apparato urinario e del cuore (McLeavy et al., 2021).

Sono state utilizzate soluzioni basate sull’IA anche per ridurre la durata dell’esame mantenendone la qualità diagnostica. La riduzione della durata dell’esame radiologico non solo migliora l'efficienza complessiva, ma contribuisce anche a una migliore esperienza del paziente e alla conformità con l’esame di imaging. In uno studio multicentrico sulla RM della colonna vertebrale, un algoritmo di ricostruzione delle immagini basato sul deep learning, che ha migliorato le immagini tramite filtraggio e una riduzione del rumore tale da conservare meglio i dettagli, ha ridotto i tempi di durata dell’esame del 40% (Bash, Johnson, et al., 2021). Per le RM cerebrali pesate in T1, un algoritmo simile che migliorava la nitidezza delle immagini e ne riduceva il rumore ha ridotto la durata dell’esame del 60% mantenendo l'accuratezza della volumetria della regione cerebrale rispetto agli esami standard (Bash, Wang, et al., 2021).

Nella prassi radiologica di routine, le immagini spesso contengono artefatti che ne riducono l’interpretabilità. Questi artefatti sono il risultato di alcune caratteristiche della specifica modalità di imaging o del protocollo utilizzato, oppure di fattori intrinseci al paziente sottoposto all’esame, come la presenza di corpi estranei o eventuali movimenti fatti dal paziente durante la scansione. Soprattutto nel caso della RM, i protocolli di imaging che richiedono una scansione rapida spesso introducono alcuni artefatti nell'immagine ricostruita. In uno studio, un algoritmo basato sul deep learning ha ridotto gli artefatti di banding (scalettatura) associati a sequenze di RM a precessione libera allo stato stazionario bilanciate del cervello e del ginocchio (K.H. Kim e Park, 2017). Per l'imaging in tempo reale del cuore mediante RM, un altro studio ha rilevato che gli artefatti di aliasing (ribaltamento) introdotti dal sottocampionamento (undersampling) dei dati sono stati ridotti utilizzando un approccio basato sul deep learning (Hauptmann et al., 2019). La presenza di corpi estranei metallici, come impianti dentali, ortopedici o vascolari, è un fattore legato al paziente che si correla spesso alla formazione di artefatti nelle immagini sia di TC che di RM (Boas & Fleischmann, 2012; Hargreaves et al., 2011). Per ridurre questi artefatti sono stati studiati diversi approcci basati sul deep learning, ma nessuno si è ancora ben consolidato (Ghani & Clem Karl, 2019; Puvanasunthararajah et al., 2021; Zhang & Yu, 2018). Si stanno sperimentando approcci simili per ridurre gli artefatti legati al movimento nella RM (Tamada et al., 2020; B. Zhao et al., 2022).

Le soluzioni basate sull’IA per il monitoraggio della qualità delle immagini possono ridurre la necessità di dover richiamare i pazienti per ripetere gli esami, un problema piuttosto diffuso (Schreiber-Zinaman & Rosenkrantz, 2017). Un algoritmo basato sul deep learning che identifica la vista radiografica acquisita ed estrae misure relative alla qualità dalle radiografie della caviglia è stato in grado di prevedere la qualità delle immagini con un’accuratezza del 94% circa (Mairhöfer et al., 2021). Un altro approccio basato sul deep learning è stato in grado di prevedere esami di RM epatici non diagnostici con un valore predittivo negativo compreso tra l’86% e il 94% (Esses et al., 2018). Questo controllo di qualità automatizzato in tempo reale consente potenzialmente ai tecnici radiologi di rieseguire gli esami o di eseguire ulteriori esami con un maggior valore diagnostico.

Prioritizzazione di lettura delle immagini

Con la carenza di personale e l'aumento del numero di esami eseguiti, i radiologi si trovano a dover leggere una gran quantità di immagini. Per ottimizzare l'efficienza e l’assistenza del paziente, sono state suggerite soluzioni basate sull'IA per dare priorità alle immagini che i radiologi devono esaminare e refertare per prime, solitamente selezionando le immagini acquisite in base a risultati che richiedono un intervento urgente (O'Connor & Bhalla, 2021). Questo tipo di soluzioni è stato studiato più approfonditamente in neuroradiologia, dove lo spostamento in cima all’elenco delle TC che presentavano un’emorragia intracranica mediante uno strumento basato sull’IA ha ridotto di vari minuti il tempo trascorso prima che il radiologo le visionasse (O’Neill et al., 2021). In un altro studio, grazie alla prioritizzazione mediante lista di lavoro, il tempo necessario per arrivare alla diagnosi (ossia il tempo intercorso tra l'acquisizione delle immagini e la loro visualizzazione, lettura e refertazione da parte del radiologo) è stato ridotto da 512 a 19 minuti in ambito ambulatoriale (Arbabshirani et al., 2018). In uno studio di simulazione, in cui è stata utilizzata la prioritizzazione mediante lista di lavoro basata sull’IA per identificare i risultati che richiedevano un intervento urgente su alcune radiografie del torace (come pneumotorace, versamenti pleurici e corpi estranei), si è avuta una sostanziale riduzione del tempo necessario per visualizzare e refertare le immagini rispetto a quello necessario con la prioritizzazione del flusso di lavoro standard (Baltruschat et al., 2021).

Interpretazione delle immagini

Attualmente la maggior parte delle soluzioni per l’imaging medico basate sull’IA disponibili in commercio sono mirate ad alcuni aspetti dell’analisi e dell’interpretazione delle immagini (Rezazade Mehrizi et al., 2021; van Leeuwen et al., 2021), come segmentare l’immagine in parti o partizione (per il targeting chirurgico o radioterapico, ad esempio), portare aree sospette all'attenzione del radiologo estraendo biomarcatori di imaging (radiomica), confrontare le immagini nel tempo e raggiungere diagnosi di imaging specifiche.

Neurologia

  • 29-38% delle applicazioni basate sull'IA in commercio in radiologia (Rezazade Mehrizi et al., 2021; van Leeuwen et al., 2021).

La maggior parte delle soluzioni basate sull’IA disponibili in commercio studiate per interpretare i dati di neuroimaging mirano a rilevare e caratterizzare l’ictus ischemico, l’emorragia intracranica, la demenza e la sclerosi multipla (Olthof et al., 2020). Vari studi hanno dimostrato un’eccellente accuratezza dei metodi basati sull’IA nel rilevamento e nella classificazione dell’emorragia intraparenchimale, subaracnoidea e subdurale alla TC cerebrale (Flanders et al., 2020; Ker et al., 2019; Kuo et al., 2019). Studi successivi hanno dimostrato che, rispetto ai radiologi, alcune soluzioni basate sull’IA hanno tassi di falsi positivi e negativi sostanzialmente inferiori (Ginat, 2020; Rao et al., 2021). Nell’ictus ischemico, le soluzioni basate sull’IA si sono in gran parte concentrate sulla quantificazione del core infartuale (Goebel et al., 2018; Maegerlein et al., 2019), sul rilevamento delle occlusioni dei grandi vasi (Matsoukas et al., 2022; Morey et al., 2021; Murray et al., 2020; Shlobin et al., 2022) e sulla previsione degli esiti dell'ictus (Bacchi et al., 2020; Nielsen et al., 2018; Y. Yu et al., 2020, 2021).

Nella sclerosi multipla, l’IA è stata utilizzata per identificare e segmentare le lesioni (Nair et al., 2020; S.-H. Wang et al., 2018), una prassi che può essere particolarmente utile nel follow-up longitudinale dei pazienti. L’IA è stata utilizzata anche per estrarre caratteristiche di imaging associate alla progressione della malattia e alla conversione da sindrome clinicamente isolata a sclerosi multipla definita (Narayana et al., 2020; Yoo et al., 2019). Altre applicazioni dell'IA in neuroradiologia sono il rilevamento degli aneurismi intracranici (Faron et al., 2020; Nakao et al., 2018; Ueda et al., 2019) e la segmentazione dei tumori cerebrali (Kao et al., 2019; Mlynarski et al., 2019; Zhou et al., 2020), nonché la previsione dei marcatori genetici del tumore al cervello da dati di imaging (Choi et al., 2019; J. Zhao et al., 2020).

Torace

  • 24%-31% delle applicazioni basate sull’IA disponibili in commercio in radiologia (Rezazade Mehrizi et al., 2021; van Leeuwen et al., 2021).

Nell’interpretazione delle radiografie del torace, i radiologi hanno rilevato risultati sostanzialmente più critici e urgenti quando sono stati aiutati da un algoritmo basato sul deep learning e lo hanno fatto molto più velocemente che senza l'algoritmo (Nam et al., 2021). È stato inoltre riscontrato che gli algoritmi di interpretazione delle immagini basati sul deep learning migliorano dal 66% al 73% la sensibilità degli specializzandi in radiologia nel rilevare reperti urgenti nelle radiografie del torace (E. J. Hwang, Nam et al., 2019). Anche un altro studio condotto su una serie più ampia di reperti di radiografie del torace ha rilevato che i radiologi aiutati da un algoritmo basato sul deep learning avevano una maggiore accuratezza diagnostica rispetto ai radiologi che leggevano le radiografie senza assistenza (Seah et al., 2021). Gli usi dell’IA nella radiologia del torace si estendono anche all’imaging trasversale come la TC. Un algoritmo di deep learning ha rilevato l'embolia polmonare nelle TC con un’elevata accuratezza (AUC = 0,85) (Huang, Kothari, et al., 2020). Inoltre, un algoritmo di deep learning ha mostrato un’accuratezza del 90% nel rilevare la dissezione aortica con esami TC senza mezzo di contrasto, un risultato simile alle prestazioni dei radiologi (Hata et al., 2021).

Al di fuori del contesto di emergenza, le soluzioni basate sull’IA sono state ampiamente testate e implementate nello screening della tubercolosi su radiografie del torace (E. J. Hwang, Park, et al., 2019; S. Hwang et al., 2016; Khan et al., 2020; Qin et al., 2019; WHO Operational Handbook on Tuberculosis Module 2: Screening – Systematic Screening for Tuberculosis Disease, senza data). Inoltre, queste soluzioni sono state utili anche nello screening del cancro del polmone, sia in termini di rilevamento di noduli polmonari alla TC (Setio et al., 2017) e alla radiografia del torace (Li et al., 2020), sia nella classificazione dei noduli in maligni o benigni (Ardila et al., 2019; Bonavita et al., 2020; Ciompi et al., 2017; B. Wu et al., 2018). Le soluzioni basate sull’IA si dimostrano molto promettenti anche nella diagnosi di polmonite, broncopneumopatia cronica ostruttiva e malattia polmonare interstiziale (F. Liu et al., 2021).

Senologia

  • 11% delle applicazioni basate sull’IA disponibili in commercio in radiologia (Rezazade Mehrizi et al., 2021; van Leeuwen et al., 2021).

Finora molti degli algoritmi di IA utilizzati nella diagnostica per immagini in senologia hanno mirato a ridurre il carico di lavoro di lettura delle mammografie da parte dei radiologi. Questo è stato fatto, ad esempio, usando algoritmi di IA studiati per individuare le mammografie negative, con conseguente riduzione di quasi un quinto del carico di lavoro dei radiologi in uno studio (Yala et al., 2019). Altri studi in cui la seconda lettura delle mammografie è stata sostituita con algoritmi di IA hanno dimostrato come ciò porti a un minor numero di falsi positivi e falsi negativi e riduca dell’88% il carico di lavoro del secondo lettore (McKinney et al., 2020).

È stato anche scoperto che le soluzioni basate sull’IA per la mammografia aumentano l’accuratezza diagnostica dei radiologi (McKinney et al., 2020; Rodríguez-Ruiz et al., 2019; Watanabe et al., 2019), e alcune di esse si sono rivelate estremamente accurate nel rilevare e classificare le lesioni mammarie in modo indipendente (Agnes et al., 2019; Al-Antari et al., 2020; Rodriguez-Ruiz et al., 2019).
Tuttavia, una recente revisione sistematica di 36 algoritmi di IA ha rilevato che questi studi erano di scarsa qualità metodologica e che tutti gli algoritmi erano stati meno accurati rispetto al consenso di due o più radiologi (Freeman et al., 2021). Gli algoritmi di IA hanno comunque dimostrato di avere delle potenzialità nell’estrarre dalle mammografie le caratteristiche predittive del cancro al di là della densità mammografica (Arefan et al., 2020; Dembrower et al., 2020; Hinton et al., 2019). Oltre alla mammografia, sono state sviluppate soluzioni basate sull’IA per rilevare e classificare le lesioni mammarie nelle ecografie (Akkus et al., 2019; Park et al., 2019; G.-G. Wu et al., 2019) e nelle RM (Herent et al., 2019).

Cardiologia

  • 11% delle applicazioni basate sull’IA disponibili in commercio in radiologia (Rezazade Mehrizi et al., 2021; van Leeuwen et al., 2021)..

La radiologia cardiaca è sempre stata particolarmente impegnativa a causa delle difficoltà insite nell’acquisizione di immagini di un organo in costante movimento. Tuttavia, proprio per questo ha tratto enormi vantaggi dai progressi della tecnologia di imaging e sembra destinata a trarre grandi benefici anche dall’IA (Sermesant et al., 2021). La maggior parte delle applicazioni basate sull’IA per il sistema cardiovascolare utilizzano dati ottenuti mediante RM, TC o ecografia (Weikert et al., 2021). Alcuni esempi importanti sono: il calcolo automatico della frazione di eiezione nell'ecocardiografia, la quantificazione delle calcificazioni nelle arterie coronarie nella TC cardiaca, la determinazione del volume del ventricolo destro nell’angio-TAC polmonare e la determinazione delle dimensioni e dello spessore della camera cardiaca nella RM cardiaca (Medical AI Evaluation, senza data, The Medical Futurist, senza data). Anche le soluzioni basate sull’IA per la previsione della probabile risposta favorevole dei pazienti agli interventi cardiaci, come la terapia di resincronizzazione cardiaca, basata su imaging e parametri clinici, si sono mostrate molto promettenti (Cikes et al., 2019; Hu et al., 2019). Le alterazioni non immediatamente visibili al lettore umano nella RM cardiaca, che possono essere utili per differenziare tra i vari tipi di cardiomiopatia, possono essere rilevate anche utilizzando l’IA attraverso l’analisi della texture (Neisius et al., 2019; J. Wang et al., 2020) e altri approcci radiomici (Mancio et al., 2022).

Sistema muscoloscheletrico

  • 7-11% delle applicazioni basate sull’IA disponibili in commercio in radiologia (Rezazade Mehrizi et al., 2021; van Leeuwen et al., 2021)..

Le applicazioni promettenti dell'IA nella valutazione di muscoli, ossa e articolazioni includono applicazioni in cui i lettori umani generalmente mostrano scarsa affidabilità intra-valutatore e inter-valutatore, come la determinazione dell'età ossea mediante radiografia (Halabi et al., 2019; Thodberg et al., 2009) e lo screening dell'osteoporosi in radiografie (Kathirvelu et al., 2019; J.-S. Lee et al., 2019) e TC (Pan et al., 2020). Le soluzioni basate sull’IA si sono mostrate promettenti anche nel rilevare fratture in radiografie e TC (Lindsey et al., 2018; Olczak et al., 2017; Urakawa et al., 2019). Una revisione sistematica delle soluzioni basate sull’IA per il rilevamento delle fratture in diverse parti del corpo ha mostrato AUC comprese tra 0,94 e 1,00 e accuratezze comprese tra il 77% e il 98% (Langerhuizen et al., 2019). Le soluzioni basate sull'IA hanno inoltre raggiunto accuratezze simili a quelle dei radiologi nella classificazione della gravità delle modifiche degenerative della colonna vertebrale (Jamaludin et al., 2017) e delle articolazioni degli arti (F. Liu et al., 2018; Thomas et al., 2020). Sono state sviluppate soluzioni basate sull'IA anche per la determinazione dell'origine delle metastasi scheletriche (Lang et al., 2019) e per la classificazione dei tumori ossei primari (Do et al., 2017).

Addome e bacino

  • 4% delle applicazioni basate sull’IA disponibili in commercio in radiologia (Rezazade Mehrizi et al., 2021; van Leeuwen et al., 2021)..

Gran parte della ricerca svolta sull’utilizzo dell’IA nell’imaging addominale si è finora concentrata sulla segmentazione automatica di organi come il fegato (Dou et al., 2017), la milza (Moon et al., 2019), il pancreas (Oktay et al., 2018) e i reni (Sharma et al., 2017). Inoltre, una revisione sistematica di 11 studi che hanno utilizzato il deep learning per il rilevamento di masse epatiche maligne ha mostrato accuratezze fino al 97% e AUC fino a 0,92 (Azer, 2019).

Altre applicazioni di IA in radiologia addominale hanno incluso il rilevamento della fibrosi epatica (He et al., 2019; Yasaka et al., 2018), della steatosi epatica e del contenuto di ferro epatico, il rilevamento del gas addominale libero alla TC e la volumetria e segmentazione automatiche della prostata (AI for Radiology, senza data).

Ostacoli all'implementazione

Nonostante le sue grandi potenzialità, l’IA nell’imaging medico non ha ancora trovato un’implementazione e un impatto diffusi nella pratica clinica di routine. Il passaggio dalla ricerca alla pratica clinica è ostacolato da vari aspetti complessi e interconnessi che, direttamente o indirettamente, riducono la probabilità che vengano adottate soluzioni basate sull’IA. Uno di questi aspetti è la mancanza di fiducia nelle soluzioni basate sull’IA da parte delle principali parti interessate, come gli enti normativi, gli operatori sanitari e i pazienti (Cadario et al., 2021; Esmaeilzadeh, 2020; J.P. Richardson et al., 2021; Tucci et al., 2022).

Generalizzabilità

Una delle sfide più importanti è sviluppare soluzioni basate sull’IA che continuino a funzionare bene in nuovi scenari reali. In un’ampia revisione sistematica, per quasi la metà degli algoritmi di imaging medico basati sull’IA studiati e testati su nuovi dati è stata riportata una riduzione superiore a 0,05 dell’AUC (A. C. Yu et al., 2022). Questa mancanza di generalizzabilità può avere effetti negativi sulle prestazioni del modello in uno scenario reale.

Se una soluzione funziona male quando viene testata su un set di dati con una distribuzione simile o identica al set di dati usato nell’addestramento dell’algoritmo, si dice che non ha una stretta generalizzabilità, e ciò è spesso conseguenza di un overfitting o sovradattamento (Eche et al., 2021). Le potenziali soluzioni per l'overfitting includono l’uso di set di dati di addestramento più grandi e la riduzione della complessità del modello. Se una soluzione funziona male quando viene testata su un set di dati con una distribuzione diversa rispetto al set di dati di addestramento (ad es., una diversa distribuzione delle etnie dei pazienti), si dice che manca di un'ampia generalizzabilità (Eche et al., 2021). Le soluzioni per questa limitazione comprendono lo stress test del modello su set di dati con distribuzioni diverse dal set di dati di addestramento (Eche et al., 2021).

Le soluzioni di IA vengono spesso sviluppate in ambienti ricchi di risorse, come le grandi aziende tecnologiche e i centri medici accademici nei Paesi ricchi. È probabile che i risultati e le prestazioni in questi contesti con risorse elevate non possano essere generalizzati a contesti con meno risorse, come ospedali più piccoli, aree rurali o Paesi più poveri (Price & Nicholson, 2019), il che complica ulteriormente la questione.

Rischio di bias

Nelle soluzioni basate sull’IA si possono avere dei bias dovuti a dati o fattori umani. Il primo caso si verifica quando i dati utilizzati per addestrare la soluzione di IA non rappresentano adeguatamente la popolazione target. I set di dati possono non essere rappresentativi quando sono troppo piccoli o sono stati raccolti in un modo che non rappresenta una determinata categoria della popolazione. Le soluzioni di IA addestrate con dati non rappresentativi perpetuano i bias e hanno scarse prestazioni nelle categorie di popolazione male o sotto-rappresentate nei dati di addestramento. La presenza di tali bias è stata dimostrata empiricamente in molti studi sull’imaging medico basato sull’IA (Larrazabal et al., 2020; Seyyed-Kalantari et al., 2021).

Le soluzioni basate sull’IA sono soggette a numerose decisioni soggettive e talvolta distorte implicitamente o esplicitamente durante il loro sviluppo da parte dell’uomo. Questi fattori umani includono il modo in cui vengono selezionati i dati di addestramento, come vengono etichettati e come viene presa la decisione di concentrarsi sul problema specifico che la soluzione basata sull'IA intende risolvere (Norori et al., 2021). Sono disponibili alcune raccomandazioni e strumenti per ridurre al minimo il rischio di bias nella ricerca sull’IA ( AIF360: A Comprehensive Set of Fairness Metrics for Datasets and Machine Learning Models, Explanations for These Metrics, and Algorithms to Mitigate Bias in Datasets and Models, senza data, IBM Watson Studio - Model Risk Management, senza data; Silberg & Manyika, 2019).

Quantità, qualità e varietà dei dati

I problemi come i bias e la scarsa generalizzabilità possono essere mitigati garantendo che i dati usati per l’addestramento degli algoritmi siano di quantità, qualità e varietà sufficienti. Tuttavia, ciò non è facile perché i pazienti sono spesso riluttanti a condividere i propri dati per scopi commerciali (Aggarwal, Farag, et al., 2021; Ghafur et al., 2020; Trinidad et al., 2020), gli ospedali e le cliniche non sono solitamente attrezzati per rendere disponibili questi dati in modo utilizzabile e sicuro, e organizzare ed etichettare i dati richiede tempo e denaro.

Molti set di dati possono essere utilizzati per molteplici scopi diversi, e la condivisione dei dati tra aziende può contribuire a rendere più efficiente il processo di raccolta e organizzazione dei dati, oltre ad aumentare la quantità di dati disponibili per ciascuna applicazione. Tuttavia, gli sviluppatori, per restare competitivi, sono spesso riluttanti a condividere i dati tra loro o addirittura a rivelare la fonte esatta dei propri dati.

Protezione dei dati e riservatezza

Per sviluppare e implementare soluzioni basate sull’IA occorre, inoltre, che i pazienti siano esplicitamente informati e diano il loro consenso all’uso dei propri dati per finalità specifiche e da parte di determinati soggetti. Questi dati devono inoltre essere adeguatamente protetti da violazioni e usi impropri. L’impossibilità di garantire tutto questo mina notevolmente la fiducia del pubblico nelle soluzioni basate sull’IA e ne ostacola l’adozione. Sebbene, secondo le norme che regolamentano la riservatezza dei dati sanitari, la raccolta di dati completamente anonimizzati non richiede il consenso esplicito del paziente (Regolamento generale sulla protezione dei dati [GDPR] – Testo legale ufficiale, 2016; Office for Civil Rights [OCR], 2012) e in teoria vige una protezione dall’uso improprio dei dati, non è ancora stato raggiunto un consenso in merito a se i dati di imaging possano o non possano essere completamente anonimizzati (Lotan et al., 2020; Murdoch, 2021). È inoltre controverso se il consenso possa essere veramente informato, considerando la complessità dei dati acquisiti e la conseguente miriade di potenziali usi futuri degli stessi (Vayena & Blasimme, 2017).

Infrastruttura IT

Tra i reparti ospedalieri, la radiologia è da sempre in prima linea nella digitalizzazione. Le soluzioni basate sull’IA incentrate sull’elaborazione e sull’interpretazione delle immagini troveranno probabilmente l’infrastruttura necessaria nella maggior parte dei reparti di radiologia, ad esempio quanto necessario per collegare le apparecchiature di imaging ai computer per l’analisi e per l’archiviazione delle immagini e di altri output. Tuttavia, è probabile che la maggior parte dei reparti di radiologia abbia bisogno di aggiornamenti significativi dell’infrastruttura per altre applicazioni di IA, soprattutto per quelle che richiedono l’integrazione di informazioni provenienti da più fonti e con output complessi. Inoltre, è importante tenere presente che la distribuzione delle infrastrutture necessarie è fortemente diseguale tra diversi Paesi e anche all’interno dello stesso Paese (Health Ethics & Governance, 2021).

Per quanto riguarda la potenza di calcolo, i reparti di radiologia dovranno investire risorse nell’hardware e nel personale necessari per utilizzare queste soluzioni basate sull’IA oppure optare per soluzioni basate sul cloud. I primi avranno un costo maggiore, ma consentiranno l'elaborazione dei dati entro i confini della rete locale dell'ospedale o della clinica. Le soluzioni informatiche basate sul cloud (note come “infrastruttura come servizio” o “IaaS”) sono spesso considerate l’opzione meno sicura e meno affidabile, ma ciò dipende da una serie di fattori e non è quindi sempre vero (Baccianella & Gough, senza data). Esistono delle linee guida che indicano cosa occorre considerare quando si acquistano soluzioni basate sul cloud nel settore sanitario (Cloud Security for Healthcare Services, 2021).

Mancanza di standardizzazione, interoperabilità e integrabilità

Il problema dell’infrastruttura diventa ancora più complicato se si considera quanto sia attualmente frammentato il mercato dell’imaging medico basato sull’IA (Alexander et al., 2020). È quindi probabile che, nel prossimo futuro, in un singolo reparto siano in esecuzione contemporaneamente diverse dozzine di soluzioni basate sull’IA di diversi fornitori. Avere un'infrastruttura separata e autonoma (ad esempio una workstation o un server) per ciascuna di esse sarebbe incredibilmente complicato e difficile da gestire. Le soluzioni a questo problema che sono state suggerite includono i “mercati” di soluzioni di IA, che sono delle specie di app store (Advanced AI Solutions for Radiology, senza data, Curated Marketplace, 2018, Imaging AI Marketplace - Overview, senza data, Sectra Amplifier Marketplace, 2021, The Nuance AI Marketplace for Diagnostic Imaging, senza data) e lo sviluppo di infrastrutture onnicomprensive non legate a specifici fornitori (Leiner et al., 2021). Per un’efficace implementazione di tali soluzioni occorre un’intensa collaborazione tra sviluppatori di soluzioni di IA, fornitori di servizi di imaging e società di tecnologia dell’informazione.

Interpretabilità

Spesso è impossibile capire esattamente come le soluzioni basate sull’IA giungano alle loro conclusioni, soprattutto nel caso di approcci complessi come il deep learning, e questo riduce la trasparenza del processo decisionale per l’acquisizione e l’approvazione di queste soluzioni, rende complicata l’identificazione dei bias e rende più difficile per i medici spiegare i risultati di queste soluzioni ai loro pazienti e decidere se una soluzione funzioni bene o non abbia funzionato correttamente (Char et al., 2018; Reddy et al., 2020; Vayena et al., 2018; Whittlestone et al., 2019). Alcuni hanno suggerito che per affrontare queste difficoltà possono rivelarsi utili proprio le tecniche che aiutano gli esseri umani a capire come gli algoritmi di IA prendono determinate decisioni o fanno certe previsioni (IA “interpretabile” o “spiegabile”). Tuttavia, altri hanno sostenuto che le tecniche attualmente disponibili non sono adatte per comprendere le singole decisioni di un algoritmo e hanno messo in guardia dal fare affidamento su di esse per garantire che gli algoritmi funzionino in modo sicuro e affidabile (Ghassemi et al., 2021).

Responsabilità

Nei sistemi sanitari vige un quadro giuridico in cui gli operatori sanitari e gli istituti medici possono essere ritenuti responsabili degli effetti avversi derivanti dalle loro azioni. La questione su chi dovrebbe essere ritenuto responsabile dei fallimenti di una soluzione basata sull’IA è complicata. Per i prodotti farmaceutici, ad esempio, la responsabilità dei difetti intrinseci del prodotto o del suo utilizzo è spesso in capo al produttore o a chi ha prescritto il farmaco. Una differenza fondamentale è che i sistemi basati sull’IA sono in costante evoluzione e apprendimento, pertanto funzionano intrinsecamente in un modo che non dipende da ciò che i loro sviluppatori avrebbero potuto prevedere (Yeung, 2018). Per l’utente finale, come l’operatore sanitario, la soluzione basata sull’IA potrebbe essere poco trasparente e, quindi, l’utente potrebbe non essere in grado di capire quando la soluzione non funziona correttamente o è inaccurata (Habli et al., 2020; Yeung, 2018).

Fragilità

Nonostante i sostanziali progressi nello sviluppo degli ultimi anni, gli algoritmi di deep learning sono ancora sorprendentemente fragili. Ciò significa che, quando l’algoritmo si trova ad affrontare uno scenario sostanzialmente diverso da quello affrontato durante l’addestramento, non riesce a contestualizzarlo e spesso genera risultati senza senso o inaccurati. Ciò accade perché, a differenza della mente umana, l’algoritmo per lo più impara a percepire le cose entro i confini di determinati presupposti, ma non riesce a generalizzare al di fuori di questi ultimi. Per fare un esempio di come si potrebbe abusare di questo con intento doloso, bastano delle minime modifiche alle immagini mediche, impercettibili da parte dell’uomo, per rendere inaccurati i risultati degli algoritmi di classificazione delle malattie (Finlayson et al., 2018). La mancanza di interpretabilità di molte soluzioni basate sull’IA aggrava questo problema, perché rende difficile capire come siano arrivate alla conclusione sbagliata.

Il processo decisionale di acquisto

Finora più di 100 prodotti basati sull’IA hanno ottenuto il marchio di conformità europea (CE) o l’autorizzazione della Food and Drug Administration (FDA) statunitense. Questi prodotti possono essere trovati in database online che vengono continuamente aggiornati, nei quali è possibile effettuare ricerche, che sono curati dalla FDA (Center for Devices & Radiological Health, senza data), dall’American College of Radiology (Assess-AI, senza data) e da altri enti (AI for Radiology, senza data, The Medical Futurist, senza data; E. Wu et al., 2021). Il numero crescente di prodotti disponibili, la complessità intrinseca di molte di queste soluzioni e il fatto che molte persone che solitamente prendono decisioni di acquisto negli ospedali non hanno familiarità con la valutazione di tali prodotti rendono importante e necessaria una riflessione attenta sulle decisioni di acquisto. Tali decisioni dovranno essere prese sulla base del contributo degli operatori sanitari, dei professionisti informatici (IT) e dei professionisti della gestione, della finanza, del diritto e delle risorse umane all’interno degli ospedali.

Decidere se acquistare una soluzione basata sull’IA in radiologia, nonché quale tra il crescente numero di soluzioni disponibili in commercio acquistare, include considerazioni relative a qualità, sicurezza e aspetti economici. Negli ultimi anni sono state emanate varie linee guida per aiutare i potenziali acquirenti a prendere queste decisioni (A Buyer's Guide to AI in Health and Care, 2020; Omoumi et al., 2021; Reddy et al., 2021), e queste linee guida probabilmente si evolveranno in futuro al mutare delle aspettative dei clienti, degli organismi di regolamentazione e dei soggetti interessati che prenderanno parte ai processi decisionali relativi ai rimborsi.

Innanzitutto, il potenziale acquirente deve avere ben chiaro qual è il problema e se l’IA è l’approccio giusto per risolverlo o se esistono alternative tutto sommato più vantaggiose. Se l'IA è l'approccio appropriato, gli acquirenti devono sapere esattamente qual è l'ambito d’azione della soluzione proposta da un potenziale prodotto basato sull'IA, ovvero per quale problema specifico è stata progettata tale soluzione e in quali circostanze specifiche. Ciò significa che occorre considerare se la soluzione sia stata pensata per lo screening, per la diagnosi, per il monitoraggio, per consigliare trattamenti o per altri scopi. Va quindi considerato anche chi saranno gli utilizzatori della soluzione e che tipo di qualifiche o di formazione specifica devono avere per essere in grado di utilizzare la soluzione e interpretarne i risultati. Deve essere chiaro all’acquirente se la soluzione sia destinata a sostituire determinate mansioni che normalmente verrebbero eseguite dall’utente finale o a fungere da doppio lettore, come meccanismo di triage o per altre attività come il controllo di qualità. L’acquirente deve anche capire se la soluzione è destinata a fornire "nuove" informazioni (ovvero informazioni che altrimenti non sarebbero disponibili senza la soluzione), a migliorare le prestazioni di un'attività esistente oltre le prestazioni di una soluzione umana o di un'altra soluzione non basata sull'IA o a ottenere un risparmio in termini di tempo o altre risorse.

L’acquirente deve inoltre avere accesso a informazioni che gli consentano di valutare i potenziali vantaggi della soluzione di IA, e in ciò dovrebbe essere supportato da evidenze scientifiche pubblicate sull’efficacia e sull’efficienza economica della soluzione. Il modo in cui ciò verrà fatto dipenderà in gran parte dalla soluzione stessa e dal contesto in cui si prevede che venga implementata, ma sono disponibili linee guida al riguardo (National Institute for Health and Care Excellence [NICE], senza data). Alcune domande da porre in questi casi sarebbero: Quanto influirà la soluzione sulla gestione dei pazienti? Migliorerà le prestazioni diagnostiche? Farà risparmiare tempo e denaro? Influirà sulla qualità della vita dei pazienti? Deve inoltre essere chiaro all’acquirente chi esattamente dovrebbe trarre beneficio dall’uso di questa soluzione (i radiologi? i medici clinici? i pazienti? il sistema sanitario o la società tutta?).

Come per qualsiasi intervento sanitario, tutte le soluzioni basate sull’IA comportano dei rischi, che vanno chiariti all’acquirente. Alcuni di questi rischi potrebbero avere conseguenze legali, come il rischio di diagnosi errate. Questi rischi devono essere quantificati e i potenziali acquirenti devono elaborare un piano per affrontarli, delineando anche un quadro di responsabilità all’interno delle organizzazioni che implementano queste soluzioni. Gli acquirenti devono inoltre essere certi di aver ben compreso i potenziali effetti negativi sulla formazione dei radiologi e i cambiamenti che si avranno nel flusso di lavoro dei radiologi con l’uso di queste soluzioni.

Anche le specifiche del design della soluzione di IA sono importanti ai fini della decisione se acquistarla o meno. Queste sono: la robustezza della soluzione, le differenze tra fornitori e parametri di scansione, le circostanze in cui è stato addestrato l'algoritmo (compresi i possibili fattori confondenti) e il modo in cui sono state valutate le prestazioni. Deve essere chiaro agli acquirenti anche se e come sono state prese in considerazione le potenziali fonti di bias durante lo sviluppo. Poiché una caratteristica fondamentale delle soluzioni basate sull'IA è la loro capacità di apprendere continuamente da nuovi dati, all'acquirente dovrebbe essere chiaro anche se e come esattamente verrà incorporato questo riaddestramento nella soluzione nel tempo, nonché se sarà necessaria o meno una nuova approvazione normativa ad ogni iterazione e se sarà necessario o meno un riaddestramento, ad esempio, in seguito a modifiche alle apparecchiature di imaging presso la struttura dell'acquirente.

I principali punti di forza di molte soluzioni basate sull’IA sono la facilità d’uso e il miglioramento dei flussi di lavoro, pertanto i potenziali acquirenti devono esaminare attentamente il modo in cui queste soluzioni dovranno essere integrate nei flussi di lavoro esistenti, compresa la loro interoperabilità con i sistemi PACS e di gestione delle cartelle cliniche elettroniche. Anche sapere se la soluzione richiede o meno altro hardware (ad es., unità di elaborazione grafica) o software (ad es., per la visualizzazione dei risultati della soluzione) o se può essere facilmente integrata nell'infrastruttura informatica esistente dell'organizzazione dell’acquirente influisce sul costo complessivo della soluzione per l’acquirente ed è quindi un altro aspetto di fondamentale importanza da considerare. Inoltre, l’acquirente dovrebbe sapere che livello di interazione manuale è richiesto, sia in circostanze normali che per la risoluzione dei problemi. Devono essere coinvolti nel processo di acquisto tutti i probabili utilizzatori della soluzione di IA, per essere certi che la conoscano bene e che la soluzione soddisfi i loro standard etici e le loro esigenze professionali.

Dal punto di vista normativo, dovrebbe essere chiaro all’acquirente se la soluzione sia conforme alle normative sui dispositivi medici e sulla protezione dei dati. La soluzione è stata approvata nel Paese dell'acquirente? Se sì, in quale classe di rischio? Gli acquirenti devono anche prendere in considerazione l’ipotesi di predisporre mappe del flusso di dati che mostrino il percorso dei dati in tutte le fasi del funzionamento della soluzione basata sull’IA, includendo tutti coloro che avranno accesso ai dati.

Infine, ci sono altri fattori da considerare che non sono necessariamente aspetti esclusivi delle soluzioni basate sull’IA, ma che gli acquirenti potrebbero conoscere già da altri tipi di soluzioni: ad esempio, il tipo di licenza della soluzione, le modalità di formazione degli utenti sull'uso della soluzione, le modalità di manutenzione della soluzione, le modalità di gestione dei problemi e se sono previsti costi aggiuntivi in caso di scalabilità dell'implementazione della soluzione (ad es., se si passa ad utilizzare la soluzione per un numero maggiore di apparecchiature o utenti). In questo modo, il potenziale acquirente potrà prevedere i costi attuali e futuri della soluzione.

Tendenze future

Conclusione

L’IA si è dimostrata promettente nell’influenzare positivamente quasi ogni aspetto del lavoro di un reparto di radiologia, dalla programmazione degli esami dei pazienti e la determinazione del protocollo all’interpretazione delle immagini e alla formulazione delle diagnosi. Tuttavia, la promettente ricerca sugli strumenti basati sull’IA in radiologia non è stata ancora ampiamente tradotta in adozione nella pratica di routine a causa di una serie di questioni complesse e parzialmente interconnesse. Esistono potenziali soluzioni per molte di queste problematiche, ma molte di queste soluzioni richiedono ulteriori perfezionamenti e test. Nel frattempo, stanno emergendo linee guida per aiutare i potenziali utenti di soluzioni basate sull’IA in radiologia a orientarsi nel crescente numero di prodotti commerciali disponibili. Questo ne incoraggia l’adozione in scenari reali, consentendo così di scoprire le loro vere potenzialità e di identificare e affrontare i loro punti deboli in modo sicuro ed efficace. A mano a mano che verranno apportati questi miglioramenti, questi strumenti probabilmente si evolveranno per gestire dati più diversificati, si integreranno in flussi di lavoro consolidati, diventeranno più trasparenti e, in definitiva, più utili per aumentare l’efficienza e migliorare l’assistenza dei pazienti.

Bibliografia 

AAMC Report Reinforces Mounting Physician Shortage. (2021). AAMC. https://www.aamc.org/news-insights/press- releases/aamc-report-reinforces-mounting-physician-shortage

A buyer’s guide to AI in health and care. (2020). NHS Transformation Directorate. https://www.nhsx.nhs.uk/ai-lab/ explore-all-resources/adopt-ai/a-buyers-guide-to-ai-in-health- and-care/

Advanced AI solutions for radiology. (n.d.). Calantic Website. Retrieved July 3, 2022, from https://aivisions.calantic.com/

Aggarwal, R., Farag, S., Martin, G., Ashrafian, H., & Darzi, A. (2021). Patient Perceptions on Data Sharing and Applying Artificial Intelligence to Health Care Data: Cross-sectional Survey. Journal of Medical Internet Research, 23(8), e26162. https://doi.org/10.2196/26162

Aggarwal, R., Sounderajah, V., Martin, G., Ting, D. S. W., Karthikesalingam, A., King, D., Ashrafian, H., & Darzi, A. (2021). Diagnostic accuracy of deep learning in medical imaging: a systematic review and meta-analysis. NPJ Digital Medicine, 4(1), 65. https://doi.org/10.1038/s41746-021-00438-z

Agnes, S. A., Anitha, J., Pandian, S. I. A., & Peter, J. D. (2019). Classification of Mammogram Images Using Multiscale all Convolutional Neural Network (MA-CNN). Journal of Medical Systems, 44(1), 30. https://doi.org/10.1007/s10916-019-1494-z

AIF360: A comprehensive set of fairness metrics for datasets and machine learning models, explanations for these metrics, and algorithms to mitigate bias in datasets and models. (n.d.). Github. Retrieved June 11, 2022, from https://github.com/Trusted-AI/AIF360

AI for radiology. (n.d.). Retrieved June 26, 2022, from https://grand-challenge.org/aiforradiology/?subspeciality=Abdomen&modality=All&ce_ under=All&ce_class=All&fda_class=All&sort_by=last %20 modified&search=

Akagi, M., Nakamura, Y., Higaki, T., Narita, K., Honda, Y., Zhou, J., Yu, Z., Akino, N., & Awai, K. (2019). Deep learning reconstruction improves image quality of abdominal ultra- high-resolution CT. European Radiology, 29(11), 6163–6171. https://doi.org/10.1007/s00330-019-06170-3

Akkus, Z., Cai, J., Boonrod, A., Zeinoddini, A., Weston, A. D., Philbrick, K. A., & Erickson, B. J. (2019). A Survey of Deep- Learning Applications in Ultrasound: Artificial Intelligence- Powered Ultrasound for Improving Clinical Workflow. Journal of the American College of Radiology: JACR, 16(9 Pt B), 1318–1328. https://doi.org/10.1016/j.jacr.2019.06.004

Al-Antari, M. A., Al-Masni, M. A., & Kim, T.-S. (2020). Deep Learning Computer-Aided Diagnosis for Breast Lesion in Digital Mammogram. Advances in Experimental Medicine and Biology, 1213, 59–72. https://doi.org/10.1007/978-3-030-33128-3_4

Alexander, A., Jiang, A., Ferreira, C., & Zurkiya, D. (2020). An Intelligent Future for Medical Imaging: A Market Outlook on Artificial Intelligence for Medical Imaging. Journal of the American College of Radiology: JACR, 17(1 Pt B), 165–170. https:// doi.org/10.1016/j.jacr.2019.07.019

Allen, B., Agarwal, S., Coombs, L., Wald, C., & Dreyer, K. (2021). 2020 ACR Data Science Institute Artificial Intelligence Journal of the American College of Radiology: JACR

Alzubaidi, L., Zhang, J., Humaidi, A. J., Al-Dujaili, A., Duan, Y., Al-Shamma, O., Santamaría, J., Fadhel, M. A., Al-Amidie, M., & Farhan, L. (2021). Review of deep learning: concepts, CNN architectures, challenges, applications, future directions.

Arbabshirani, M. R., Fornwalt, B. K., Mongelluzzo, G. J., Suever, J. D., Geise, B. D., Patel, A. A., & Moore, G. J. (2018). Advanced machine learning in action: identification of intracranial hemorrhage on computed tomography scans of the head with clinical workflow integration. NPJ Digital Medicine, 1, 9. https://doi.org/10.1038/s41746-017-0015-z

Ardila, D., Kiraly, A. P., Bharadwaj, S., Choi, B., Reicher, J. J., Peng, L., Tse, D., Etemadi, M., Ye, W., Corrado, G., Naidich, D. P., & Shetty, S. (2019). End-to-end lung cancer screening with three-dimensional deep learning on low-dose chest computed tomography. Nature Medicine, 25(6), 954–961. https://doi. org/10.1038/s41591-019-0447-x

Arefan, D., Mohamed, A. A., Berg, W. A., Zuley, M. L., Sumkin, J. H., & Wu, S. (2020). Deep learning modeling using normal mammograms for predicting breast cancer risk. Medical Physics, 47(1), 110–118. https://doi.org/10.1002/mp.13886

Assess-AI. (n.d.). Retrieved July 2, 2022, from https://www. acrdsi.org/DSI-Services/Assess-AI

Azer, S. A. (2019). Deep learning with convolutional neural networks for identification of liver masses and hepatocellular carcinoma: A systematic review. World Journal of Gastrointestinal Oncology, 11(12), 1218–1230. https://doi.org/10.4251/wjgo.v11. i12.1218

Bacchi, S., Zerner, T., Oakden-Rayner, L., Kleinig, T., Patel, S., & Jannes, J. (2020). Deep Learning in the Prediction of Ischaemic Stroke Thrombolysis Functional Outcomes: A Pilot Study. Academic Radiology, 27(2), e19–e23. https://doi. org/10.1016/j.acra.2019.03.015

Baccianella, S., & Gough, T. (n.d.). Why cloud computing is the best option for hospitals adopting AI. Retrieved June 11, 2022, from https://www.aidence.com/articles/cloud-best-option- imaging-ai/

Balint, B. J., Steenburg, S. D., Lin, H., Shen, C., Steele, J. L., & Gunderman, R. B. (2014). Do telephone call interruptions have an impact on radiology resident diagnostic accuracy? Academic Radiology, 21(12), 1623–1628. https://doi.org/10.1016/j. acra.2014.08.001

Baltruschat, I., Steinmeister, L., Nickisch, H., Saalbach, A., Grass, M., Adam, G., Knopp, T., & Ittrich, H. (2021). Smart chest X-ray worklist prioritization using artificial intelligence: a clinical workflow simulation. European Radiology, 31(6), 3837– 3845. https://doi.org/10.1007/s00330-020-07480-7

Bash, S., Johnson, B., Gibbs, W., Zhang, T., Shankaranarayanan, A., & Tanenbaum, L. N. (2021). Deep Learning Image Processing Enables 40 % Faster Spinal MR Scans Which Match or Exceed Quality of Standard of Care : A Prospective Multicenter Multireader Study. Clinical Neuroradiology. https://doi.org/10.1007/s00062-021-01121-2

Bash, S., Wang, L., Airriess, C., Zaharchuk, G., Gong, E., Shankaranarayanan, A., & Tanenbaum, L. N. (2021). Deep Learning Enables 60 % Accelerated Volumetric Brain MRI While Preserving Quantitative Performance: A Prospective, Multicenter, Multireader Trial. AJNR. American Journal of Neuroradiology, 42(12), 2130–2137. https://doi.org/10.3174/ajnr.A7358

Boas, F. E., & Fleischmann, D. (2012). CT artifacts: causes and reduction techniques. Imaging in Medicine, 4(2), 229–240. https://doi.org/10.2217/iim.12.13

Boland, G. W., Duszak, R., Jr, & Kalra, M. (2014). Protocol design and optimization. Journal of the American College of Radiology: JACR, 11(5), 440–441. https://doi.org/10.1016/j. jacr.2014.01.021

Bonavita, I., Rafael-Palou, X., Ceresa, M., Piella, G., Ribas, V., & González Ballester, M. A. (2020). Integration of convolutional neural networks for pulmonary nodule malignancy assessment in a lung cancer classification pipeline. Computer Methods and Programs in Biomedicine, 185, 105172. https://doi.org/10.1016/j.cmpb.2019.105172

Bozkurt, S., Cahan, E. M., Seneviratne, M. G., Sun, R., Lossio- Ventura, J. A., Ioannidis, J. P. A., & Hernandez-Boussard, T. (2020). Reporting of demographic data and representativeness in machine learning models using electronic health records.
Journal of the American Medical Informatics Association: JAMIA, 27(12), 1878–1884. https://doi.org/10.1093/jamia/ocaa164

Brown, A. D., & Marotta, T. R. (2018). Using machine learning for sequence-level automated MRI protocol selection in neuroradiology. Journal of the American Medical Informatics Association: JAMIA, 25(5), 568–571. https://doi.org/10.1093/ jamia/ocx125

Bruls, R. J. M., & Kwee, R. M. (2020). Workload for radiologists during on-call hours: dramatic increase in the past 15 years.
Insights into Imaging, 11(1), 121. https://doi.org/10.1186/ s13244-020-00925-z

Buruk, B., Ekmekci, P. E., & Arda, B. (2020). A critical perspective on guidelines for responsible and trustworthy artificial intelligence. Medicine, Health Care, and Philosophy, 23(3), 387–399. https://doi.org/10.1007/s11019-020-09948-1

Cadario, R., Longoni, C., & Morewedge, C. K. (2021). Understanding, explaining, and utilizing medical artificial intelligence. Nature Human Behaviour, 5(12), 1636–1642. https://doi.org/10.1038/s41562-021-01146-0

Center for Devices, & Radiological Health. (n.d.). Artificial Intelligence and Machine Learning (AI/ML)-Enabled Medical Devices. U.S. Food and Drug Administration; FDA. Retrieved July 2, 2022, from https://www.fda.gov/medical-devices/software- medical-device-samd/artificial-intelligence-and-machine- learning-aiml-enabled-medical-devices

Char, D. S., Shah, N. H., & Magnus, D. (2018). Implementing Machine Learning in Health Care - Addressing Ethical Challenges. The New England Journal of Medicine, 378(11), 981–983. https://doi.org/10.1056/NEJMp1714229

Chen, H., Zhang, Y., Kalra, M. K., Lin, F., Chen, Y., Liao, P., Zhou, J., & Wang, G. (2017). Low-Dose CT With a Residual Encoder-Decoder Convolutional Neural Network. IEEE Transactions on Medical Imaging, 36(12), 2524–2535. https://doi.org/10.1109/TMI.2017.2715284

Chen, Y., Stavropoulou, C., Narasinkan, R., Baker, A., & Scarbrough, H. (2021). Professionals’ responses to the introduction of AI innovations in radiology and their implications for future adoption: a qualitative study. BMC Health Services Research, 21(1), 813. https://doi.org/10.1186/ s12913-021-06861-y

Choe, J., Lee, S. M., Do, K.-H., Lee, G., Lee, J.-G., Lee, S. M., & Seo, J. B. (2019). Deep Learning-based Image Conversion of CT Reconstruction Kernels Improves Radiomics Reproducibility for Pulmonary Nodules or Masses. Radiology, 292(2), 365–373. https://doi.org/10.1148/radiol.2019181960

Choi, K. S., Choi, S. H., & Jeong, B. (2019). Prediction of IDH genotype in gliomas with dynamic susceptibility contrast perfusion MR imaging using an explainable recurrent neural network. Neuro-Oncology, 21(9), 1197–1209. https://doi.org/10.1093/neuonc/noz095

Chong, L. R., Tsai, K. T., Lee, L. L., Foo, S. G., & Chang, P. C. (2020). Artificial Intelligence Predictive Analytics in the Management of Outpatient MRI Appointment No-Shows. AJR. American Journal of Roentgenology, 215(5), 1155–1162. https://doi.org/10.2214/AJR.19.22594

Cikes, M., Sanchez-Martinez, S., Claggett, B., Duchateau, N., Piella, G., Butakoff, C., Pouleur, A. C., Knappe, D., Biering- Sørensen, T., Kutyifa, V., Moss, A., Stein, K., Solomon, S. D., & Bijnens, B. (2019). Machine learning-based phenogrouping in heart failure to identify responders to cardiac resynchronization therapy. European Journal of Heart Failure, 21(1), 74–85. https://doi.org/10.1002/ejhf.1333

Ciompi, F., Chung, K., van Riel, S. J., Setio, A. A. A., Gerke, P. K., Jacobs, C., Scholten, E. T., Schaefer-Prokop, C., Wille,
M. M. W., Marchianò, A., Pastorino, U., Prokop, M., & van Ginneken, B. (2017). Towards automatic pulmonary nodule management in lung cancer screening with deep learning. Scientific Reports, 7, 46479. https://doi.org/10.1038/srep46479

Clinical radiology UK workforce census 2019 report. (2019). https://www.rcr.ac.uk/publication/clinical-radiology-uk- workforce-census-2019-report

Cloud security for healthcare services. (2021, January 14). ENISA. https://www.enisa.europa.eu/publications/cloud- security-for-healthcare-services/

CONSORT-AI and SPIRIT-AI Steering Group. (2019). Reporting guidelines for clinical trials evaluating artificial intelligence interventions are needed. Nature Medicine, 25(10), 1467–1468. https://doi.org/10.1038/s41591-019-0603-3

Curated marketplace. (2018, May 22). Blackford. https://www.blackfordanalysis.com/applications/

Dance, A. (2021). AI spots cell structures that humans can’t. Nature. 592 (7852), 154–155.

Dantas, L. F., Fleck, J. L., Cyrino Oliveira, F. L., & Hamacher, S. (2018). No-shows in appointment scheduling - a systematic literature review. Health Policy, 122(4), 412–421. https://doi.org/10.1016/j.healthpol.2018.02.002

Deák, Z., Grimm, J. M., Treitl, M., Geyer, L. L., Linsenmaier, U., Körner, M., Reiser, M. F., & Wirth, S. (2013). Filtered back projection, adaptive statistical iterative reconstruction, and a model-based iterative reconstruction in abdominal CT: an experimental clinical study. Radiology, 266(1), 197–206. https://doi.org/10.1148/radiol.12112707

Dembrower, K., Liu, Y., Azizpour, H., Eklund, M., Smith, K., Lindholm, P., & Strand, F. (2020). Comparison of a Deep
Learning Risk Score and Standard Mammographic Density Score for Breast Cancer Risk Prediction. Radiology, 294(2), 265–272. https://doi.org/10.1148/radiol.2019190872

Do, B. H., Langlotz, C., & Beaulieu, C. F. (2017). Bone Tumor Diagnosis Using a Naïve Bayesian Model of Demographic and Radiographic Features. Journal of Digital Imaging, 30(5), 640–647. https://doi.org/10.1007/s10278-017-0001-7

Dou, Q., Yu, L., Chen, H., Jin, Y., Yang, X., Qin, J., & Heng, P.-A. (2017). 3D deeply supervised network for automated segmentation of volumetric medical images. Medical Image Analysis, 41, 40–54. https://doi.org/10.1016/j. media.2017.05.001

Eche, T., Schwartz, L. H., Mokrane, F.-Z., & Dercle, L. (2021). Toward Generalizability in the Deployment of Artificial Intelligence in Radiology: Role of Computation Stress Testing to Overcome Underspecification. Radiology. Artificial Intelligence, 3(6), e210097. https://doi.org/10.1148/ryai.2021210097

England, N. H. S., & Improvement, N. H. S. (2019). NHS diagnostic waiting times and activity data. NHS. https://www. england.nhs.uk/statistics/wp-content/uploads/sites/2/2021/12/ DWTA-Report-October-2021_M43D4.pdf

Esmaeilzadeh, P. (2020). Use of AI-based tools for healthcare purposes: a survey study from consumers’ perspectives. BMC Medical Informatics and Decision Making, 20(1), 170. https://doi. org/10.1186/s12911-020-01191-1

Esses, S. J., Lu, X., Zhao, T., Shanbhogue, K., Dane, B., Bruno, M., & Chandarana, H. (2018). Automated image quality evaluation of T2 -weighted liver MRI utilizing deep learning architecture. Journal of Magnetic Resonance Imaging: JMRI, 47(3), 723–728. https://doi.org/10.1002/jmri.25779

European Society of Radiology (ESR). (2022). Current practical experience with artificial intelligence in clinical radiology:
a survey of the European Society of Radiology. Insights into Imaging, 13(1), 107. https://doi.org/10.1186/s13244-022- 01247-y

Faron, A., Sichtermann, T., Teichert, N., Luetkens, J. A., Keulers, A., Nikoubashman, O., Freiherr, J., Mpotsaris, A., & Wiesmann, M. (2020). Performance of a Deep-Learning Neural Network to Detect Intracranial Aneurysms from 3D TOF-MRA Compared to Human Readers. Clinical Neuroradiology, 30(3), 591–598. https://doi.org/10.1007/s00062-019-00809-w

Feng, J., Phillips, R. V., Malenica, I., Bishara, A., Hubbard, A. E., Celi, L. A., & Pirracchio, R. (2022). Clinical artificial intelligence quality improvement: towards continual monitoring and updating of AI algorithms in healthcare. NPJ Digital Medicine, 5(1), 66. https://doi.org/10.1038/s41746-022- 00611-y

Finlayson, S. G., Chung, H. W., Kohane, I. S., & Beam, A. L. (2018). Adversarial Attacks Against Medical Deep Learning Systems. In arXiv [cs.CR]. arXiv. https://doi.org/10.1145/nnnnnnn. 

Flanders, A. E., Prevedello, L. M., Shih, G., Halabi, S. S., Kalpathy-Cramer, J., Ball, R., Mongan, J. T., Stein, A., Kitamura, F. C., Lungren, M. P., Choudhary, G., Cala, L., Coelho, L., Mogensen, M., Morón, F., Miller, E., Ikuta, I., Zohrabian, V., McDonnell, O., … RSNA-ASNR 2019 Brain Hemorrhage CT Annotators. (2020). Construction of a Machine Learning Dataset through Collaboration: The RSNA 2019 Brain CT Hemorrhage Challenge. Radiology. Artificial Intelligence, 2(3), e190211. https://doi.org/10.1148/ryai.2020190211

Freeman, K., Geppert, J., Stinton, C., Todkill, D., Johnson, S., Clarke, A., & Taylor-Phillips, S. (2021). Use of artificial intelligence for image analysis in breast cancer screening programmes: systematic review of test accuracy. BMJ , 374, n1872. https://doi.org/10.1136/bmj.n1872

General Data Protection Regulation (GDPR) – Official Legal Text. (2016, July 13). General Data Protection Regulation (GDPR). https://gdpr-info.eu/

Ghafur, S., Van Dael, J., Leis, M., Darzi, A., & Sheikh, A. (2020). Public perceptions on data sharing: key insights from the UK and the USA. The Lancet. Digital Health, 2(9), e444–e446. https://doi.org/10.1016/S2589-7500(20)30161-8

Ghani, M. U., & Clem Karl, W. (2019). Fast Enhanced CT Metal Artifact Reduction using Data Domain Deep Learning. In arXiv [cs.CV]. arXiv. http://arxiv.org/abs/1904.04691

Ghassemi, M., Oakden-Rayner, L., & Beam, A. L. (2021). The false hope of current approaches to explainable artificial intelligence in health care. The Lancet. Digital Health, 3(11), e745–e750. https://doi.org/10.1016/S2589-7500(21)00208-9

Ginat, D. T. (2020). Analysis of head CT scans flagged by deep learning software for acute intracranial hemorrhage. Neuroradiology, 62(3), 335–340. https://doi.org/10.1007/ s00234-019-02330-w

Goebel, J., Stenzel, E., Guberina, N., Wanke, I., Koehrmann, M., Kleinschnitz, C., Umutlu, L., Forsting, M., Moenninghoff, C., & Radbruch, A. (2018). Automated ASPECT rating: comparison between the Frontier ASPECT Score software and the Brainomix software. Neuroradiology, 60(12), 1267–1272. https://doi.org/10.1007/s00234-018-2098-x

Habli, I., Lawton, T., & Porter, Z. (2020). Artificial intelligence in health care: accountability and safety. Bulletin of the World Health Organization, 98(4), 251–256. https://doi.org/10.2471/ BLT.19.237487

Halabi, S. S., Prevedello, L. M., Kalpathy-Cramer, J., Mamonov, A. B., Bilbily, A., Cicero, M., Pan, I., Pereira, L. A., Sousa, R. T., Abdala, N., Kitamura, F. C., Thodberg, H. H., Chen, L., Shih, G., Andriole, K., Kohli, M. D., Erickson, B. J., & Flanders, A. E. (2019). The RSNA Pediatric Bone Age Machine Learning Challenge. Radiology, 290(2), 498–503. https://doi. org/10.1148/radiol.2018180736

Hargreaves, B. A., Worters, P. W., Pauly, K. B., Pauly, J. M., Koch, K. M., & Gold, G. E. (2011). Metal-induced artifacts in MRI. AJR. American Journal of Roentgenology, 197(3), 547–555. https://doi.org/10.2214/AJR.11.7364

Harry, E., Sinsky, C., Dyrbye, L. N., Makowski, M. S., Trockel, M., Tutty, M., Carlasare, L. E., West, C. P., & Shanafelt, T. D. (2021). Physician Task Load and the Risk of Burnout Among US Physicians in a National Survey. Joint Commission Journal on Quality and Patient Safety / Joint Commission Resources, 47(2), 76–85. https://doi.org/10.1016/j.jcjq.2020.09.011

Hata, A., Yanagawa, M., Yamagata, K., Suzuki, Y., Kido, S., Kawata, A., Doi, S., Yoshida, Y., Miyata, T., Tsubamoto, M., Kikuchi, N., & Tomiyama, N. (2021). Deep learning algorithm for detection of aortic dissection on non-contrast-enhanced CT. European Radiology, 31(2), 1151–1159. https://doi.org/10.1007/ s00330-020-07213-w

Hauptmann, A., Arridge, S., Lucka, F., Muthurangu, V., & Steeden, J. A. (2019). Real-time cardiovascular MR with
spatio-temporal artifact suppression using deep learning-proof of concept in congenital heart disease. Magnetic Resonance in Medicine: Official Journal of the Society of Magnetic Resonance in Medicine / Society of Magnetic Resonance in Medicine, 81(2), 1143–1156. https://doi.org/10.1002/mrm.27480

Health Ethics & Governance. (2021, June 28). Ethics and governance of artificial intelligence for health. World Health Organization. https://www.who.int/publications/i/item/9789240029200

He, L., Li, H., Dudley, J. A., Maloney, T. C., Brady, S. L., Somasundaram, E., Trout, A. T., & Dillman, J. R. (2019). Machine Learning Prediction of Liver Stiffness Using Clinical and T2-Weighted MRI Radiomic Data. AJR. American Journal of Roentgenology, 213(3), 592–601. https://doi.org/10.2214/ AJR.19.21082

Herent, P., Schmauch, B., Jehanno, P., Dehaene, O., Saillard, C., Balleyguier, C., Arfi-Rouche, J., & Jégou, S. (2019).
Detection and characterization of MRI breast lesions using deep learning. Diagnostic and Interventional Imaging, 100(4), 219–225. https://doi.org/10.1016/j.diii.2019.02.008

Hinton, B., Ma, L., Mahmoudzadeh, A. P., Malkov, S., Fan, B., Greenwood, H., Joe, B., Lee, V., Kerlikowske, K., & Shepherd,
J. (2019). Deep learning networks find unique mammographic differences in previous negative mammograms between interval and screen-detected cancers: a case-case study. Cancer Imaging: The Official Publication of the International Cancer Imaging Society, 19(1), 41. https://doi.org/10.1186/s40644-019-0227-3

Holzinger, A., Biemann, C., Pattichis, C. S., & Kell, D. B. (2017). What do we need to build explainable AI systems for the medical domain? In arXiv [cs.AI]. arXiv. http://arxiv.org/ abs/1712.09923

Hötker, A. M., Da Mutten, R., Tiessen, A., Konukoglu, E., & Donati, O. F. (2021). Improving workflow in prostate MRI:
AI-based decision-making on biparametric or multiparametric MRI. Insights into Imaging, 12(1), 112. https://doi.org/10.1186/ s13244-021-01058-7

Huang, S.-C., Kothari, T., Banerjee, I., Chute, C., Ball, R. L., Borus, N., Huang, A., Patel, B. N., Rajpurkar, P., Irvin, J., Dunnmon, J., Bledsoe, J., Shpanskaya, K., Dhaliwal, A., Zamanian, R., Ng, A. Y., & Lungren, M. P. (2020). PENet-a scalable deep-learning model for automated diagnosis of pulmonary embolism using volumetric CT imaging. NPJ Digital Medicine, 3, 61. https://doi.org/10.1038/s41746-020-0266-y

Huang, S.-C., Pareek, A., Seyyedi, S., Banerjee, I., & Lungren, M. P. (2020). Fusion of medical imaging and electronic health records using deep learning: a systematic review and implementation guidelines. NPJ Digital Medicine, 3, 136. https:// doi.org/10.1038/s41746-020-00341-z

Huisman, M., Ranschaert, E., Parker, W., Mastrodicasa, D., Koci, M., Pinto de Santos, D., Coppola, F., Morozov, S., Zins, M., Bohyn, C., Koç, U., Wu, J., Veean, S., Fleischmann, D., Leiner, T., & Willemink, M. J. (2021). An international survey on AI in radiology in 1,041 radiologists and radiology residents part 1: fear of replacement, knowledge, and attitude. European Radiology, 31(9), 7058–7066. https://doi.org/10.1007/s00330-021-07781-5

Hu, S.-Y., Santus, E., Forsyth, A. W., Malhotra, D., Haimson, J., Chatterjee, N. A., Kramer, D. B., Barzilay, R., Tulsky,
J. A., & Lindvall, C. (2019). Can machine learning improve patient selection for cardiac resynchronization therapy? PloS One, 14(10), e0222397. https://doi.org/10.1371/journal. pone.0222397

Hwang, E. J., Nam, J. G., Lim, W. H., Park, S. J., Jeong, Y. S., Kang, J. H., Hong, E. K., Kim, T. M., Goo, J. M., Park, S., Kim, K. H., & Park, C. M. (2019). Deep Learning for Chest Radiograph Diagnosis in the Emergency Department. Radiology, 293(3), 573–580. https://doi.org/10.1148/radiol.2019191225

Hwang, E. J., Park, S., Jin, K.-N., Kim, J. I., Choi, S. Y., Lee, J. H., Goo, J. M., Aum, J., Yim, J.-J., Park, C. M., & Deep Learning- Based Automatic Detection Algorithm Development and Evaluation Group. (2019). Development and Validation of a Deep Learning-based Automatic Detection Algorithm for Active Pulmonary Tuberculosis on Chest Radiographs. Clinical Infectious Diseases: An Official Publication of the Infectious Diseases Society of America, 69(5), 739–747. https://doi.org/10.1093/cid/ciy967

Hwang, S., Kim, H.-E., Jeong, J., & Kim, H.-J. (2016). A novel approach for tuberculosis screening based on deep convolutional neural networks. In G. D. Tourassi & S. G. Armato (Eds.), Medical Imaging 2016: Computer-Aided Diagnosis. SPIE. https://doi.org/10.1117/12.2216198

IBM Watson Studio - Model Risk Management. (n.d.). Retrieved June 11, 2022, from https://www.ibm.com/cloud/ watson-studio/model-risk-management

Imaging AI Marketplace - overview. (n.d.). Retrieved June 11, 2022, from https://www.ibm.com/products/imaging-ai- marketplace

Jamaludin, A., Lootus, M., Kadir, T., Zisserman, A., Urban, J., Battié, M. C., Fairbank, J., McCall, I., & Genodisc Consortium. (2017). ISSLS PRIZE IN BIOENGINEERING SCIENCE 2017:
Automation of reading of radiological features from magnetic resonance images (MRIs) of the lumbar spine without human intervention is comparable with an expert radiologist. European Spine Journal: Official Publication of the European Spine Society, the European Spinal Deformity Society, and the European Section of the Cervical Spine Research Society, 26(5), 1374–1383. https:// doi.org/10.1007/s00586-017-4956-3

Kaissis, G. A., Makowski, M. R., Rückert, D., & Braren, R. F. (2020). Secure, privacy-preserving and federated machine learning in medical imaging. Nature Machine Intelligence, 2(6), 305–311. https://doi.org/10.1038/s42256-020-0186-1

Kaissis, G., Ziller, A., Passerat-Palmbach, J., Ryffel, T., Usynin, D., Trask, A., Lima, I., Mancuso, J., Jungmann, F., Steinborn, M.-M., Saleh, A., Makowski, M., Rueckert, D., & Braren, R. (2021). End-to-end privacy preserving deep learning on multi-institutional medical imaging. Nature Machine Intelligence, 3(6), 473–484. https://doi.org/10.1038/s42256-021-00337-8

Kalra, A., Chakraborty, A., Fine, B., & Reicher, J. (2020). Machine Learning for Automation of Radiology Protocols for Quality and Efficiency Improvement. Journal of the American College of Radiology: JACR, 17(9), 1149–1158. https://doi.org/10.1016/j.jacr.2020.03.012

Kao, P.-Y., Chen, J. W., & Manjunath, B. S. (2019). Improving 3D U-Net for Brain Tumor Segmentation by Utilizing Lesion Prior. In arXiv [cs.CV]. arXiv. http://arxiv.org/abs/1907.00281

Kapoor, N., Lacson, R., & Khorasani, R. (2020). Workflow Applications of Artificial Intelligence in Radiology and an Overview of Available Tools. Journal of the American College of Radiology: JACR, 17(11), 1363–1370. https://doi.org/10.1016/j. jacr.2020.08.016

Kathirvelu, D., Vinupritha, P., & Kalpana, V. (2019). A computer aided diagnosis system for measurement of mandibular cortical thickness on dental panoramic radiographs in prediction of women with low bone mineral density. Journal of Medical Systems, 43(6), 148. https://doi.org/10.1007/s10916-019-1268-7

Ker, J., Singh, S. P., Bai, Y., Rao, J., Lim, T., & Wang, L. (2019). Image Thresholding Improves 3-Dimensional Convolutional Neural Network Diagnosis of Different Acute Brain Hemorrhages on Computed Tomography Scans. Sensors, 19(9). https://doi.org/10.3390/s19092167

Khan, F. A., Majidulla, A., Tavaziva, G., Nazish, A., Abidi, S. K., Benedetti, A., Menzies, D., Johnston, J. C., Khan, A. J., & Saeed, S. (2020). Chest x-ray analysis with deep learning- based software as a triage test for pulmonary tuberculosis: a prospective study of diagnostic accuracy for culture-confirmed disease. The Lancet. Digital Health, 2(11), e573–e581. https://doi.org/10.1016/S2589-7500(20)30221-1

Kim, D. W., Jang, H. Y., Kim, K. W., Shin, Y., & Park, S. H. (2019). Design Characteristics of Studies Reporting the Performance of Artificial Intelligence Algorithms for Diagnostic Analysis of Medical Images: Results from Recently Published Papers. Korean Journal of Radiology: Official Journal of the Korean Radiological Society, 20(3), 405–410. https://doi.org/10.3348/ kjr.2019.0025

Kim, K. H., & Park, S.-H. (2017). Artificial neural network for suppression of banding artifacts in balanced steady-state free precession MRI. Magnetic Resonance Imaging, 37, 139–146. https://doi.org/10.1016/j.mri.2016.11.020

Korteling, J. E. H., van de Boer-Visschedijk, G. C., Blankendaal, R. A. M., Boonekamp, R. C., & Eikelboom, A. R. (2021). Human- versus Artificial Intelligence. Frontiers in Artificial Intelligence 4, 622364. https://doi.org/10.3389/ frai.2021.622364

Kühl, N., Goutier, M., Baier, L., Wolff, C., & Martin, D. (2020). Human vs. supervised machine learning: Who learns patterns faster? In arXiv [cs.AI] arXiv. http://arxiv.org/abs/2012.03661

Kuo, W., Häne, C., Mukherjee, P., Malik, J., & Yuh, E. L. (2019). Expert-level detection of acute intracranial hemorrhage on head computed tomography using deep learning. Proceedings of the National Academy of Sciences of the United States of America, 116(45), 22737–22745. https://doi.org/10.1073/ pnas.1908021116

Langerhuizen, D. W. G., Janssen, S. J., Mallee, W. H., van den Bekerom, M. P. J., Ring, D., Kerkhoffs, G. M. M. J., Jaarsma,
R. L., & Doornberg, J. N. (2019). What Are the Applications and Limitations of Artificial Intelligence for Fracture Detection and Classification in Orthopaedic Trauma Imaging? A Systematic Review. Clinical Orthopaedics and Related Research, 477(11), 2482–2491. https://doi.org/10.1097/CORR.0000000000000848

Lang, N., Zhang, Y., Zhang, E., Zhang, J., Chow, D., Chang, P., Yu, H. J., Yuan, H., & Su, M.-Y. (2019). Differentiation of spinal metastases originated from lung and other cancers using radiomics and deep learning based on DCE-MRI. Magnetic Resonance Imaging, 64, 4–12. https://doi.org/10.1016/j. mri.2019.02.013

Larrazabal, A. J., Nieto, N., Peterson, V., Milone, D. H., & Ferrante, E. (2020). Gender imbalance in medical imaging datasets produces biased classifiers for computer-aided diagnosis. Proceedings of the National Academy of Sciences of the United States of America, 117(23), 12592–12594. https://doi.org/10.1073/pnas.1919012117

Lee, J.-S., Adhikari, S., Liu, L., Jeong, H.-G., Kim, H., & Yoon, S.-J. (2019). Osteoporosis detection in panoramic radiographs using a deep convolutional neural network-based computer- assisted diagnosis system: a preliminary study. Dento Maxillo Facial Radiology, 48(1), 20170344. https://doi.org/10.1259/ dmfr.20170344

Lee, Y. H. (2018). Efficiency Improvement in a Busy Radiology Practice: Determination of Musculoskeletal Magnetic Resonance Imaging Protocol Using Deep-Learning Convolutional Neural Networks. Journal of Digital Imaging, 31(5), 604–610. https://doi.org/10.1007/s10278-018-0066-y

Leiner, T., Bennink, E., Mol, C. P., Kuijf, H. J., & Veldhuis, W. B. (2021). Bringing AI to the clinic: blueprint for a vendor-neutral AI deployment infrastructure. Insights into Imaging, 12(1), 11. https://doi.org/10.1186/s13244-020-00931-1

Lekadir, K., Osuala, R., Gallin, C., Lazrak, N., Kushibar, K., Tsakou, G., Aussó, S., Alberich, L. C., Marias, K., Tsiknakis, M., Colantonio, S., Papanikolaou, N., Salahuddin, Z., Woodruff, H. C., Lambin, P., & Martí-Bonmatí, L. (2021). FUTURE-AI: Guiding Principles and Consensus Recommendations for Trustworthy Artificial Intelligence in Medical Imaging. In arXiv [cs.CV]. arXiv. http://arxiv.org/ abs/2109.09658

Letourneau-Guillon, L., Camirand, D., Guilbert, F., & Forghani, R. (2020). Artificial Intelligence Applications for Workflow, Process Optimization and Predictive Analytics. Neuroimaging Clinics of North America, 30(4), e1–e15. https://doi.org/10.1016/j.nic.2020.08.008

Levin, D. C., Parker, L., & Rao, V. M. (2017). Recent Trends in Imaging Use in Hospital Settings: Implications for Future Planning. Journal of the American College of Radiology: JACR, 14(3), 331–336. https://doi.org/10.1016/j.jacr.2016.08.025

Lindsey, R., Daluiski, A., Chopra, S., Lachapelle, A., Mozer, M., Sicular, S., Hanel, D., Gardner, M., Gupta, A., Hotchkiss, R., & Potter, H. (2018). Deep neural network improves fracture detection by clinicians. Proceedings of the National Academy of Sciences of the United States of America, 115(45), 11591–11596. https://doi.org/10.1073/pnas.1806905115

Liu, F., Tang, J., Ma, J., Wang, C., Ha, Q., Yu, Y., & Zhou, Z. (2021). The application of artificial intelligence to chest medical image analysis. Intelligent Medicine, 1(3), 104–117. https://doi.org/10.1016/j.imed.2021.06.004

Liu, F., Zhou, Z., Samsonov, A., Blankenbaker, D., Larison, W., Kanarek, A., Lian, K., Kambhampati, S., & Kijowski, R. (2018). Deep Learning Approach for Evaluating Knee MR Images: Achieving High Diagnostic Performance for Cartilage Lesion Detection. Radiology, 289(1), 160–169. https://doi.org/10.1148/ radiol.2018172986

Liu, X., Cruz Rivera, S., Moher, D., Calvert, M. J., Denniston, A. K., & SPIRIT-AI and CONSORT-AI Working Group. (2020). Reporting guidelines for clinical trial reports for interventions involving artificial intelligence: the CONSORT-AI extension. Nature Medicine, 26(9), 1364–1374. https://doi.org/10.1038/ s41591-020-1034-x

Liu, X., Faes, L., Kale, A. U., Wagner, S. K., Fu, D. J., Bruynseels, A., Mahendiran, T., Moraes, G., Shamdas, M., Kern, C., Ledsam, J. R., Schmid, M. K., Balaskas, K., Topol, E. J., Bachmann, L. M., Keane, P. A., & Denniston, A. K. (2019). A comparison of deep learning performance against health-care professionals in detecting diseases from medical imaging: a systematic review and meta-analysis. The Lancet. Digital Health, 1(6), e271–e297. https://doi.org/10.1016/S2589- 7500(19)30123-2

Li, X., Shen, L., Xie, X., Huang, S., Xie, Z., Hong, X., & Yu, J. (2020). Multi-resolution convolutional networks for chest X-ray radiograph based lung nodule detection. Artificial Intelligence in Medicine, 103, 101744. https://doi.org/10.1016/j. artmed.2019.101744

Lotan, E., Tschider, C., Sodickson, D. K., Caplan, A. L., Bruno, M., Zhang, B., & Lui, Y. W. (2020). Medical Imaging and Privacy in the Era of Artificial Intelligence: Myth, Fallacy, and the Future. Journal of the American College of Radiology: JACR, 17(9), 1159–1162. https://doi.org/10.1016/j.jacr.2020.04.007

Maegerlein, C., Fischer, J., Mönch, S., Berndt, M., Wunderlich, S., Seifert, C. L., Lehm, M., Boeckh-Behrens, T., Zimmer, C., & Friedrich, B. (2019). Automated Calculation of the Alberta Stroke Program Early CT Score: Feasibility and
Reliability. Radiology, 291(1), 141–148. https://doi.org/10.1148/ radiol.2019181228

Mairhöfer, D., Laufer, M., Simon, P. M., Sieren, M., Bischof, A., Käster, T., Barth, E., Barkhausen, J., & Martinetz, T. (2021). An AI-based Framework for Diagnostic Quality Assessment of Ankle Radiographs. https://openreview.net/ pdf?id=bj04hJss_xZ

Mancio, J., Pashakhanloo, F., El-Rewaidy, H., Jang, J., Joshi, G., Csecs, I., Ngo, L., Rowin, E., Manning, W., Maron, M., & Nezafat, R. (2022). Machine learning phenotyping of scarred myocardium from cine in hypertrophic cardiomyopathy. European Heart Journal Cardiovascular Imaging, 23(4), 532–542. https://doi.org/10.1093/ehjci/jeab056

Matsoukas, S., Morey, J., Lock, G., Chada, D., Shigematsu, T., Marayati, N. F., Delman, B. N., Doshi, A., Majidi, S.,
De Leacy, R., Kellner, C. P., & Fifi, J. T. (2022). AI software detection of large vessel occlusion stroke on CT angiography: a real-world prospective diagnostic test accuracy study. Journal of Neurointerventional Surgery. https://doi.org/10.1136/ neurintsurg-2021-018391

McKinney, S. M., Sieniek, M., Godbole, V., Godwin, J., Antropova, N., Ashrafian, H., Back, T., Chesus, M., Corrado, G. S., Darzi, A., Etemadi, M., Garcia-Vicente, F., Gilbert, F. J., Halling-Brown, M., Hassabis, D., Jansen, S., Karthikesalingam, A., Kelly, C. J., King, D., … Shetty, S. (2020). International evaluation of an AI system for breast cancer screening. Nature, 577(7788), 89–94. https://doi.org/10.1038/s41586-019-1799-6

McLeavy, C. M., Chunara, M. H., Gravell, R. J., Rauf, A., Cushnie, A., Staley Talbot, C., & Hawkins, R. M. (2021). The future of CT: deep learning reconstruction. Clinical Radiology, 76(6), 407–415. https://doi.org/10.1016/j.crad.2021.01.010

Medical AI evaluation. (n.d.). Retrieved June 26, 2022, from https://ericwu09.github.io/medical-ai-evaluation/

Mlynarski, P., Delingette, H., Criminisi, A., & Ayache, N. (2019). Deep learning with mixed supervision for brain tumor segmentation. Journal of Medical Imaging (Bellingham, Wash.), 6(3), 034002. https://doi.org/10.1117/1.JMI.6.3.034002

Mongan, J., Moy, L., & Kahn, C. E., Jr. (2020). Checklist for Artificial Intelligence in Medical Imaging (CLAIM): A Guide for Authors and Reviewers. Radiology. Artificial Intelligence, 2(2), e200029. https://doi.org/10.1148/ryai.2020200029

Moon, H., Huo, Y., Abramson, R. G., Peters, R. A., Assad, A., Moyo, T. K., Savona, M. R., & Landman, B. A. (2019). Acceleration of spleen segmentation with end-to-end deep learning method and automated pipeline. Computers in Biology and Medicine, 107, 109–117. https://doi.org/10.1016/j.compbiomed.2019.01.018

Morey, J. R., Zhang, X., Yaeger, K. A., Fiano, E., Marayati, N. F., Kellner, C. P., De Leacy, R. A., Doshi, A., Tuhrim, S., & Fifi, J. T. (2021). Real-World Experience with Artificial Intelligence- Based Triage in Transferred Large Vessel Occlusion Stroke Patients. Cerebrovascular Diseases, 50(4), 450–455. https://doi. org/10.1159/000515320

Murdoch, B. (2021). Privacy and artificial intelligence: challenges for protecting health information in a new era. BMC Medical Ethics, 22(1), 122. https://doi.org/10.1186/s12910-021-00687-3

Murray, N. M., Unberath, M., Hager, G. D., & Hui, F. K. (2020). Artificial intelligence to diagnose ischemic stroke and identify large vessel occlusions: a systematic review. Journal of Neurointerventional Surgery, 12(2), 156–164. https://doi.org/10.1136/neurintsurg-2019-015135

Nagendran, M., Chen, Y., Lovejoy, C. A., Gordon, A. C., Komorowski, M., Harvey, H., Topol, E. J., Ioannidis, J. P. A., Collins, G. S., & Maruthappu, M. (2020). Artificial intelligence versus clinicians: systematic review of design, reporting standards, and claims of deep learning studies. BMJ, 368. https://doi.org/10.1136/bmj.m689

Nair, T., Precup, D., Arnold, D. L., & Arbel, T. (2020). Exploring uncertainty measures in deep networks for Multiple sclerosis lesion detection and segmentation. Medical Image Analysis, 59, 101557. https://doi.org/10.1016/j.media.2019.101557

Nakao, T., Hanaoka, S., Nomura, Y., Sato, I., Nemoto, M., Miki, S., Maeda, E., Yoshikawa, T., Hayashi, N., & Abe, O. (2018). Deep neural network-based computer-assisted detection of cerebral aneurysms in MR angiography. Journal of Magnetic Resonance Imaging: JMRI, 47(4), 948–953. https://doi.org/10.1002/jmri.25842

Nam, J. G., Kim, M., Park, J., Hwang, E. J., Lee, J. H., Hong, J. H., Goo, J. M., & Park, C. M. (2021). Development and validation of a deep learning algorithm detecting 10 common abnormalities on chest radiographs. The European Respiratory Journal: Official Journal of the European Society for Clinical Respiratory Physiology, 57(5). https://doi.org/10.1183/13993003.03061-2020

Narayana, P. A., Coronado, I., Sujit, S. J., Wolinsky, J. S., Lublin, F. D., & Gabr, R. E. (2020). Deep Learning for Predicting Enhancing Lesions in Multiple Sclerosis from Noncontrast. MRI. Radiology, 294(2), 398–404. https://doi.org/10.1148/ radiol.2019191061

National Institute for Health and Care Excellence (NICE). (n.d.). Evidence standards framework for digital health technologies. Retrieved June 10, 2022, from https://www.nice.org.uk/corporate/ecd7

Neisius, U., El-Rewaidy, H., Nakamori, S., Rodriguez, J., Manning, W. J., & Nezafat, R. (2019). Radiomic Analysis of Myocardial Native T1 Imaging Discriminates Between Hypertensive Heart Disease and Hypertrophic Cardiomyopathy. JACC. Cardiovascular Imaging, 12(10), 1946–1954. https://doi. org/10.1016/j.jcmg.2018.11.024

Nelson, A., Herron, D., Rees, G., & Nachev, P. (2019). Predicting scheduled hospital attendance with artificial intelligence. Npj Digital Medicine, 2(1), 26. https://doi.org/10.1038/s41746-019-0103-3

Nielsen, A., Hansen, M. B., Tietze, A., & Mouridsen, K. (2018). Prediction of Tissue Outcome and Assessment of Treatment Effect in Acute Ischemic Stroke Using Deep Learning. Stroke; a Journal of Cerebral Circulation, STROKEAHA.117.019740. https://doi.org/10.1161/STROKEAHA.117.019740

Norori, N., Hu, Q., Aellen, F. M., Faraci, F. D., & Tzovara, A. (2021). Addressing bias in big data and AI for health care: A call for open science. Patterns (New York, N.Y.), 2(10), 100347. https://doi.org/10.1016/j.patter.2021.100347

O’Connor, S. D., & Bhalla, M. (2021). Should Artificial Intelligence Tell Radiologists Which Study to Read Next? [Review of Should Artificial Intelligence Tell Radiologists Which Study to Read Next?]. Radiology. Artificial Intelligence, 3(2), e210009. https://doi.org/10.1148/ryai.2021210009

Office for Civil Rights (OCR). (2012, September 7). Guidance Regarding Methods for De-identification of Protected Health Information in Accordance with the Health Insurance Portability and Accountability Act (HIPAA) Privacy Rule. HHS.gov; US Department of Health and Human Services. https://www.hhs.gov/hipaa/for-professionals/privacy/special-topics/de- identification/index.html

Oktay, O., Schlemper, J., Le Folgoc, L., Lee, M., Heinrich, M., Misawa, K., Mori, K., McDonagh, S., Hammerla, N. Y.,
Kainz, B., Glocker, B., & Rueckert, D. (2018). Attention U-Net: Learning Where to Look for the Pancreas. In arXiv [cs.CV]. arXiv. http://arxiv.org/abs/1804.03999

Olczak, J., Fahlberg, N., Maki, A., Razavian, A. S., Jilert, A., Stark, A., Sköldenberg, O., & Gordon, M. (2017). Artificial intelligence for analyzing orthopedic trauma radiographs. Acta Orthopaedica, 88(6), 581–586. https://doi.org/10.1080/1745367
4.2017.1344459

Olthof, A. W., van Ooijen, P. M. A., & Rezazade Mehrizi, M. H. (2020). Promises of artificial intelligence in neuroradiology: a systematic technographic review. Neuroradiology, 62(10), 1265–1278. https://doi.org/10.1007/s00234-020-02424-w

Omoumi, P., Ducarouge, A., Tournier, A., Harvey, H., Kahn, C. E., Jr, Louvet-de Verchère, F., Pinto Dos Santos, D., Kober, T., & Richiardi, J. (2021). To buy or not to buy-evaluating commercial AI solutions in radiology (the ECLAIR guidelines). European Radiology, 31(6), 3786–3796. https://doi.org/10.1007/ s00330-020-07684-x

O’Neill, T. J., Xi, Y., Stehel, E., Browning, T., Ng, Y. S., Baker, C., & Peshock, R. M. (2021). Active Reprioritization of the Reading Worklist Using Artificial Intelligence Has a Beneficial Effect on the Turnaround Time for Interpretation of Head CT with Intracranial Hemorrhage. Radiology. Artificial Intelligence, 3(2), e200024. https://doi.org/10.1148/ryai.2020200024

Ooi, S. K. G., Makmur, A., Soon, A. Y. Q., Fook-Chong, S., Liew, C., Sia, S. Y., Ting, Y. H., & Lim, C. Y. (2021). Attitudes toward artificial intelligence in radiology with learner needs assessment within radiology residency programmes: a national multi-programme survey. Singapore Medical Journal, 62(3), 126–134. https://doi.org/10.11622/smedj.2019141

Pan, Y., Shi, D., Wang, H., Chen, T., Cui, D., Cheng, X., & Lu, Y. (2020). Automatic opportunistic osteoporosis screening using low-dose chest computed tomography scans obtained for lung cancer screening. European Radiology, 30(7), 4107–4116. https://doi.org/10.1007/s00330-020-06679-y

Park, H. J., Kim, S. M., La Yun, B., Jang, M., Kim, B., Jang, J. Y., Lee, J. Y., & Lee, S. H. (2019). A computer-aided diagnosis system using artificial intelligence for the diagnosis and characterization of breast masses on ultrasound: Added value for the inexperienced breast radiologist. Medicine, 98(3), e14146. https://doi.org/10.1097/MD.0000000000014146

Price, I. I., & Nicholson, W. (2019). Medical AI and Contextual Bias. https://papers.ssrn.com/abstract=3347890

Puvanasunthararajah, S., Fontanarosa, D., Wille, M.-L., & Camps, S. M. (2021). The application of metal artifact reduction methods on computed tomography scans for radiotherapy applications: A literature review. Journal of Applied Clinical Medical Physics / American College of Medical Physics, 22(6), 198–223. https://doi.org/10.1002/acm2.13255

Qin, Z. Z., Sander, M. S., Rai, B., Titahong, C. N., Sudrungrot, S., Laah, S. N., Adhikari, L. M., Carter, E. J., Puri, L., Codlin, A. J., & Creswell, J. (2019). Using artificial intelligence to read chest radiographs for tuberculosis detection: A multi-site evaluation of the diagnostic accuracy of three deep learning systems. Scientific Reports, 9(1), 15000. https://doi.org/10.1038/ s41598-019-51503-3

Ramspek, C. L., Jager, K. J., Dekker, F. W., Zoccali, C., & van Diepen, M. (2021). External validation of prognostic models: what, why, how, when and where?

Rao, B., Zohrabian, V., Cedeno, P., Saha, A., Pahade, J., & Davis, M. A. (2021). Utility of Artificial Intelligence Tool as a Prospective Radiology Peer Reviewer - Detection of Unreported Intracranial Hemorrhage. Academic Radiology, 28(1), 85–93. https://doi.org/10.1016/j.acra.2020.01.035

Reddy, S., Allan, S., Coghlan, S., & Cooper, P. (2020). A governance model for the application of AI in health care. Journal of the American Medical Informatics Association: JAMIA, 27(3), 491–497. https://doi.org/10.1093/jamia/ocz192

Reddy, S., Rogers, W., Makinen, V.-P., Coiera, E., Brown, P., Wenzel, M., Weicken, E., Ansari, S., Mathur, P., Casey, A., & Kelly, B. (2021). Evaluation framework to guide implementation of AI systems into healthcare settings. BMJ Health & Care Informatics, 28(1). https://doi.org/10.1136/ bmjhci-2021-100444

Reyes, M., Meier, R., Pereira, S., Silva, C. A., Dahlweid, F.-M., von Tengg-Kobligk, H., Summers, R. M., & Wiest, R. (2020). On the Interpretability of Artificial Intelligence in Radiology: Challenges and Opportunities. Radiology. Artificial Intelligence, 2(3), e190043. https://doi.org/10.1148/ryai.2020190043

Rezazade Mehrizi, M. H., van Ooijen, P., & Homan, M. (2021). Applications of artificial intelligence (AI) in diagnostic radiology: a technography study. European Radiology, 31(4), 1805–1811. https://doi.org/10.1007/s00330-020-07230-9

Richardson, J. P., Smith, C., Curtis, S., Watson, S., Zhu, X., Barry, B., & Sharp, R. R. (2021). Patient apprehensions about the use of artificial intelligence in healthcare. NPJ Digital Medicine, 4(1), 140. https://doi.org/10.1038/s41746-021-00509-1

Richardson, M. L., Garwood, E. R., Lee, Y., Li, M. D., Lo, H. S., Nagaraju, A., Nguyen, X. V., Probyn, L., Rajiah, P., Sin, J., Wasnik, A. P., & Xu, K. (2021). Noninterpretive Uses of Artificial Intelligence in Radiology. Academic Radiology, 28(9), 1225– 1235. https://doi.org/10.1016/j.acra.2020.01.012

Rockenbach, M. A. B. (2021, June 13). Multimodal AI in healthcare: Closing the gaps. CodeX. https://medium.com/codex/ multimodal-ai-in-healthcare-1f5152e83be2

Rodríguez-Ruiz, A., Krupinski, E., Mordang, J.-J., Schilling, K., Heywang-Köbrunner, S. H., Sechopoulos, I., & Mann, R. M. (2019). Detection of Breast Cancer with Mammography: Effect of an Artificial Intelligence Support System. Radiology, 290(2), 305–314. https://doi.org/10.1148/radiol.2018181371

Rodriguez-Ruiz, A., Lång, K., Gubern-Merida, A., Broeders, M., Gennaro, G., Clauser, P., Helbich, T. H., Chevalier, M., Tan, T., Mertelmeier, T., Wallis, M. G., Andersson, I., Zackrisson, S., Mann, R. M., & Sechopoulos, I. (2019). Stand- Alone Artificial Intelligence for Breast Cancer Detection in Mammography: Comparison With 101 Radiologists. Journal of the National Cancer Institute, 111(9), 916–922. https://doi.org/10.1093/jnci/djy222

Santomartino, S. M., & Yi, P. H. (2022). Systematic Review of Radiologist and Medical Student Attitudes on the Role and Impact of AI in Radiology. Academic Radiology. https://doi.org/10.1016/j.acra.2021.12.032

Schemmel, A., Lee, M., Hanley, T., Pooler, B. D., Kennedy, T., Field, A., Wiegmann, D., & Yu, J.-P. J. (2016). Radiology Workflow Disruptors: A Detailed Analysis. Journal of the American College of Radiology: JACR, 13(10), 1210–1214. https://doi.org/10.1016/j.jacr.2016.04.009

Schreiber-Zinaman, J., & Rosenkrantz, A. B. (2017). Frequency and reasons for extra sequences in clinical abdominal MRI examinations. Abdominal Radiology (New York), 42(1), 306–311. https://doi.org/10.1007/s00261-016-0877-6

Scott, I. A., Carter, S. M., & Coiera, E. (2021). Exploring stakeholder attitudes towards AI in clinical practice. BMJ Health & Care Informatics, 28(1). https://doi.org/10.1136/ bmjhci-2021-100450

Seah, J. C. Y., Tang, C. H. M., Buchlak, Q. D., Holt, X. G., Wardman, J. B., Aimoldin, A., Esmaili, N., Ahmad, H., Pham, H., Lambert, J. F., Hachey, B., Hogg, S. J. F., Johnston, B. P., Bennett, C., Oakden-Rayner, L., Brotchie, P., & Jones, C. M. (2021). Effect of a comprehensive deep-learning model on the accuracy of chest x-ray interpretation by radiologists: a retrospective, multireader multicase study. The Lancet. Digital Health, 3(8), e496–e506. https://doi.org/10.1016/S2589- 7500(21)00106-0

Sectra Amplifier Marketplace. (2021, July 5). Sectra Medical. https://medical.sectra.com/product/sectra-amplifier- marketplace/

Sermesant, M., Delingette, H., Cochet, H., Jaïs, P., & Ayache, N. (2021). Applications of artificial intelligence in cardiovascular imaging. Nature Reviews. Cardiology, 18(8), 600–609. https://doi.org/10.1038/s41569-021-00527-2

Setio, A. A. A., Traverso, A., de Bel, T., Berens, M. S. N., van den Bogaard, C., Cerello, P., Chen, H., Dou, Q., Fantacci, M. E., Geurts, B., Gugten, R. van der, Heng, P. A., Jansen, B., de Kaste, M. M. J., Kotov, V., Lin, J. Y.-H., Manders, J. T. M. C., Sóñora-Mengana, A., García-Naranjo, J. C., … Jacobs, C. (2017). Validation, comparison, and combination of algorithms for automatic detection of pulmonary nodules in computed tomography images: The LUNA16 challenge. Medical Image Analysis, 42, 1–13. https://doi.org/10.1016/j.media.2017.06.015

Seyyed-Kalantari, L., Zhang, H., McDermott, M. B. A., Chen, I. Y., & Ghassemi, M. (2021). Underdiagnosis bias of artificial intelligence algorithms applied to chest radiographs in under- served patient populations. Nature Medicine, 27(12), 2176– 2182. https://doi.org/10.1038/s41591-021-01595-0

Shan, H., Padole, A., Homayounieh, F., Kruger, U., Khera, R. D., Nitiwarangkul, C., Kalra, M. K., & Wang, G. (2019). Competitive performance of a modularized deep neural network compared to commercial algorithms for low-dose CT image reconstruction. Nature Machine Intelligence, 1(6), 269–276. https://doi.org/10.1038/s42256-019-0057-9

Sharma, K., Rupprecht, C., Caroli, A., Aparicio, M. C., Remuzzi, A., Baust, M., & Navab, N. (2017). Automatic Segmentation of Kidneys using Deep Learning for Total Kidney Volume Quantification in Autosomal Dominant Polycystic Kidney Disease. Scientific Reports, 7(1), 2049. https://doi.org/10.1038/s41598-017-01779-0

Shelmerdine, S. C., Arthurs, O. J., Denniston, A., & Sebire, N. J. (2021). Review of study reporting guidelines for clinical studies using artificial intelligence in healthcare. BMJ Health & Care Informatics, 28(1). https://doi.org/10.1136/ bmjhci-2021-100385

Shinagare, A. B., Ip, I. K., Abbett, S. K., Hanson, R., Seltzer, S. E., & Khorasani, R. (2014). Inpatient imaging utilization: trends of the past decade. AJR. American Journal of Roentgenology, 202(3), W277–W283. https://doi.org/10.2214/AJR.13.10986

Shlobin, N. A., Baig, A. A., Waqas, M., Patel, T. R., Dossani, R. H., Wilson, M., Cappuzzo, J. M., Siddiqui, A. H., Tutino, V. M., & Levy, E. I. (2022). Artificial Intelligence for Large-Vessel Occlusion Stroke: A Systematic Review. World Neurosurgery, 159, 207–220.e1. https://doi.org/10.1016/j.wneu.2021.12.004

Silberg, J., & Manyika, J. (2019, June 6). Tackling bias in artificial intelligence (and in humans). McKinsey & Company. https://www.mckinsey.com/featured-insights/artificial- intelligence/tackling-bias-in-artificial-intelligence-and-in- humans

Singh, S., Kalra, M. K., Hsieh, J., Licato, P. E., Do, S., Pien, H. H., & Blake, M. A. (2010). Abdominal CT: comparison of adaptive statistical iterative and filtered back projection reconstruction techniques. Radiology, 257(2), 373–383. https://doi.org/10.1148/radiol.10092212

Smith-Bindman, R., Kwan, M. L., Marlow, E. C., Theis, M. K., Bolch, W., Cheng, S. Y., Bowles, E. J. A., Duncan, J. R.,
Greenlee, R. T., Kushi, L. H., Pole, J. D., Rahm, A. K., Stout, N. K., Weinmann, S., & Miglioretti, D. L. (2019). Trends in Use of Medical Imaging in US Health Care Systems and in Ontario, Canada, 2000-2016. JAMA: The Journal of the American Medical Association, 322(9), 843–856. https://doi.org/10.1001/ jama.2019.11456

Sutherland, G., Russell, N., Gibbard, R., & Dobrescu, A. (n.d.). The value of radiology, part II. https://car.ca/wp-content/ uploads/2019/07/value-of-radiology-part-2-en.pdf

Tamada, D., Kromrey, M.-L., Ichikawa, S., Onishi, H., & Motosugi, U. (2020). Motion Artifact Reduction Using a Convolutional Neural Network for Dynamic Contrast Enhanced MR Imaging of the Liver. Magnetic Resonance in Medical Sciences: MRMS: An Official Journal of Japan Society of Magnetic Resonance in Medicine, 19(1), 64–76. https://doi.org/10.2463/ mrms.mp.2018-0156

The Medical Futurist. (n.d.). The Medical Futurist. Retrieved February 23, 2022, from https://medicalfuturist.com/fda- approved-ai-based-algorithms/

The Nuance AI Marketplace for Diagnostic Imaging. (n.d.). https://www.nuance.com/content/dam/nuance/en_us/collateral/ healthcare/data-sheet/ds-ai-marketplace-for-diagnostic- imaging-en-us.pdf

Thodberg, H. H., Kreiborg, S., Juul, A., & Pedersen, K. D. (2009). The BoneXpert method for automated determination of skeletal maturity. IEEE Transactions on Medical Imaging, 28(1), 52–66. https://doi.org/10.1109/TMI.2008.926067

Thomas, K. A., Kidziński, Ł., Halilaj, E., Fleming, S. L., Venkataraman, G. R., Oei, E. H. G., Gold, G. E., & Delp, S. L. (2020). Automated Classification of Radiographic Knee Osteoarthritis Severity Using Deep Neural Networks. Radiology. Artificial Intelligence, 2(2), e190065. https://doi.org/10.1148/ ryai.2020190065

Towards trustable machine learning. (2018). Nature Biomedical Engineering, 2(10), 709–710. https://doi.org/10.1038/ s41551-018-0315-x

Trinidad, M. G., Platt, J., & Kardia, S. L. R. (2020). The public’s comfort with sharing health data with third-party commercial companies. Humanities and Social Sciences Communications, 7(1), 1–10. https://doi.org/10.1057/s41599-020-00641-5

Trivedi, H., Mesterhazy, J., Laguna, B., Vu, T., & Sohn, J. H. (2018). Automatic Determination of the Need for Intravenous Contrast in Musculoskeletal MRI Examinations Using IBM Watson’s Natural Language Processing Algorithm. Journal of Digital Imaging, 31(2), 245–251. https://doi.org/10.1007/ s10278-017-0021-3

Tsao, D. N. (2020, July 27). AI in medical diagnostics 2020- 2030: Image recognition, players, clinical applications, forecasts: IDTechEx. https://www.idtechex.com/en/research-report/ai-in- medical-diagnostics-2020-2030-image-recognition-players- clinical-applications-forecasts/766

Tucci, V., Saary, J., & Doyle, T. E. (2022). Factors influencing trust in medical artificial intelligence for healthcare professionals: a narrative review. Journal of Medical Artificial Intelligence, 5, 4–4. https://doi.org/10.21037/jmai-21-25

Ueda, D., Yamamoto, A., Nishimori, M., Shimono, T., Doishita, S., Shimazaki, A., Katayama, Y., Fukumoto, S., Choppin, A., Shimahara, Y., & Miki, Y. (2019). Deep Learning for MR Angiography: Automated Detection of Cerebral Aneurysms. Radiology, 290(1), 187–194. https://doi.org/10.1148/ radiol.2018180901

Urakawa, T., Tanaka, Y., Goto, S., Matsuzawa, H., Watanabe, K., & Endo, N. (2019). Detecting intertrochanteric hip fractures with orthopedist-level accuracy using a deep convolutional neural network. Skeletal Radiology, 48(2), 239–244. https://doi.org/10.1007/s00256-018-3016-3

van Leeuwen, K. G., Schalekamp, S., Rutten, M. J. C. M., van Ginneken, B., & de Rooij, M. (2021). Artificial intelligence in radiology: 100 commercially available products and their scientific evidence. European Radiology, 31(6), 3797–3804. https://doi.org/10.1007/s00330-021-07892-z

Vayena, E., & Blasimme, A. (2017). Biomedical Big Data: New Models of Control Over Access, Use and Governance. Journal of Bioethical Inquiry, 14(4), 501–513. https://doi.org/10.1007/ s11673-017-9809-6

Vayena, E., Blasimme, A., & Cohen, I. G. (2018). Machine learning in medicine: Addressing ethical challenges. PLoS Medicine, 15(11), e1002689. https://doi.org/10.1371/journal. pmed.1002689

Wang, J., Yang, F., Liu, W., Sun, J., Han, Y., Li, D., Gkoutos, G. V., Zhu, Y., & Chen, Y. (2020). Radiomic Analysis of Native T1 Mapping Images Discriminates Between MYH7 and MYBPC3- Related Hypertrophic Cardiomyopathy. Journal of Magnetic Resonance Imaging: JMRI, 52(6), 1714–1721. https://doi.org/10.1002/jmri.27209

Wang, S.-H., Tang, C., Sun, J., Yang, J., Huang, C., Phillips, P., & Zhang, Y.-D. (2018). Multiple Sclerosis Identification by 14-Layer Convolutional Neural Network With Batch Normalization, Dropout, and Stochastic Pooling. Frontiers in Neuroscience, 12, 818. https://doi.org/10.3389/fnins.2018.00818

Watanabe, A. T., Lim, V., Vu, H. X., Chim, R., Weise, E., Liu, J., Bradley, W. G., & Comstock, C. E. (2019). Improved
Cancer Detection Using Artificial Intelligence: a Retrospective Evaluation of Missed Cancers on Mammography. Journal
of Digital Imaging, 32(4), 625–637. https://doi.org/10.1007/ s10278-019-00192-5

Weikert, T., Francone, M., Abbara, S., Baessler, B., Choi, B. W., Gutberlet, M., Hecht, E. M., Loewe, C., Mousseaux, E., Natale, L., Nikolaou, K., Ordovas, K. G., Peebles, C., Prieto, C., Salgado, R., Velthuis, B., Vliegenthart, R., Bremerich, J., & Leiner, T. (2021). Machine learning in cardiovascular radiology: ESCR position statement on design requirements, quality assessment, current applications, opportunities, and challenges. European Radiology, 31(6), 3909–3922. https://doi.org/10.1007/ s00330-020-07417-0

Whittlestone, J., Nyrup, R., Alexandrova, A., Dihal, K., & Cave, S. (2019). Ethical and societal implications of algorithms, data, and artificial intelligence: a roadmap for research.
London: Nuffield Foundation. https://www.nuffieldfoundation. org/sites/default/files/files/Ethical-and-Societal-Implications-of- Data-and-AI-report-Nuffield-Foundat.pdf

WHO operational handbook on tuberculosis Module 2: Screening – Systematic screening for tuberculosis disease. (n.d.). Retrieved June 19, 2022, from https://www.who.int/ publications-detail-redirect/9789240022614

Willemink, M. J., & Noël, P. B. (2019). The evolution of image reconstruction for CT-from filtered back projection to artificial intelligence. European Radiology, 29(5), 2185–2195. https://doi.org/10.1007/s00330-018-5810-7

Winder, M., Owczarek, A. J., Chudek, J., Pilch-Kowalczyk, J., & Baron, J. (2021). Are We Overdoing It? Changes in Diagnostic Imaging Workload during the Years 2010-2020 including the Impact of the SARS-CoV-2 Pandemic. Healthcare (Basel, Switzerland), 9(11). https://doi.org/10.3390/healthcare9111557

Wong, T. T., Kazam, J. K., & Rasiej, M. J. (2019). Effect of Analytics-Driven Worklists on Musculoskeletal MRI
Interpretation Times in an Academic Setting. AJR. American Journal of Roentgenology, 1–5. https://doi.org/10.2214/ AJR.18.20434

Wu, B., Zhou, Z., Wang, J., & Wang, Y. (2018). Joint learning for pulmonary nodule segmentation, attributes and malignancy prediction. 2018 IEEE 15th International Symposium on Biomedical Imaging (ISBI 2018), 1109–1113. https://doi.org/10.1109/ISBI.2018.8363765

Wu, E., Wu, K., Daneshjou, R., Ouyang, D., Ho, D. E., & Zou, J. (2021). How medical AI devices are evaluated: limitations and recommendations from an analysis of FDA approvals. Nature Medicine, 27(4), 582–584. https://doi.org/10.1038/s41591-021-01312-x

Wu, G.-G., Zhou, L.-Q., Xu, J.-W., Wang, J.-Y., Wei, Q., Deng, Y.-B., Cui, X.-W., & Dietrich, C. F. (2019). Artificial intelligence in breast ultrasound. World Journal of Radiology, 11(2), 19–26. https://doi.org/10.4329/wjr.v11.i2.19

Yala, A., Schuster, T., Miles, R., Barzilay, R., & Lehman, C. (2019). A Deep Learning Model to Triage Screening Mammograms: A Simulation Study. Radiology, 293(1), 38–46. https://doi.org/10.1148/radiol.2019182908

Yasaka, K., Akai, H., Kunimatsu, A., Abe, O., & Kiryu, S. (2018). Liver Fibrosis: Deep Convolutional Neural Network for Staging by Using Gadoxetic Acid-enhanced Hepatobiliary Phase MR Images. Radiology, 287(1), 146–155. https://doi.org/10.1148/ radiol.2017171928

Yeung, K. (2018). A Study of the Implications of Advanced Digital Technologies (Including AI Systems) for the Concept of
Responsibility Within a Human Rights Framework. https://papers. ssrn.com/abstract=3286027

Yoo, Y., Tang, L. Y. W., Li, D. K. B., Metz, L., Kolind, S., Traboulsee, A. L., & Tam, R. C. (2019). Deep learning of brain lesion patterns and user-defined clinical and MRI features for predicting conversion to multiple sclerosis from clinically isolated syndrome. Computer Methods in Biomechanics and Biomedical Engineering: Imaging & Visualization, 7(3), 250–259. https://doi.org/10.1080/21681163.2017.1356750

Yu, A. C., Mohajer, B., & Eng, J. (2022). External Validation of Deep Learning Algorithms for Radiologic Diagnosis: A Systematic Review. Radiology. Artificial Intelligence, 4(3), e210064. https://doi.org/10.1148/ryai.210064

Yu, J.-P. J., Kansagra, A. P., & Mongan, J. (2014). The radiologist’s workflow environment: evaluation of disruptors and potential implications. Journal of the American College of Radiology: JACR, 11(6), 589–593. https://doi.org/10.1016/j. jacr.2013.12.026

Yusuf, M., Atal, I., Li, J., Smith, P., Ravaud, P., Fergie, M., Callaghan, M., & Selfe, J. (2020). Reporting quality of studies using machine learning models for medical diagnosis: a systematic review. BMJ Open, 10(3), e034568. https://doi.org/10.1136/bmjopen-2019-034568

Yu, Y., Xie, Y., Thamm, T., Gong, E., Ouyang, J., Christensen, S., Marks, M. P., Lansberg, M. G., Albers, G. W., & Zaharchuk,
G. (2021). Tissue at Risk and Ischemic Core Estimation Using Deep Learning in Acute Stroke. AJNR. American Journal of Neuroradiology, 42(6), 1030–1037. https://doi.org/10.3174/ajnr.A7081

Yu, Y., Xie, Y., Thamm, T., Gong, E., Ouyang, J., Huang, C., Christensen, S., Marks, M. P., Lansberg, M. G., Albers, G. W., & Zaharchuk, G. (2020). Use of Deep Learning to Predict Final Ischemic Stroke Lesions From Initial Magnetic Resonance Imaging. JAMA Network Open, 3(3), e200772. https://doi.org/10.1001/jamanetworkopen.2020.0772

Zhang, Y., & Yu, H. (2018). Convolutional Neural Network Based Metal Artifact Reduction in X-Ray Computed Tomography. IEEE Transactions on Medical Imaging, 37(6), 1370–1381. https://doi.org/10.1109/TMI.2018.2823083

Zhao, B., Liu, Z., Ding, S., Liu, G., Cao, C., & Wu, H. (2022). Motion artifact correction for MR images based on convolutional neural network. Optoelectronics Letters, 18(1), 54–58. https://doi.org/10.1007/s11801-022-1084-z

Zhao, J., Huang, Y., Song, Y., Xie, D., Hu, M., Qiu, H., & Chu, J. (2020). Diagnostic accuracy and potential covariates for machine learning to identify IDH mutations in glioma patients: evidence from a meta-analysis. European Radiology, 30(8), 4664–4674. https://doi.org/10.1007/s00330-020-06717-9

Zhou, C., Ding, C., Wang, X., Lu, Z., & Tao, D. (2020). One-pass Multi-task Networks with Cross-task Guided Attention for Brain Tumor Segmentation. IEEE Transactions on Image Processing: A Publication of the IEEE Signal Processing Society. https://doi.org/10.1109/TIP.2020.2973510

Zicari, R. V., Brodersen, J., Brusseau, J., Düdder, B., Eichhorn, T., Ivanov, T., Kararigas, G., Kringen, P., McCullough, M., Möslein, F., Mushtaq, N., Roig, G., Stürtz, N., Tolle, K., Tithi, J. J., van Halem, I., & Westerlund, M. (2021). Z-Inspection®: A Process to Assess Trustworthy AI. IEEE Transactions on Technology and Society, 1–1. https://doi.org/10.1109/TTS.2021.3066209
 

 

Guide to Artificial Intelligence in Radiology

    Artificial intelligence (AI) is playing a growing role in all our lives and has shown promise in addressing some of the greatest current and upcoming societal challenges we face. The healthcare industry, though notoriously complex and resistant to disruption, potentially has a lot to gain from the use of AI. With an established history of leading digital transformation in healthcare and an urgent need for improved efficiency, radiology has been at the forefront of harnessing AI’s potential.

    This book covers how and why AI can address challenges faced by radiology departments, provides an overview of the fundamental concepts related to AI, and describes some of the most promising use cases for AI in radiology. In addition, the major challenges associated with the adoption of AI into routine radiological practice are discussed. The book also covers some crucial points radiology departments should keep in mind when deciding on which AI-based solutions to purchase. Finally, it provides an outlook on what new and evolving aspects of AI in radiology to expect in the near future.

    The healthcare industry has experienced a number of trends over the past few decades that demand a change in the way certain things are done. These trends are particularly salient in radiology, where the diagnostic quality of imaging scans has improved dramatically while scan times have decreased. As a result, the amount and complexity of medical imaging data acquired have increased substantially over the past few decades (Smith-Bindman et al., 2019; Winder et al., 2021) and are expected to continue to increase (Tsao, 2020). This issue is complicated by a widespread global shortage of radiologists (AAMC Report Reinforces Mounting Physician Shortage, 2021, Clinical Radiology UK Workforce Census 2019 Report, 2019). Healthcare workers, including radiologists, have an increasing workload (Bruls & Kwee, 2020; Levin et al., 2017) that contributes to burnout and medical errors (Harry et al., 2021). Being an essential service provider to virtually all other hospital departments, staff shortages within radiology have significant effects that spread throughout the hospital and to society as a whole (England & Improvement, 2019; Sutherland et al., n.d.).

    With an ageing global population and a rising burden of chronic illnesses, these issues are expected to pose even more of a challenge to the healthcare industry in the future.

    AI-based medical imaging solutions have the potential to ameliorate these challenges for several reasons. They are particularly suited to handling large, complex datasets (Alzubaidi et al., 2021). Moreover, they are well suited to automate some of the tasks traditionally performed by radiologists and radiographers, potentially freeing up time and making workflows within radiology departments more efficient (Allen et al., 2021; Baltruschat et al., 2021; Kalra et al., 2020; O’Neill et al., 2021; van Leeuwen et al., 2021; Wong et al., 2019). AI is also capable of detecting complex patterns in data that humans cannot necessarily find or quantify (Dance, 2021; Korteling et al., 2021; Kühl et al., 2020).

    The term “artificial intelligence” refers to the use of computer systems to solve specific problems in a way that simulates human reasoning. One fundamental characteristic of AI is that, like humans, these systems can tailor their solutions to changing circumstances. Note that, while these systems are meant to mimic on a fundamental level how humans think, their capacity to do so (e.g. in terms of the amount of data they can handle at one time, the nature and amount of patterns they can find in the data, and the speed at which they do so) often exceeds that of humans.

    AI solutions come in the form of computer algorithms, which are pieces of computer code representing instructions to be followed to solve a specific problem. In its most fundamental form, the algorithm takes data as an input, performs some computation on that data, and returns an output.

    An AI algorithm can be explicitly programmed to solve a specific task, analogous to a step-by-step recipe for baking a cake. On the other hand, the algorithm can be programmed to look for patterns within the data in order to solve the problem. These types of algorithms are known as machine learning algorithms. Thus, all machine learning algorithms are AI, but not all AI is machine learning. The patterns in the data that the algorithm can be explicitly programmed to look for or that it can “discover” by itself are known as features. An important characteristic of machine learning is that such algorithms learn from the data itself, and their performance improves the more data they are given.

    One of the most common uses of machine learning is in classification - assigning a piece of data a particular label. For example, a machine learning algorithm might be used to tell if a photo (the input) shows a dog or a cat (the label). The algorithm can learn to do so in a supervised or unsupervised way.

    Supervised learning

    In supervised learning, the machine learning algorithm is given data that has been labelled with the ground truth, in this example, photos of dogs and cats that have been labelled as such. The process then goes through the following phases:

    1.Training phase: The algorithm learns the features associated with dogs and cats using the aforementioned data (training data).
    2.Test phase: The algorithm is then given a new set of photos (the test data), it labels them and the performance of the algorithm on that data is assessed.

    In some cases, there is a phase in between training and test, known as the validation phase. In this phase, the algorithm is given a new set of photos (not included in either the training or test data), its performance is assessed on this data, and the model is tweaked and retrained on the training data. This is repeated until some predefined performance-based criterion is reached, and the algorithm then enters the test phase.

    Unsupervised learning

    In unsupervised learning, the algorithm identifies features within the input data that allow it to assign classes to the individual data points without being told explicitly what those classes are or should be. Such algorithms can identify patterns or group data points together without human intervention and include clustering and dimensionality reduction algorithms. Not all machine learning algorithms perform classification. Some are used to predict a continuous metric (e.g. the temperature in four weeks’ time) instead of a discrete label (e.g. cats vs dogs). These are known as regression algorithms.

    Neural networks and deep learning

    A neural network is made up of an input layer and an output layer, which are themselves composed of nodes. In simple neural networks, features that are manually derived from a dataset are fed into the input layer, which performs some computations, the results of which are relayed to the output layer. In deep learning, multiple “hidden” layers exist between the input and output layers. Each node of the hidden layers performs calculations using certain weights and relays the output to the next hidden layer until the output layer is reached.

    In the beginning, random values are assigned to the weights and the accuracy of the algorithm is calculated. The values of the weights are then iteratively adjusted until a set of weight values that maximize accuracy is found. This iterative adjustment of the weight values is usually done by moving backwards from the output layer to the input layer, a technique called backpropagation. This entire process is done on the training data.

    Performance evaluation

    Understanding how the performance of AI algorithms is assessed is key to interpreting the AI literature. Several performance metrics exist for assessing how well a model performs certain tasks. No single metric is perfect, so a combination of several metrics provides a fuller picture of model performance.

    In regression, the most commonly used metrics include:

    • Mean absolute error (MAE): the average difference between the predicted values and the ground truth.
    • Root mean square error (RMSE): the differences between the predicted values and the ground truth are squared and then averaged over the sample. Then the square root of the average is taken. Unlike the MAE, the RMSE thus gives higher weight to larger differences.
    • R2: the proportion of the total variance in the ground truth explained by the variance in the predicted values. It ranges from 0 to 1.

    The following metrics are commonly used in classification tasks:

    • Accuracy: this is the proportion of all predictions that were predicted correctly. It ranges from 0 to 1.
    • Sensitivity: also known as the true positive rate (TPR) or recall, this is the proportion of true positives that were predicted correctly. It ranges from 0 to 1.
    • Specificity: Also known as the true negative rate (TNR), this is the proportion of true negatives that were predicted correctly. It ranges from 0 to 1.
    • Precision: also known as positive predictive value (PPV), this is the proportion of positive classifications that were predicted correctly. It ranges from 0 to 1.

    An inherent trade-off exists between sensitivity and specificity. The relevant importance of each, as well as their interpretation, highly depends on the specific research question and classification task.

    Importantly, although classification models are meant to reach a binary conclusion, they are inherently probability-based. This means that these models will output a probability that a data point belongs to one class or another. In order to reach a conclusion on the most likely class, a threshold is used. Metrics such as accuracy, sensitivity, specificity and precision refer to the performance of the algorithm based on a certain threshold. The area under the receiver operating characteristic curve (AUC) is a threshold-independent performance metric. The AUC can be interpreted as the probability that a random positive example is ranked higher by the algorithm than a random negative example.

    In image segmentation tasks, which are a type of classification task, the following metrics are commonly used:

    • Dice similarity coefficient: a measure of overlap between two sets (e.g. two images) that is calculated as two times the number of elements common to the sets divided by the sum of the number of elements in each set. It ranges from 0 (no overlap) to 1 (perfect overlap).
    • Hausdorff distance: a measure of how far two sets (e.g. two images) within a space are far from each other. It is basically the largest distance from one point in one set to the closest point in the other set.

    Internal and external validity

    Internally valid models perform well in their task on the data being used to train and validate them. The degree to which they are internally valid is assessed using the performance metrics outlined above and depends on the characteristics of the model itself and the quality of the data that the model was trained and validated on.

    Externally valid models perform well in their tasks on new data (Ramspek et al., 2021). The better the model performs on data that differs from the data the models were trained and validated on, the higher the external validity. In practice, this often requires the performance of the models to be tested on data from hospitals or geographical areas that were not part of the model’s training and validation datasets.

    Guidelines for evaluating AI research

    Several guidelines have been developed to assess the evidence behind AI-based interventions in healthcare (X. Liu et al., 2020; Mongan et al., 2020; Shelmerdine et al., 2021; Weikert et al., 2021). These provide a template for those doing AI research in healthcare and ensure that relevant information is reported transparently and comprehensively, but can also be used by other stakeholders to assess the quality of published research. This helps ensure that AI-based solutions with substantial potential or actual limitations, particularly those caused by poor reporting (Bozkurt et al., 2020; D. W. Kim et al., 2019; X. Liu et al., 2019; Nagendran et al., 2020; Yusuf et al., 2020), are not prematurely adopted (CONSORT-AI and SPIRIT-AI Steering Group, 2019). Guidelines have also been proposed for evaluating the trustworthiness of AI-based solutions in terms of transparency, confidentiality, security, and accountability (Buruk et al., 2020; Lekadir et al., 2021; Zicari et al., 2021).

    Over the past few years, AI has shown great potential in addressing a broad range of tasks within a medical imaging department, including many that happen before the patient is scanned. Implementations of AI to improve the efficiency of radiology workflows prior to patient scanning are sometimes referred to as “upstream AI” (Kapoor et al., 2020; M. L. Richardson et al., 2021).

    Scheduling

    One promising upstream AI application is predicting whichpatients arelikelytomisstheirscanappointments. Missed appointments are associated with significantly increased workload and costs (Dantas et al., 2018). Using a Gradient Boosting approach, Nelson et al. predicted missed hospital magnetic resonance imaging (MRI) appointments in the United Kingdom’s National Health Service (NHS) with high accuracy (Nelson et al., 2019). Their simulations also suggested that acting on the predictions of this model by targeting patients who are likely to miss their appointments would potentially yield a net benefit of several pounds per appointment across a range of model thresholds and missed appointment rates (Nelson et al., 2019). Similar results were recently found in a study of a single hospital in Singapore. For the 6-month period following the deployment of the predictive tool they were able to significantly reduce the no show rate from 19.3 % tp 15.9 % which translated into a potential economic benefit of $180,000 (Chong et. al., 2020).

    Scheduling scans in a radiology department is a challenging endeavour because, although it is largely an administrative task, it depends heavily on medical information. The task of assigning patients to specific appointments thus often requires the input of someone with domain knowledge, which stipulates that either the person making the appointments must be a radiologist or radiology technician, or these people will have to provide input regularly. In either scenario, the process is somewhat inefficient and can potentially be streamlined using AI-based algorithms that check scan indications and contraindications and provide the people scheduling the scans with information about scan urgency (Letourneau-Guillon et al., 2020).

    Protocolling

    Depending on hospital or clinic policy, the decision on what exact scan protocol a patient receives is usually made based on the information on the referring physician’s scan request and the judgement of the radiologist. This is often supplemented by direct communication between the referring physician and radiologist and the radiologist’s review of the patient’s medical information. This process improves patient care (Boland et al., 2014) but can be time-consuming and inefficient, particularly with modalities like MRI, where a large number of protocol permutations exist. In one study, protocolling alone accounted for about 6 % of the radiologist’s working time (Schemmel et al., 2016). Radiologists are also often interrupted by tasks such as protocolling when interpreting images, despite the fact that the latter is considered a radiologist’s primary responsibility (Balint et al., 2014; J.-P. J. Yu et al., 2014).

    Interpretation of the narrative text of the referring physician’s scan request has been attempted using natural language classifiers, the same technology used in chatbots and virtual assistants. Natural language classifiers based on deep learning have shown promise in assigning patients to either a contrast-enhanced or non-enhanced MRI protocol for musculoskeletal MRI, with an accuracy of 83 % (Trivedi et al., 2018) and 94 % (Y. H. Lee, 2018). Similar algorithms have shown an accuracy of 95 % for predicting the appropriate brain MRI protocol using a combination of up to 41 different MRI sequences (Brown & Marotta, 2018). Across a wide range of body regions, a deep-learning-based natural language classifier decided based on the narrative text of the scan requests whether to automatically assign a specific computed tomography (CT) or MRI protocol (which it did with 95 % accuracy) or, in more difficult cases, recommend a list of three most appropriate protocols to the radiologist (which it did with 92 % accuracy) (Kalra et al., 2020).

    AI has also been used to decide whether already protocolled scans need to be extended, a decision which has to be made in real-time while the patient is inside the scanner. One such example is in prostate MRI, where a decision on whether to administer a contrast agent is often made after the non-contrast sequences. Hötker et al. found that a convolutional neural network (CNN) assigned 78 % of patients to the appropriate prostate MRI protocol (Hötker et al., 2021). The sensitivity of the CNN for the need for contrast was 94.4 % with a specificity of 68.8 % and only 2 % of patients in their study would have had to be called back for a contrast- enhanced scan (Hötker et al., 2021).

    Image quality improvement and monitoring

    Many AI-based solutions that work in the background of radiology workflows to improve image quality have recently been established. These include solutions for monitoring image quality, reducing image artefacts, improving spatial resolution, and speeding up scans.

    Such solutions are entering the radiology mainstream, particularly for computed tomography, which for decades used established but artefact-prone methods for reconstructing interpretable images from the raw sensor data (Deák et al., 2013; Singh et al., 2010).

    These are gradually being replaced by deep-learning- based reconstruction methods, which improve image quality while maintaining low radiation doses (Akagi et al., 2019; H. Chen et al., 2017; Choe et al., 2019; Shan et al., 2019). This reconstruction is performed on supercomputers on the CT scanner itself or on the cloud. The balance between radiation dose and image quality can be adjusted on a protocol-specific basis to tailor scans to individual patients and clinical scenarios (McLeavy et al., 2021; Willemink & Noël, 2019). Such approaches have found particular use when scanning children, pregnant women, and obese patients as well as CT scans of the urinary tract and heart (McLeavy et al., 2021).

    AI-based solutions have also been used to speed up scans while maintaining diagnostic quality. Scan time reduction not only improves overall efficiency but also contributes to an overall better patient experience and compliance with imaging examination. A multi- centre study of spine MRI showed that a deep-learning- based image reconstruction algorithm that enhanced images using filtering and detail-preserving noise reduction reduced scan times by 40 % (Bash, Johnson, et al., 2021). For T1-weighted MRI scans of the brain, a similar algorithm that improves image sharpness and reduces image noise reduced scan times by 60 % while maintaining the accuracy of brain region volumetry compared to standard scans (Bash, Wang, et al., 2021).

    In routine radiological practice, images often contain artefacts that reduce their interpretability. These artefacts are the result of characteristics of the specific imaging modality or protocol used or factors intrinsic to the patient being scanned, such as the presence of foreign bodies or the patient moving during the scan. Particularly with MRI, imaging protocols that demand fast scanning often introduce certain artefacts to the reconstructed image. In one study, a deep-learning- based algorithm reduced banding artefacts associated with balanced steady-state free precession MRI sequences of the brain and knee (K. H. Kim & Park, 2017). For real-time imaging of the heart using MRI, another study found that the aliasing artefacts introduced by the data undersampling were reduced by using a deep-learning-based approach (Hauptmann et al., 2019). The presence of metallic foreign bodies such as dental, orthopaedic or vascular implants is a common patient-related factor causing image artefacts in both CT and MRI (Boas & Fleischmann, 2012; Hargreaves et al., 2011). Although not yet well established, several deep-learning-based approaches for reducing these artefacts have been investigated (Ghani & Clem Karl, 2019; Puvanasunthararajah et al., 2021; Zhang & Yu, 2018). Similar approaches are being tested for reducing motion-related artefacts in MRI (Tamada et al., 2020; B. Zhao et al., 2022).

    AI-based solutions for monitoring image quality potentially reduce the need to call patients back to repeat imaging examinations, which is a common problem (Schreiber-Zinaman & Rosenkrantz, 2017). A deep-learning-based algorithm that identifies the radiographic view acquired and extracts quality-related metrics from ankle radiographs was able to predict image quality with about 94 % accuracy (Mairhöfer et al., 2021). Another deep-learning-based approach was capable of predicting nondiagnostic liver MRI scans with a negative predictive value of between 86 % and 94 % (Esses et al., 2018). This real-time automated quality control potentially allows radiology technicians to rerun scans or run additional scans with greater diagnostic value.

    Scan reading prioritization

    With staff shortages and increasing scan numbers, radiologists face long reading lists. To optimize efficiency and patient care, AI-based solutions have been suggested as a way to prioritize which scans radiologists read and report first, usually by screening acquired images for findings that require urgent intervention (O’Connor & Bhalla, 2021). This has been most extensively studied in neuroradiology, where moving CT scans that were found to have intracranial haemorrhage by an AI-based tool to the top of the reading list reduced the time it took radiologists to view the scans by several minutes (O’Neill et al., 2021). Another study found that the time-to diagnosis (which includes the time from image acquisition to viewing by the radiologist and the time to read and report the scans) was reduced from 512 to 19 minutes in an outpatient setting when such a worklist prioritization was used (Arbabshirani et al., 2018). A simulation study using AI-based worklist prioritization based on identifying urgent findings on chest radiographs (such as pneumothorax, pleural effusions, and foreign bodies) also found a substantial reduction in the time it took to view and report the scans compared to standard workflow prioritization (Baltruschat et al., 2021).

    Image interpretation

    Currently, the majority of commercially available AI- based solutions in medical imaging focus on some aspect of analyzing and interpreting images (Rezazade Mehrizi et al., 2021; van Leeuwen et al., 2021). This includes segmenting parts of the image (for surgical or radiation therapy targeting, for example), bringing suspicious areas to radiologists’ attention, extracting imaging biomarkers (radiomics), comparing images across time, and reaching specific imaging diagnoses.

    Neurology

    ¡ 29–38 % of commercially available AI-based applications in radiology (Rezazade Mehrizi et al., 2021; van Leeuwen et al., 2021).

    Most commercially available AI-based solutions targeted at neuroimaging data aim to detect and characterize ischemic stroke, intracranial haemorrhage, dementia, and multiple sclerosis (Olthof et al., 2020). Several studies have shown excellent accuracy of AI- based methods for the detection and classification of intraparenchymal, subarachnoid, and subdural haemorrhage on head CT (Flanders et al., 2020; Ker et al., 2019; Kuo et al., 2019). Subsequent studies showed that, compared to radiologists, some AI-based solutions have substantially lower false positive and negative rates (Ginat, 2020; Rao et al., 2021). In ischemic stroke, AI-based solutions have largely focused on the quantification of the infarct core (Goebel et al., 2018; Maegerlein et al., 2019), the detection of large vessel occlusion (Matsoukas et al., 2022; Morey et al., 2021; Murray et al., 2020; Shlobin et al., 2022), and the prediction of stroke outcomes (Bacchi et al., 2020; Nielsen et al., 2018; Y. Yu et al., 2020, 2021).

    In multiple sclerosis, AI has been used to identify and segment lesions (Nair et al., 2020; S.-H. Wang et al., 2018), which can be particularly helpful for the longitudinal follow-up of patients. It has also been used to extract imaging features associated with progressive disease and conversion from clinically isolated syndrome to definite multiple sclerosis (Narayana et al., 2020; Yoo et al., 2019). Other applications of AI in neuroradiology include the detection of intracranial aneurysms (Faron et al., 2020; Nakao et al., 2018; Ueda et al., 2019) and the segmentation of brain tumours (Kao et al., 2019; Mlynarski et al., 2019; Zhou et al., 2020) as well as the prediction of brain tumour genetic markers from imaging data (Choi et al., 2019; J. Zhao et al., 2020)

    Chest

    ¡ 24 %–31 % of commercially available AI-based applications in radiology (Rezazade Mehrizi et al., 2021; van Leeuwen et al., 2021).

    When interpreting chest radiographs, radiologists detected substantially more critical and urgent findings when aided by a deep-learning-based algorithm, and did so much faster than without the algorithm (Nam et al., 2021). Deep-learning-based image interpretation algorithms have also been found to improve radiology residents’ sensitivity for detecting urgent findings on chest radiographs from 66 % to 73 % (E. J. Hwang, Nam, et al., 2019). Another study which focused on a broader range of findings on chest radiographs also found that radiologists aided by a deep-learning-based algorithm had higher diagnostic accuracy than radiologists who read the radiographs without assistance (Seah et al., 2021). The uses of AI in chest radiology also extend to cross-sectional imaging like CT. A deep learning algorithm was found to detect pulmonary embolism on CT scans with high accuracy (AUC = 0.85) (Huang, Kothari, et al., 2020). Moreover, a deep learning algorithm was 90 % accurate in detecting aortic dissection on non-contrast-enhanced CT scans, similar to the performance of radiologists (Hata et al., 2021).

    Outside the emergency setting, AI-based solutions have been widely tested and implemented for tuberculosis screening on chest radiographs (E. J. Hwang, Park, et al., 2019; S. Hwang et al., 2016; Khan et al., 2020; Qin et al., 2019; WHO Operational Handbook on Tuberculosis Module 2: Screening – Systematic Screening for Tuberculosis Disease, n.d.). In addition, they have been useful for lung cancer screening both in terms of detecting lung nodules on CT (Setio et al., 2017) and chest radiographs (Li et al., 2020) and by classifying whether nodules are likely to be malignant or benign (Ardila et al., 2019; Bonavita et al., 2020; Ciompi et al., 2017; B. Wu et al., 2018). AI-based solutions also show great promise for the diagnosis of pneumonia, chronic obstructive pulmonary disease, and interstitial lung disease (F. Liu et al., 2021).

    Breast

    ¡ 11 % of commercially available AI-based applications in radiology (Rezazade Mehrizi et al., 2021; van Leeuwen et al., 2021).

    So far, many of the AI-based algorithms targeting breast imaging aim to reduce the workload of radiologists reading mammograms. Ways to do this have included using AI-based algorithms to triage out negative mammograms, which in one study was associated with a reduction in radiologists’ workload by almost one-fifth (Yala et al., 2019). Other studies that have replaced second readers of mammograms with AI- based algorithms have shown that this leads to fewer false positives and false negatives as well as reduces the workload of the second reader by 88 % (McKinney et al., 2020).

    AI-based solutions for mammography have also been found to increase the diagnostic accuracy of radiologists (McKinney et al., 2020; Rodríguez-Ruiz et al., 2019; Watanabe et al., 2019) and some have been found to be highly accurate in independently detecting and classifying breast lesions (Agnes et al., 2019; Al- Antari et al., 2020; Rodriguez-Ruiz et al., 2019).
    Despite this, a recent systematic review of 36 AI- based algorithms found that these studies were of poor methodological quality and that all algorithms were less accurate than the consensus of two or more radiologists (Freeman et al., 2021). AI-based algorithms have nonetheless shown potential for extracting cancer-predictive features from mammograms beyond mammographic breast density (Arefan et al., 2020; Dembrower et al., 2020; Hinton et al., 2019). Beyond mammography, AI-based solutions have been developed for detecting and classifying breast lesions on ultrasound (Akkus et al., 2019; Park et al., 2019; G.- G. Wu et al., 2019) and MRI (Herent et al., 2019).

    Cardiac

    ¡ 11 % of commercially available AI-based applications in radiology (Rezazade Mehrizi et al., 2021; van Leeuwen et al., 2021).

    Cardiac radiology has always been particularly challenging because of the difficulties inherent in acquiring images of a constantly moving organ. Because of this, it has benefited immensely from advances in imaging technology and seems set to benefit greatly from AI as well (Sermesant et al., 2021). Most of the AI-based applications of the cardiovascular system use MRI, CT or ultrasound data (Weikert et al., 2021). Prominent examples include the automated calculation of ejection fraction on echocardiography, quantification of coronary artery calcification on cardiac CT, determination of right ventricular volume on CT pulmonary angiography, and determination of heart chamber size and thickness on cardiac MRI (Medical AI Evaluation, n.d., The Medical Futurist, n.d.). AI-based solutions for the prediction of patients likely to respond favourably to cardiac interventions, such as cardiac resynchronization therapy, based on imaging and clinical parameters have also shown great promise (Cikes et al., 2019; Hu et al., 2019). Changes in cardiac MRI not readily visible to human readers but potentially useful for differentiating different types of cardiomyopathies can also be detected using AI through texture analysis (Neisius et al., 2019; J. Wang et al., 2020) and other radiomic approaches (Mancio et al., 2022).

    Musculoskeletal

    ¡ 7–11 % of commercially available AI-based applications in radiology (Rezazade Mehrizi et al., 2021; van Leeuwen et al., 2021).

    Promising applications of AI in the assessment of muscles, bones and joints include applications where human readers generally show poor between- and within-rater reliability, such as the determination of skeletal age based on bone radiographs (Halabi et al., 2019; Thodberg et al., 2009) and screening for osteoporosis on radiographs (Kathirvelu et al., 2019; J.-S. Lee et al., 2019) and CT (Pan et al., 2020). AI- based solutions have also shown promise for detecting fractures on radiographs and CT (Lindsey et al., 2018; Olczak et al., 2017; Urakawa et al., 2019). One systematic review of AI-based solutions for fracture detection in several different body parts showed AUCs ranging from 0.94 to 1.00 and accuracies of 77 % to 98 % (Langerhuizen et al., 2019). AI-based solutions have also achieved accuracies similar to radiologists for classification of the severity of degenerative changes of the spine (Jamaludin et al., 2017) and extremity joints (F. Liu et al., 2018; Thomas et al., 2020). AI-based solutions have also been developed to determine the origin of skeletal metastases (Lang et al., 2019) and the classification of primary bone tumours (Do et al., 2017).

    Abdomen and pelvis

    ¡ 4 % of commercially available AI-based applications in radiology (Rezazade Mehrizi et al., 2021; van Leeuwen et al., 2021).

    Much of the efforts in using AI in abdominal imaging have thus far concentrated on the automated segmentation of organs such as the liver (Dou et al., 2017), spleen (Moon et al., 2019), pancreas (Oktay et al., 2018), and kidneys (Sharma et al., 2017). In addition, a systematic review of 11 studies using deep learning for the detection of malignant liver masses showed accuracies of up to 97 % and AUCs of up to 0.92 (Azer, 2019).

    Other applications of AI in abdominal radiology include the detection of liver fibrosis (He et al., 2019; Yasaka et al., 2018), fatty liver disease, hepatic iron content, the detection of free abdominal gas on CT, and automated volumetry and segmentation of the prostate (AI for Radiology, n.d.).

    Despite the great potential of AI in medical imaging, it has yet to find widespread implementation and impact in routine clinical practice. This research-to- clinic translation is being hindered by several complex and interrelated issues that directly or indirectly lower the likelihood of AI-based solutions being adopted. One major way they do so is by creating a lack of trust in AI- based solutions by key stakeholders such as regulators, healthcare professionals and patients (Cadario et al., 2021; Esmaeilzadeh, 2020; J. P. Richardson et al., 2021; Tucci et al., 2022).

    Generalizability

    One major challenge is to develop AI-based solutions that continue to perform well in new, real-world scenarios. In a large systematic review, almost half of the studied AI-based medical imaging algorithms reported a greater than 0.05 decrease in the AUC when tested on new data (A. C. Yu et al., 2022). This lack of generalizability can lead to adverse effects on how well the model performs in a real-world scenario.

    If a solution performs poorly when tested on a dataset with a similar or identical distribution to the training dataset, it is said to lack narrow generalizability and is often a consequence of overfitting (Eche et al., 2021). Potential solutions for overfitting are using larger training datasets and reducing the model’s complexity. If a solution performs poorly when tested on a dataset with a different distribution to the training dataset (e.g. a different distribution of patient ethnicities), it is said to lack broad generalizability (Eche et al., 2021). Solutions to poor broad generalizability include stress-testing the model on datasets with different distributions from the training dataset (Eche et al., 2021).

    AI solutions are often developed in a high-resource environment such as large technology companies and academic medical centres in wealthy countries. It is likely that findings and performance in these high-resource contexts will fail to generalize to lower- resource contexts such as smaller hospitals, rural areas or poorer countries (Price & Nicholson, 2019), which complicates the issue further.

    Risk of bias

    Biases can arise in AI-based solutions due to data or human factors. The former occurs when the data used to train the AI solution does not adequately represent the target population. Datasets can be unrepresentative when they are too small or have been collected in a way that misrepresents a certain population category. AI solutions trained on unrepresentative data perpetuate biases and perform poorly in the population categories underrepresented or misrepresented in the training data. The presence of such biases has been empirically shown in many AI-based medical imaging studies (Larrazabal et al., 2020; Seyyed-Kalantari et al., 2021).

    AI-based solutions are prone to several subjective and sometimes implicitly or explicitly prejudiced decisions during their development by humans. These human factors include how the training data is selected, how it is labelled, and how the decision is made to focus on the specific problem the AI-based solution intends to solve (Norori et al., 2021). Some recommendations and tools are available to help minimize the risk of bias in AI research (AIF360: A Comprehensive Set of Fairness Metrics for Datasets and Machine Learning Models, Explanations for These Metrics, and Algorithms to Mitigate Bias in Datasets and Models, n.d., IBM Watson Studio - Model Risk Management, n.d.; Silberg & Manyika, 2019).

    Data quantity, quality and variety

    Problems such as bias and lack of generalizability can be mitigated by ensuring that training data is of sufficient quantity, quality and variety. However, this is difficult to do because patients are often reluctant to share their data for commercial purposes (Aggarwal, Farag, et al., 2021; Ghafur et al., 2020; Trinidad et al., 2020), hospitals and clinics are usually not equipped to make this data available in a useable and secure manner, and organizing and labelling the data is time- consuming and expensive.

    Many datasets can be used for a number of different purposes, and sharing data between companies can help make the process of data collection and organization more efficient, as well as increase the amount of data available for each application. However, developers are often reluctant to share data with each other, or even reveal the exact source of their data, to stay competitive.

    Data protection and privacy

    The development and implementation of AI-based solutions require that patients are explicitly informed about, and give their consent to, the use of their data for a particular purpose and by certain people. This data also has to be adequately protected from data breaches and misuse. Failure to ensure this greatly undermines the public’s trust in AI-based solutions and hinders their adoption. While regulations governing health data privacy state that the collection of fully anonymized data does not require explicit patient consent (General Data Protection Regulation (GDPR) – Official Legal Text, 2016; Office for Civil Rights (OCR), 2012) and in theory protects from the data being misused, whether or not imaging data can be fully anonymized is controversial (Lotan et al., 2020; Murdoch, 2021). Whether consent can be truly informed considering the complexity of the data being acquired, and the resulting myriad of potential future uses of the data, is also disputed (Vayena & Blasimme, 2017).

    IT infrastructure

    Among hospital departments, radiology has always been at the forefront ofdigitalization. AI-based solutions that focus on image processing and interpretation are likely to find the prerequisite infrastructure in most radiology departments, for example for linking imaging equipment to computers for analysis and for archiving images and other outputs. However, most radiology departments are likely to require significant infrastructure upgrades for other applications of AI, particularly those requiring the integration of information from multiple sources and having complex outputs. Moreover, it is important to keep in mind that the distribution of necessary infrastructure is highly unequal across and within countries (Health Ethics & Governance, 2021).

    In terms of computing power, radiology departments will either have to invest resources into the hardware and personnel necessary to run these AI-based solutions or opt for cloud-based solutions. The former comes with an extra cost but allows data processing within the confines of the hospital or clinic’s local network. Cloud-based solutions for computing (known as “infrastructure as a service” or “IaaS”) are often considered the less secure and less trustworthy option, but this depends on a number of factors and is thus not always true (Baccianella & Gough, n.d.). Guidelines on what to consider when procuring cloud-based solutions in healthcare are available (Cloud Security for Healthcare Services, 2021).

    Lack of standardization, interoperability, and integrability

    The problem of infrastructure becomes even more complicated when considering how fragmented the AI medical imaging market currently is (Alexander et al., 2020). It is therefore likely that in the near future a single department will have several dozen AI-based solutions from different vendors running simultaneously. Having a separate self-contained infrastructure (e.g. a workstation or server) for each of these would be incredibly complicated and difficult to manage. Suggested solutions for this have included AI solution “marketplaces”, similar to app stores (Advanced AI Solutions for Radiology, n.d., Curated Marketplace, 2018, Imaging AI Marketplace - Overview, n.d., Sectra Amplifier Marketplace, 2021, The Nuance AI Marketplace for Diagnostic Imaging, n.d.), and development of an overarching vendor-neutral infrastructure (Leiner et al., 2021). The successful implementation of such solutions requires close partnerships between AI solution developers, imaging vendors and information technology companies.

    Interpretability

    It is often impossible to understand exactly how AI- based solutions come to their conclusions, particularly with complex approaches like deep learning. This reduces how transparent the decision-making process for procuring and approving these solutions can be, makes the identification of biases difficult, and makes it harder for clinicians to explain the outputs of these solutions to their patients and to determine whether a solution is working properly or has malfunctioned (Char et al., 2018; Reddy et al., 2020; Vayena et al., 2018; Whittlestone et al., 2019). Some have suggested that techniques that help humans understand how AI- based algorithms made certain decisions or predictions (“interpretable” or “explainable” AI) might help mitigate these challenges. However, others have argued that currently available techniques are unsuitable for understanding individual decisions of an algorithm and have warned against relying on them for ensuring that algorithms work in a safe and reliable way (Ghassemi et al., 2021).

    Liability

    In healthcare systems, a framework of accountability ensures that healthcare workers and medical institutions can be held responsible for adverse effects resulting from their actions. The question of who should be held accountable for the failures of an AI- based solution is complicated. For pharmaceuticals, for example, the accountability for inherent failures in the product or its use often lies with either the manufacturer or the prescriber. One key difference is that AI-based systems are continuously evolving and learning, and so inherently work in a way that is independent of what their developers could have foreseen (Yeung, 2018). To the end-user such as the healthcare worker, the AI- based solution may be opaque and so they may not be able to tell when the solution is malfunctioning or inaccurate (Habli et al., 2020; Yeung, 2018).

    Brittleness

    Despite substantial progress in their development over the past few years, deep learning algorithms are still surprising brittle. This means that, when the algorithm faces a scenario that differs substantially from what it faced during training, it cannot contextualize and often produces nonsensical or inaccurate results. This happens because, unlike humans, most algorithms learn to perceive things within the confines of certain assumptions, but fail to generalize outside these assumptions. As an example of how this can be abused with malicious intent, subtle changes to medical images, imperceptible by humans, can render the results of disease-classifying algorithms inaccurate (Finlayson et al., 2018). The lack of interpretability of many AI-based solutions compounds this problem because it makes it difficult to troubleshoot how they reached the wrong conclusion.

    So far, more than 100 AI-based products have gained conformité européenne (CE) marking or Food and Drug Adminstration (FDA) clearance. These products can be found in continuously updated and searchable online databases curated by the FDA (Center for Devices & Radiological Health, n.d.), the American College of Radiology (Assess-AI, n.d.), and others (AI for Radiology, n.d., The Medical Futurist, n.d.; E. Wu et al., 2021). The increasing number of available products, the inherent complexity of many of these solutions, and the fact that many people who usually make purchasing decisions in hospitals are not familiar with evaluating such products make it important to think carefully when deciding on which product to purchase. Such decisions will need to be made after incorporating input from healthcare workers, information technology (IT) professionals, as well as management, finance, legal, and human resources professionals within hospitals.

    Deciding on whether to purchase an AI-based solution in radiology, as well as which of the increasing number of commercially available solutions to purchase, includes considerations of quality, safety, and finances. Over the past few years, several guidelines have emerged to help potential buyers make these decisions (A Buyer’s Guide to AI in Health and Care, 2020; Omoumi et al., 2021; Reddy et al., 2021), and these guidelines are likely to evolve in the future with changing expectations from customers, regulatory bodies, and stakeholders involved in reimbursement decisions.

    First of all, it has to be clear to the potential buyer what the problem is and whether AI is the appropriate approach to this solution, or whether alternatives exist that are more advantageous on balance. If AI is the appropriate approach, buyers should know exactly what a potential AI-based product’s scope of the solution is - i.e. what specific problem the AI-based solution is designed to solve and in what specific circumstances. This includes whether the solution is intended for screening, diagnosis, monitoring, treatment recommendation or another application. It also includes the intended users of the solution and what kind of specific qualifications or training they are expected to have in order to be able to operate the solution and interpret its outputs. It needs to be clear to buyers whether the solution is intended to replace certain tasks that would normally be performed by the end-user, act as a double-reader, as a triaging mechanism, or for other tasks like quality control. Buyers should also understand whether the solution is intended to provide “new” information (i.e. information that would otherwise be unavailable to the user without the solution), improve the performance of an existing task beyond a human’s or other non-AI-based solution’s performance or if it is intended to save time or other resources.

    Buyers should also have access to information that allows them to assess the potential benefits of the AI solution, and this should be backed up by published scientific evidence for the efficacy and cost-efficiency of the solution. How this is done will depend highly on the solution itself and the context in which it is expected to be deployed, but guidelines for this are available (National Institute for Health and Care Excellence (NICE), n.d.). Some questions to ask here would be: How much of an influence will the solution have on patient management? Will it improve diagnostic performance? Will it save time and money? Will it affect patients’ quality of life? It should also be clear to the buyer who exactly is expected to benefit from the use of this solution (Radiologists? Clinicians? Patients? The healthcare system or society as a whole?).

    As with any healthcare intervention, all AI-based solutions come with potential risks, and these should be made clear to the buyer. Some of these risks might have legal consequences, such as the potential for misdiagnosis. These risks should be quantified, and potential buyers should have a framework for dealing with them, including identifying a framework for accountability within the organizations implementing these solutions. Buyers should also ensure they clearly understand the potential negative effects on radiologists’ training and the potential disruption to radiologists’ workflows associated with the use of these solutions.

    Specifics of the AI solution’s design are also relevant to the decision on whether or not to purchase it. These include how robust the solution is to differences between vendors and scanning parameters, the circumstances under which the algorithm was trained (including potential confounding factors), and the way that performance was assessed. It should also be clear to buyers if and how potential sources of bias were accounted for during development. Because a core characteristic of AI-based solutions is their ability to continuously learn from new data, whether and how exactly this retraining is incorporated into the solution with time should also be clear to the buyer, including whether or not new regulatory approval is needed with each iteration. This also includes whether or not retraining is required, for example, due to changes in imaging equipment at the buyer’s institution.

    The main selling points of many AI-based solutions are ease-of-use and improved workflows. Therefore, potential buyers should carefully scrutinize how these solutions are to be integrated into existing workflows, including inter-operability with PACS and electronic medical record systems. Whether or not the solution requires extra hardware (e.g. graphical processing units) or software (e.g. for visualization of the solution’s outputs), or if it can readily be integrated into the existing information technology infrastructure of the buyer’s organization influences the overall cost of the solution for the buyer and is therefore also a critical consideration. In addition, the degree of manual interaction required, both under normal circumstances and for troubleshooting, should be known to the buyer. All potential users of the AI solution should be involved in the purchasing process to ensure that they are familiar with it and that it meets their professional ethical standards and suits their needs.

    From a regulatory perspective, it should be clear to the buyer whether the solution complies with medical device and data protection regulations. Has the solution been approved in the buyer’s country? If so, under which risk classification? Buyers should also consider creating data flow maps that display how the data flows in the operation of the AI-based solution, including who has access to the data.

    Finally, there are other factors to consider which are not necessarily unique to AI-based solutions and which buyers might be familiar with from purchasing other types of solutions. This includes the licensing model of the solution, how users are to be trained on using the solution, how the solution is maintained, how failures in the solution are dealt with, and whether additional costs are to be expected when scaling up the solution’s implementation (e.g. using the solution for more imaging equipment or more users). This allows the potential buyer to anticipate the current and future costs of purchasing the solution.

    The past decade of increasing interest and progress in AI-based solutions for medical imaging has set the stage for a number of trends that are likely to appear or intensify in the near future.

    Firstly, there is an increasing sentiment that, although AI holds a great deal of promise for interpretive applications (such as the detection of pathology), non-interpretive AI-based solutions might hold the most potential in terms of instilling efficiency into radiology workflows and improving patient experiences. This trend towards involving AI earlier in the patient management process is likely to extend to AI increasingly acting as a clinical decision support system to guide when and which imaging scans are performed.

    For this to happen, AI needs to be integrated into existing clinical information systems, and the specific algorithms used need to be able to handle more varied data. This will likely pave the way for the development of algorithms that are capable of integrating demographic, clinical, and laboratory patient data to make recommendations about patient management (Huang, Pareek, et al., 2020; Rockenbach, 2021). The previously mentioned natural language processing algorithms that have been used to interpret scan requests may be useful candidates for this.

    In addition, we are likely to see AI algorithms that can interpret multiple different types of imaging data from the same patient. Currently, less than 5 % of commercially available AI-based solutions in medical imaging work with more than one imaging modality (Rezazade Mehrizi et al., 2021; van Leeuwen et al., 2021) despite the fact that the typical patient in a hospital receives multiple imaging scans during their stay (Shinagare et al., 2014). With this, it is also likely that more AI-based solutions will be developed that target hitherto neglected modalities such as nuclear imaging techniques and ultrasound.

    The current market for AI-based solutions in radiology is spread across a relatively large number of companies (Alexander et al., 2020). Potential users are likely to expect a streamlined integration of these products in their workflows, which can be challenging in such a fragmented market. Improved integration can be achieved in several different ways, including with vendor-neutral marketplaces or by the gradual consolidation of providers of AI-based solutions.

    With the expanding use of AI, the issue of trust between AI developers, healthcare professionals, regulators, and patients will become more relevant. It is therefore likely that efforts will intensify to take steps towards strengthening that trust. This will potentially include raising the expected standards of evidence for AI- based solutions (Aggarwal, Sounderajah, et al., 2021; X. Liu et al., 2019; van Leeuwen et al., 2021; Yusuf et al., 2020), making them more transparent through the use and improvement of interpretable AI techniques (Holzinger et al., 2017; Reyes et al., 2020; “Towards Trustable Machine Learning,” 2018), and enhancing techniques for maintaining patient data privacy (G. Kaissis et al., 2021; G. A. Kaissis et al., 2020).

    Furthermore, while most existing regulations stipulate that AI-based algorithms cannot be modified after regulatory approval, this is likely to change in the future. The potential for these algorithms to learn from data acquired after approval and adapt to changing circumstances is a major advantage of AI. Still, frameworks for doing so have thus far been lacking in the healthcare sector. However, promising ideas have recently emerged, including adapting existing hospital quality assurance and improvement frameworks to monitor AI-based algorithms’ performance and the data they are trained on and update the algorithms accordingly (Feng et al., 2022). This will likely require the development of multidisciplinary teams within hospitals consisting of clinicians, IT professionals, and biostatisticians who closely collaborate with model developers and regulators (Feng et al., 2022).

    While the obstacles discussed in previous sections might slow down the adoption of AI in radiology somewhat, the fear of AI potentially replacing radiologists is unlikely to be one of them. A recent survey from Europe showed that most radiologists did not perceive a reduction in their clinical workload after adopting AI-based solutions (European Society of Radiology (ESR), 2022), likely because, at the same time, demand for radiologists’ services has been continuously rising. Studies from around the world have shown that radiology professionals, particularly those with AI exposure and experience, are generally optimistic about the role of AI in their practice (Y. Chen et al., 2021; Huisman et al., 2021; Ooi et al., 2021; Santomartino & Yi, 2022; Scott et al., 2021).

    AI has shown promise in positively impacting virtually every facet of a radiology department’s work - from scheduling and protocolling patient scans to interpreting images and reaching diagnoses. Promising research on AI-based tools in radiology has not yet been widely translated to adoption in routine practice, however, because of a number of complex, partially intertwined issues. Potential solutions exist for many of these challenges, but many of these solutions require further refinement and testing. In the meantime, guidelines are emerging to help potential users of AI-based solutions in radiology navigate the increasing number of commercial products. This encourages their adoption in real-world scenarios, thus allowing their true potential to be uncovered, as well as their weaknesses to be identified and addressed in a safe and effective way. As these incremental improvements are made, these tools will likely evolve to handle more varied data, become integrated into consolidated workflows, become more transparent, and ultimately more useful for increasing efficiency and improving patient care.

    AAMC Report Reinforces Mounting Physician Shortage. (2021). AAMC. https://www.aamc.org/news-insights/press- releases/aamc-report-reinforces-mounting-physician-shortage

    A buyer’s guide to AI in health and care. (2020). NHS Transformation Directorate. https://www.nhsx.nhs.uk/ai-lab/ explore-all-resources/adopt-ai/a-buyers-guide-to-ai-in-health- and-care/

    Advanced AI solutions for radiology. (n.d.). Calantic Website. Retrieved July 3, 2022, from https://aivisions.calantic.com/

    Aggarwal, R., Farag, S., Martin, G., Ashrafian, H., & Darzi, A. (2021). Patient Perceptions on Data Sharing and Applying Artificial Intelligence to Health Care Data: Cross-sectional Survey. Journal of Medical Internet Research, 23(8), e26162. https://doi.org/10.2196/26162

    Aggarwal, R., Sounderajah, V., Martin, G., Ting, D. S. W., Karthikesalingam, A., King, D., Ashrafian, H., & Darzi, A. (2021). Diagnostic accuracy of deep learning in medical imaging: a systematic review and meta-analysis. NPJ Digital Medicine, 4(1), 65. https://doi.org/10.1038/s41746-021-00438-z

    Agnes, S. A., Anitha, J., Pandian, S. I. A., & Peter, J. D. (2019). Classification of Mammogram Images Using Multiscale all Convolutional Neural Network (MA-CNN). Journal of Medical Systems, 44(1), 30. https://doi.org/10.1007/s10916-019-1494-z

    AIF360: A comprehensive set of fairness metrics for datasets and machine learning models, explanations for these metrics, and algorithms to mitigate bias in datasets and models. (n.d.). Github. Retrieved June 11, 2022, from https:// github.com/Trusted-AI/AIF360

    AI for radiology. (n.d.). Retrieved June 26, 2022, from https://grand-challenge.org/
    aiforradiology/?subspeciality=Abdomen&modality=All&ce_ under=All&ce_class=All&fda_class=All&sort_by=last %20 modified&search=

    Akagi, M., Nakamura, Y., Higaki, T., Narita, K., Honda, Y., Zhou, J., Yu, Z., Akino, N., & Awai, K. (2019). Deep learning reconstruction improves image quality of abdominal ultra- high-resolution CT. European Radiology, 29(11), 6163–6171. https://doi.org/10.1007/s00330-019-06170-3

    Akkus, Z., Cai, J., Boonrod, A., Zeinoddini, A., Weston, A. D., Philbrick, K. A., & Erickson, B. J. (2019). A Survey of Deep- Learning Applications in Ultrasound: Artificial Intelligence- Powered Ultrasound for Improving Clinical Workflow. Journal of the American College of Radiology: JACR, 16(9 Pt B), 1318–1328. https://doi.org/10.1016/j.jacr.2019.06.004

    Al-Antari, M. A., Al-Masni, M. A., & Kim, T.-S. (2020). Deep Learning Computer-Aided Diagnosis for Breast Lesion in Digital Mammogram. Advances in Experimental Medicine and Biology, 1213, 59–72. https://doi.org/10.1007/978-3-030-33128-3_4

    Alexander, A., Jiang, A., Ferreira, C., & Zurkiya, D. (2020). An Intelligent Future for Medical Imaging: A Market Outlook on Artificial Intelligence for Medical Imaging. Journal of the American College of Radiology: JACR, 17(1 Pt B), 165–170. https:// doi.org/10.1016/j.jacr.2019.07.019

    Allen, B., Agarwal, S., Coombs, L., Wald, C., & Dreyer, K. (2021). 2020 ACR Data Science Institute Artificial Intelligence Journal of the American College of Radiology: JACR

    Alzubaidi, L., Zhang, J., Humaidi, A. J., Al-Dujaili, A., Duan, Y., Al-Shamma, O., Santamaría, J., Fadhel, M. A., Al-Amidie, M., & Farhan, L. (2021). Review of deep learning: concepts, CNN architectures, challenges, applications, future directions.

    Arbabshirani, M. R., Fornwalt, B. K., Mongelluzzo, G. J., Suever, J. D., Geise, B. D., Patel, A. A., & Moore, G. J. (2018). Advanced machine learning in action: identification of intracranial hemorrhage on computed tomography scans of the head with clinical workflow integration. NPJ Digital Medicine, 1, 9. https://doi.org/10.1038/s41746-017-0015-z

    Ardila, D., Kiraly, A. P., Bharadwaj, S., Choi, B., Reicher, J. J., Peng, L., Tse, D., Etemadi, M., Ye, W., Corrado, G., Naidich, D. P., & Shetty, S. (2019). End-to-end lung cancer screening with three-dimensional deep learning on low-dose chest computed tomography. Nature Medicine, 25(6), 954–961. https://doi. org/10.1038/s41591-019-0447-x

    Arefan, D., Mohamed, A. A., Berg, W. A., Zuley, M. L., Sumkin, J. H., & Wu, S. (2020). Deep learning modeling using normal mammograms for predicting breast cancer risk. Medical Physics, 47(1), 110–118. https://doi.org/10.1002/mp.13886

    Assess-AI. (n.d.). Retrieved July 2, 2022, from https://www. acrdsi.org/DSI-Services/Assess-AI

    Azer, S. A. (2019). Deep learning with convolutional neural networks for identification of liver masses and hepatocellular carcinoma: A systematic review. World Journal of Gastrointestinal Oncology, 11(12), 1218–1230. https://doi.org/10.4251/wjgo.v11. i12.1218

    Bacchi, S., Zerner, T., Oakden-Rayner, L., Kleinig, T., Patel, S., & Jannes, J. (2020). Deep Learning in the Prediction of Ischaemic Stroke Thrombolysis Functional Outcomes: A Pilot Study. Academic Radiology, 27(2), e19–e23. https://doi. org/10.1016/j.acra.2019.03.015

    Baccianella, S., & Gough, T. (n.d.). Why cloud computing is the best option for hospitals adopting AI. Retrieved June 11, 2022, from https://www.aidence.com/articles/cloud-best-option- imaging-ai/

    Balint, B. J., Steenburg, S. D., Lin, H., Shen, C., Steele, J. L., & Gunderman, R. B. (2014). Do telephone call interruptions have an impact on radiology resident diagnostic accuracy? Academic Radiology, 21(12), 1623–1628. https://doi.org/10.1016/j. acra.2014.08.001

    Baltruschat, I., Steinmeister, L., Nickisch, H., Saalbach, A., Grass, M., Adam, G., Knopp, T., & Ittrich, H. (2021). Smart chest X-ray worklist prioritization using artificial intelligence: a clinical workflow simulation. European Radiology, 31(6), 3837– 3845. https://doi.org/10.1007/s00330-020-07480-7

    Bash, S., Johnson, B., Gibbs, W., Zhang, T., Shankaranarayanan, A., & Tanenbaum, L. N. (2021). Deep Learning Image Processing Enables 40 % Faster Spinal MR Scans Which Match or Exceed Quality of Standard of Care : A Prospective Multicenter Multireader Study. Clinical Neuroradiology. https://doi.org/10.1007/s00062-021-01121-2

    Bash, S., Wang, L., Airriess, C., Zaharchuk, G., Gong, E., Shankaranarayanan, A., & Tanenbaum, L. N. (2021). Deep Learning Enables 60 % Accelerated Volumetric Brain MRI While Preserving Quantitative Performance: A Prospective, Multicenter, Multireader Trial. AJNR. American Journal of Neuroradiology, 42(12), 2130–2137. https://doi.org/10.3174/ajnr.A7358

    Boas, F. E., & Fleischmann, D. (2012). CT artifacts: causes and reduction techniques. Imaging in Medicine, 4(2), 229–240. https://doi.org/10.2217/iim.12.13

    Boland, G. W., Duszak, R., Jr, & Kalra, M. (2014). Protocol design and optimization. Journal of the American College of Radiology: JACR, 11(5), 440–441. https://doi.org/10.1016/j. jacr.2014.01.021

    Bonavita, I., Rafael-Palou, X., Ceresa, M., Piella, G., Ribas, V., & González Ballester, M. A. (2020). Integration of convolutional neural networks for pulmonary nodule malignancy assessment in a lung cancer classification pipeline. Computer Methods and Programs in Biomedicine, 185, 105172. https://doi.org/10.1016/j.cmpb.2019.105172

    Bozkurt, S., Cahan, E. M., Seneviratne, M. G., Sun, R., Lossio- Ventura, J. A., Ioannidis, J. P. A., & Hernandez-Boussard, T. (2020). Reporting of demographic data and representativeness in machine learning models using electronic health records.
    Journal of the American Medical Informatics Association: JAMIA, 27(12), 1878–1884. https://doi.org/10.1093/jamia/ocaa164

    Brown, A. D., & Marotta, T. R. (2018). Using machine learning for sequence-level automated MRI protocol selection in neuroradiology. Journal of the American Medical Informatics Association: JAMIA, 25(5), 568–571. https://doi.org/10.1093/ jamia/ocx125

    Bruls, R. J. M., & Kwee, R. M. (2020). Workload for radiologists during on-call hours: dramatic increase in the past 15 years.
    Insights into Imaging, 11(1), 121. https://doi.org/10.1186/ s13244-020-00925-z

    Buruk, B., Ekmekci, P. E., & Arda, B. (2020). A critical perspective on guidelines for responsible and trustworthy artificial intelligence. Medicine, Health Care, and Philosophy, 23(3), 387–399. https://doi.org/10.1007/s11019-020-09948-1

    Cadario, R., Longoni, C., & Morewedge, C. K. (2021). Understanding, explaining, and utilizing medical artificial intelligence. Nature Human Behaviour, 5(12), 1636–1642. https://doi.org/10.1038/s41562-021-01146-0

    Center for Devices, & Radiological Health. (n.d.). Artificial Intelligence and Machine Learning (AI/ML)-Enabled Medical Devices. U.S. Food and Drug Administration; FDA. Retrieved July 2, 2022, from https://www.fda.gov/medical-devices/software- medical-device-samd/artificial-intelligence-and-machine- learning-aiml-enabled-medical-devices

    Char, D. S., Shah, N. H., & Magnus, D. (2018). Implementing Machine Learning in Health Care - Addressing Ethical Challenges. The New England Journal of Medicine, 378(11), 981–983. https://doi.org/10.1056/NEJMp1714229

    Chen, H., Zhang, Y., Kalra, M. K., Lin, F., Chen, Y., Liao, P., Zhou, J., & Wang, G. (2017). Low-Dose CT With a Residual Encoder-Decoder Convolutional Neural Network. IEEE Transactions on Medical Imaging, 36(12), 2524–2535. https://doi.org/10.1109/TMI.2017.2715284

    Chen, Y., Stavropoulou, C., Narasinkan, R., Baker, A., & Scarbrough, H. (2021). Professionals’ responses to the introduction of AI innovations in radiology and their implications for future adoption: a qualitative study. BMC Health Services Research, 21(1), 813. https://doi.org/10.1186/ s12913-021-06861-y

    Choe, J., Lee, S. M., Do, K.-H., Lee, G., Lee, J.-G., Lee, S. M., & Seo, J. B. (2019). Deep Learning-based Image Conversion of CT Reconstruction Kernels Improves Radiomics Reproducibility for Pulmonary Nodules or Masses. Radiology, 292(2), 365–373. https://doi.org/10.1148/radiol.2019181960

    Choi, K. S., Choi, S. H., & Jeong, B. (2019). Prediction of IDH genotype in gliomas with dynamic susceptibility contrast perfusion MR imaging using an explainable recurrent neural network. Neuro-Oncology, 21(9), 1197–1209. https://doi.org/10.1093/neuonc/noz095

    Chong, L. R., Tsai, K. T., Lee, L. L., Foo, S. G., & Chang, P. C. (2020). Artificial Intelligence Predictive Analytics in the Management of Outpatient MRI Appointment No-Shows. AJR. American Journal of Roentgenology, 215(5), 1155–1162. https://doi.org/10.2214/AJR.19.22594

    Cikes, M., Sanchez-Martinez, S., Claggett, B., Duchateau, N., Piella, G., Butakoff, C., Pouleur, A. C., Knappe, D., Biering- Sørensen, T., Kutyifa, V., Moss, A., Stein, K., Solomon, S. D., & Bijnens, B. (2019). Machine learning-based phenogrouping in heart failure to identify responders to cardiac resynchronization therapy. European Journal of Heart Failure, 21(1), 74–85. https://doi.org/10.1002/ejhf.1333

    Ciompi, F., Chung, K., van Riel, S. J., Setio, A. A. A., Gerke, P. K., Jacobs, C., Scholten, E. T., Schaefer-Prokop, C., Wille,
    M. M. W., Marchianò, A., Pastorino, U., Prokop, M., & van Ginneken, B.
    (2017). Towards automatic pulmonary nodule management in lung cancer screening with deep learning. Scientific Reports, 7, 46479. https://doi.org/10.1038/srep46479

    Clinical radiology UK workforce census 2019 report. (2019). https://www.rcr.ac.uk/publication/clinical-radiology-uk- workforce-census-2019-report

    Cloud security for healthcare services. (2021, January 14). ENISA. https://www.enisa.europa.eu/publications/cloud- security-for-healthcare-services/

    CONSORT-AI and SPIRIT-AI Steering Group. (2019). Reporting guidelines for clinical trials evaluating artificial intelligence interventions are needed. Nature Medicine, 25(10), 1467–1468. https://doi.org/10.1038/s41591-019-0603-3

    Curated marketplace. (2018, May 22). Blackford. https://www.blackfordanalysis.com/applications/

    Dance, A. (2021). AI spots cell structures that humans can’t. Nature. 592 (7852), 154–155.

    Dantas, L. F., Fleck, J. L., Cyrino Oliveira, F. L., & Hamacher, S. (2018). No-shows in appointment scheduling - a systematic literature review. Health Policy, 122(4), 412–421. https://doi.org/10.1016/j.healthpol.2018.02.002

    Deák, Z., Grimm, J. M., Treitl, M., Geyer, L. L., Linsenmaier, U., Körner, M., Reiser, M. F., & Wirth, S. (2013). Filtered back projection, adaptive statistical iterative reconstruction, and a model-based iterative reconstruction in abdominal CT: an experimental clinical study. Radiology, 266(1), 197–206. https://doi.org/10.1148/radiol.12112707

    Dembrower, K., Liu, Y., Azizpour, H., Eklund, M., Smith, K., Lindholm, P., & Strand, F. (2020). Comparison of a Deep
    Learning Risk Score and Standard Mammographic Density Score for Breast Cancer Risk Prediction. Radiology, 294(2), 265–272. https://doi.org/10.1148/radiol.2019190872

    Do, B. H., Langlotz, C., & Beaulieu, C. F. (2017). Bone Tumor Diagnosis Using a Naïve Bayesian Model of Demographic and Radiographic Features. Journal of Digital Imaging, 30(5), 640–647. https://doi.org/10.1007/s10278-017-0001-7

    Dou, Q., Yu, L., Chen, H., Jin, Y., Yang, X., Qin, J., & Heng, P.-A. (2017). 3D deeply supervised network for automated segmentation of volumetric medical images. Medical Image Analysis, 41, 40–54. https://doi.org/10.1016/j. media.2017.05.001

    Eche, T., Schwartz, L. H., Mokrane, F.-Z., & Dercle, L. (2021). Toward Generalizability in the Deployment of Artificial Intelligence in Radiology: Role of Computation Stress Testing to Overcome Underspecification. Radiology. Artificial Intelligence, 3(6), e210097. https://doi.org/10.1148/ryai.2021210097

    England, N. H. S., & Improvement, N. H. S. (2019). NHS diagnostic waiting times and activity data. NHS. https://www. england.nhs.uk/statistics/wp-content/uploads/sites/2/2021/12/ DWTA-Report-October-2021_M43D4.pdf

    Esmaeilzadeh, P. (2020). Use of AI-based tools for healthcare purposes: a survey study from consumers’ perspectives. BMC Medical Informatics and Decision Making, 20(1), 170. https://doi. org/10.1186/s12911-020-01191-1

    Esses, S. J., Lu, X., Zhao, T., Shanbhogue, K., Dane, B., Bruno, M., & Chandarana, H. (2018). Automated image quality evaluation of T2 -weighted liver MRI utilizing deep learning architecture. Journal of Magnetic Resonance Imaging: JMRI, 47(3), 723–728. https://doi.org/10.1002/jmri.25779

    European Society of Radiology (ESR). (2022). Current practical experience with artificial intelligence in clinical radiology:
    a survey of the European Society of Radiology. Insights into Imaging, 13(1), 107. https://doi.org/10.1186/s13244-022- 01247-y

    Faron, A., Sichtermann, T., Teichert, N., Luetkens, J. A., Keulers, A., Nikoubashman, O., Freiherr, J., Mpotsaris, A., & Wiesmann, M. (2020). Performance of a Deep-Learning Neural Network to Detect Intracranial Aneurysms from 3D TOF-MRA Compared to Human Readers. Clinical Neuroradiology, 30(3), 591–598. https://doi.org/10.1007/s00062-019-00809-w

    Feng, J., Phillips, R. V., Malenica, I., Bishara, A., Hubbard, A. E., Celi, L. A., & Pirracchio, R. (2022). Clinical artificial intelligence quality improvement: towards continual monitoring and updating of AI algorithms in healthcare. NPJ Digital Medicine, 5(1), 66. https://doi.org/10.1038/s41746-022- 00611-y

    Finlayson, S. G., Chung, H. W., Kohane, I. S., & Beam, A. L. (2018). Adversarial Attacks Against Medical Deep Learning Systems. In arXiv [cs.CR]. arXiv. https://doi.org/10.1145/nnnnnnn. nnnnnnn

    Flanders, A. E., Prevedello, L. M., Shih, G., Halabi, S. S., Kalpathy-Cramer, J., Ball, R., Mongan, J. T., Stein, A., Kitamura, F. C., Lungren, M. P., Choudhary, G., Cala, L., Coelho, L., Mogensen, M., Morón, F., Miller, E., Ikuta, I., Zohrabian, V., McDonnell, O., … RSNA-ASNR 2019 Brain Hemorrhage CT Annotators. (2020). Construction of a Machine Learning Dataset through Collaboration: The RSNA 2019 Brain CT Hemorrhage Challenge. Radiology. Artificial Intelligence, 2(3), e190211. https://doi.org/10.1148/ryai.2020190211

    Freeman, K., Geppert, J., Stinton, C., Todkill, D., Johnson, S., Clarke, A., & Taylor-Phillips, S. (2021). Use of artificial intelligence for image analysis in breast cancer screening programmes: systematic review of test accuracy. BMJ , 374, n1872. https://doi.org/10.1136/bmj.n1872

    General Data Protection Regulation (GDPR) – Official Legal Text. (2016, July 13). General Data Protection Regulation (GDPR). https://gdpr-info.eu/

    Ghafur, S., Van Dael, J., Leis, M., Darzi, A., & Sheikh, A. (2020). Public perceptions on data sharing: key insights from the UK and the USA. The Lancet. Digital Health, 2(9), e444–e446. https://doi.org/10.1016/S2589-7500(20)30161-8

    Ghani, M. U., & Clem Karl, W. (2019). Fast Enhanced CT Metal Artifact Reduction using Data Domain Deep Learning. In arXiv [cs.CV]. arXiv. http://arxiv.org/abs/1904.04691

    Ghassemi, M., Oakden-Rayner, L., & Beam, A. L. (2021). The false hope of current approaches to explainable artificial intelligence in health care. The Lancet. Digital Health, 3(11), e745–e750. https://doi.org/10.1016/S2589-7500(21)00208-9

    Ginat, D. T. (2020). Analysis of head CT scans flagged by deep learning software for acute intracranial hemorrhage. Neuroradiology, 62(3), 335–340. https://doi.org/10.1007/ s00234-019-02330-w

    Goebel, J., Stenzel, E., Guberina, N., Wanke, I., Koehrmann, M., Kleinschnitz, C., Umutlu, L., Forsting, M., Moenninghoff, C., & Radbruch, A. (2018). Automated ASPECT rating: comparison between the Frontier ASPECT Score software and the Brainomix software. Neuroradiology, 60(12), 1267–1272. https://doi.org/10.1007/s00234-018-2098-x

    Habli, I., Lawton, T., & Porter, Z. (2020). Artificial intelligence in health care: accountability and safety. Bulletin of the World Health Organization, 98(4), 251–256. https://doi.org/10.2471/ BLT.19.237487

    Halabi, S. S., Prevedello, L. M., Kalpathy-Cramer, J., Mamonov, A. B., Bilbily, A., Cicero, M., Pan, I., Pereira, L. A., Sousa, R. T., Abdala, N., Kitamura, F. C., Thodberg, H. H., Chen, L., Shih, G., Andriole, K., Kohli, M. D., Erickson, B. J., & Flanders, A. E. (2019). The RSNA Pediatric Bone Age Machine Learning Challenge. Radiology, 290(2), 498–503. https://doi. org/10.1148/radiol.2018180736

    Hargreaves, B. A., Worters, P. W., Pauly, K. B., Pauly, J. M., Koch, K. M., & Gold, G. E. (2011). Metal-induced artifacts in MRI. AJR. American Journal of Roentgenology, 197(3), 547–555. https://doi.org/10.2214/AJR.11.7364

    Harry, E., Sinsky, C., Dyrbye, L. N., Makowski, M. S., Trockel, M., Tutty, M., Carlasare, L. E., West, C. P., & Shanafelt, T. D. (2021). Physician Task Load and the Risk of Burnout Among US Physicians in a National Survey. Joint Commission Journal on Quality and Patient Safety / Joint Commission Resources, 47(2), 76–85. https://doi.org/10.1016/j.jcjq.2020.09.011

    Hata, A., Yanagawa, M., Yamagata, K., Suzuki, Y., Kido, S., Kawata, A., Doi, S., Yoshida, Y., Miyata, T., Tsubamoto, M., Kikuchi, N., & Tomiyama, N. (2021). Deep learning algorithm for detection of aortic dissection on non-contrast-enhanced CT. European Radiology, 31(2), 1151–1159. https://doi.org/10.1007/ s00330-020-07213-w

    Hauptmann, A., Arridge, S., Lucka, F., Muthurangu, V., & Steeden, J. A. (2019). Real-time cardiovascular MR with
    spatio-temporal artifact suppression using deep learning-proof of concept in congenital heart disease. Magnetic Resonance in Medicine: Official Journal of the Society of Magnetic Resonance in Medicine / Society of Magnetic Resonance in Medicine, 81(2), 1143–1156. https://doi.org/10.1002/mrm.27480

    Health Ethics & Governance. (2021, June 28). Ethics and governance of artificial intelligence for health. World Health Organization. https://www.who.int/publications/i/ item/9789240029200

    He, L., Li, H., Dudley, J. A., Maloney, T. C., Brady, S. L., Somasundaram, E., Trout, A. T., & Dillman, J. R. (2019). Machine Learning Prediction of Liver Stiffness Using Clinical and T2-Weighted MRI Radiomic Data. AJR. American Journal of Roentgenology, 213(3), 592–601. https://doi.org/10.2214/ AJR.19.21082

    Herent, P., Schmauch, B., Jehanno, P., Dehaene, O., Saillard, C., Balleyguier, C., Arfi-Rouche, J., & Jégou, S. (2019).
    Detection and characterization of MRI breast lesions using deep learning. Diagnostic and Interventional Imaging, 100(4), 219–225. https://doi.org/10.1016/j.diii.2019.02.008

    Hinton, B., Ma, L., Mahmoudzadeh, A. P., Malkov, S., Fan, B., Greenwood, H., Joe, B., Lee, V., Kerlikowske, K., & Shepherd,
    J.
    (2019). Deep learning networks find unique mammographic differences in previous negative mammograms between interval and screen-detected cancers: a case-case study. Cancer Imaging: The Official Publication of the International Cancer Imaging Society, 19(1), 41. https://doi.org/10.1186/s40644-019-0227-3

    Holzinger, A., Biemann, C., Pattichis, C. S., & Kell, D. B. (2017). What do we need to build explainable AI systems for the medical domain? In arXiv [cs.AI]. arXiv. http://arxiv.org/ abs/1712.09923

    Hötker, A. M., Da Mutten, R., Tiessen, A., Konukoglu, E., & Donati, O. F. (2021). Improving workflow in prostate MRI:
    AI-based decision-making on biparametric or multiparametric MRI. Insights into Imaging, 12(1), 112. https://doi.org/10.1186/ s13244-021-01058-7

    Huang, S.-C., Kothari, T., Banerjee, I., Chute, C., Ball, R. L., Borus, N., Huang, A., Patel, B. N., Rajpurkar, P., Irvin, J., Dunnmon, J., Bledsoe, J., Shpanskaya, K., Dhaliwal, A., Zamanian, R., Ng, A. Y., & Lungren, M. P. (2020). PENet-a scalable deep-learning model for automated diagnosis of pulmonary embolism using volumetric CT imaging. NPJ Digital Medicine, 3, 61. https://doi.org/10.1038/s41746-020-0266-y

    Huang, S.-C., Pareek, A., Seyyedi, S., Banerjee, I., & Lungren, M. P. (2020). Fusion of medical imaging and electronic health records using deep learning: a systematic review and implementation guidelines. NPJ Digital Medicine, 3, 136. https:// doi.org/10.1038/s41746-020-00341-z

    Huisman, M., Ranschaert, E., Parker, W., Mastrodicasa, D., Koci, M., Pinto de Santos, D., Coppola, F., Morozov, S., Zins, M., Bohyn, C., Koç, U., Wu, J., Veean, S., Fleischmann, D., Leiner, T., & Willemink, M. J. (2021). An international survey on AI in radiology in 1,041 radiologists and radiology residents part 1: fear of replacement, knowledge, and attitude. European Radiology, 31(9), 7058–7066. https://doi.org/10.1007/s00330-021-07781-5

    Hu, S.-Y., Santus, E., Forsyth, A. W., Malhotra, D., Haimson, J., Chatterjee, N. A., Kramer, D. B., Barzilay, R., Tulsky,
    J. A., & Lindvall, C.
    (2019). Can machine learning improve patient selection for cardiac resynchronization therapy? PloS One, 14(10), e0222397. https://doi.org/10.1371/journal. pone.0222397

    Hwang, E. J., Nam, J. G., Lim, W. H., Park, S. J., Jeong, Y. S., Kang, J. H., Hong, E. K., Kim, T. M., Goo, J. M., Park, S., Kim, K. H., & Park, C. M. (2019). Deep Learning for Chest Radiograph Diagnosis in the Emergency Department. Radiology, 293(3), 573–580. https://doi.org/10.1148/radiol.2019191225

    Hwang, E. J., Park, S., Jin, K.-N., Kim, J. I., Choi, S. Y., Lee, J. H., Goo, J. M., Aum, J., Yim, J.-J., Park, C. M., & Deep Learning- Based Automatic Detection Algorithm Development and Evaluation Group. (2019). Development and Validation of a Deep Learning-based Automatic Detection Algorithm for Active Pulmonary Tuberculosis on Chest Radiographs. Clinical Infectious Diseases: An Official Publication of the Infectious Diseases Society of America, 69(5), 739–747. https://doi.org/10.1093/cid/ciy967

    Hwang, S., Kim, H.-E., Jeong, J., & Kim, H.-J. (2016). A novel approach for tuberculosis screening based on deep convolutional neural networks. In G. D. Tourassi & S. G. Armato (Eds.), Medical Imaging 2016: Computer-Aided Diagnosis. SPIE. https://doi.org/10.1117/12.2216198

    IBM Watson Studio - Model Risk Management. (n.d.). Retrieved June 11, 2022, from https://www.ibm.com/cloud/ watson-studio/model-risk-management

    Imaging AI Marketplace - overview. (n.d.). Retrieved June 11, 2022, from https://www.ibm.com/products/imaging-ai- marketplace

    Jamaludin, A., Lootus, M., Kadir, T., Zisserman, A., Urban, J., Battié, M. C., Fairbank, J., McCall, I., & Genodisc Consortium. (2017). ISSLS PRIZE IN BIOENGINEERING SCIENCE 2017:
    Automation of reading of radiological features from magnetic resonance images (MRIs) of the lumbar spine without human intervention is comparable with an expert radiologist. European Spine Journal: Official Publication of the European Spine Society, the European Spinal Deformity Society, and the European Section of the Cervical Spine Research Society, 26(5), 1374–1383. https:// doi.org/10.1007/s00586-017-4956-3

    Kaissis, G. A., Makowski, M. R., Rückert, D., & Braren, R. F. (2020). Secure, privacy-preserving and federated machine learning in medical imaging. Nature Machine Intelligence, 2(6), 305–311. https://doi.org/10.1038/s42256-020-0186-1

    Kaissis, G., Ziller, A., Passerat-Palmbach, J., Ryffel, T., Usynin, D., Trask, A., Lima, I., Mancuso, J., Jungmann, F., Steinborn, M.-M., Saleh, A., Makowski, M., Rueckert, D., & Braren, R. (2021). End-to-end privacy preserving deep learning on multi-institutional medical imaging. Nature Machine Intelligence, 3(6), 473–484. https://doi.org/10.1038/s42256-021-00337-8

    Kalra, A., Chakraborty, A., Fine, B., & Reicher, J. (2020). Machine Learning for Automation of Radiology Protocols for Quality and Efficiency Improvement. Journal of the American College of Radiology: JACR, 17(9), 1149–1158. https://doi.org/10.1016/j.jacr.2020.03.012

    Kao, P.-Y., Chen, J. W., & Manjunath, B. S. (2019). Improving 3D U-Net for Brain Tumor Segmentation by Utilizing Lesion Prior. In arXiv [cs.CV]. arXiv. http://arxiv.org/abs/1907.00281

    Kapoor, N., Lacson, R., & Khorasani, R. (2020). Workflow Applications of Artificial Intelligence in Radiology and an Overview of Available Tools. Journal of the American College of Radiology: JACR, 17(11), 1363–1370. https://doi.org/10.1016/j. jacr.2020.08.016

    Kathirvelu, D., Vinupritha, P., & Kalpana, V. (2019). A computer aided diagnosis system for measurement of mandibular cortical thickness on dental panoramic radiographs in prediction of women with low bone mineral density. Journal of Medical Systems, 43(6), 148. https://doi.org/10.1007/s10916-019-1268-7

    Ker, J., Singh, S. P., Bai, Y., Rao, J., Lim, T., & Wang, L. (2019). Image Thresholding Improves 3-Dimensional Convolutional Neural Network Diagnosis of Different Acute Brain Hemorrhages on Computed Tomography Scans. Sensors, 19(9). https://doi.org/10.3390/s19092167

    Khan, F. A., Majidulla, A., Tavaziva, G., Nazish, A., Abidi, S. K., Benedetti, A., Menzies, D., Johnston, J. C., Khan, A. J., & Saeed, S. (2020). Chest x-ray analysis with deep learning- based software as a triage test for pulmonary tuberculosis: a prospective study of diagnostic accuracy for culture-confirmed disease. The Lancet. Digital Health, 2(11), e573–e581. https://doi.org/10.1016/S2589-7500(20)30221-1

    Kim, D. W., Jang, H. Y., Kim, K. W., Shin, Y., & Park, S. H. (2019). Design Characteristics of Studies Reporting the Performance of Artificial Intelligence Algorithms for Diagnostic Analysis of Medical Images: Results from Recently Published Papers. Korean Journal of Radiology: Official Journal of the Korean Radiological Society, 20(3), 405–410. https://doi.org/10.3348/ kjr.2019.0025

    Kim, K. H., & Park, S.-H. (2017). Artificial neural network for suppression of banding artifacts in balanced steady-state free precession MRI. Magnetic Resonance Imaging, 37, 139–146. https://doi.org/10.1016/j.mri.2016.11.020

    Korteling, J. E. H., van de Boer-Visschedijk, G. C., Blankendaal, R. A. M., Boonekamp, R. C., & Eikelboom, A. R. (2021). Human- versus Artificial Intelligence. Frontiers in Artificial Intelligence 4, 622364. https://doi.org/10.3389/ frai.2021.622364

    Kühl, N., Goutier, M., Baier, L., Wolff, C., & Martin, D. (2020). Human vs. supervised machine learning: Who learns patterns faster? In arXiv [cs.AI] arXiv. http://arxiv.org/abs/2012.03661

    Kuo, W., Häne, C., Mukherjee, P., Malik, J., & Yuh, E. L. (2019). Expert-level detection of acute intracranial hemorrhage on head computed tomography using deep learning. Proceedings of the National Academy of Sciences of the United States of America, 116(45), 22737–22745. https://doi.org/10.1073/ pnas.1908021116

    Langerhuizen, D. W. G., Janssen, S. J., Mallee, W. H., van den Bekerom, M. P. J., Ring, D., Kerkhoffs, G. M. M. J., Jaarsma,
    R. L., & Doornberg, J. N.
    (2019). What Are the Applications and Limitations of Artificial Intelligence for Fracture Detection and Classification in Orthopaedic Trauma Imaging? A Systematic Review. Clinical Orthopaedics and Related Research, 477(11), 2482–2491. https://doi.org/10.1097/CORR.0000000000000848

    Lang, N., Zhang, Y., Zhang, E., Zhang, J., Chow, D., Chang, P., Yu, H. J., Yuan, H., & Su, M.-Y. (2019). Differentiation of spinal metastases originated from lung and other cancers using radiomics and deep learning based on DCE-MRI. Magnetic Resonance Imaging, 64, 4–12. https://doi.org/10.1016/j. mri.2019.02.013

    Larrazabal, A. J., Nieto, N., Peterson, V., Milone, D. H., & Ferrante, E. (2020). Gender imbalance in medical imaging datasets produces biased classifiers for computer-aided diagnosis. Proceedings of the National Academy of Sciences of the United States of America, 117(23), 12592–12594. https://doi.org/10.1073/pnas.1919012117

    Lee, J.-S., Adhikari, S., Liu, L., Jeong, H.-G., Kim, H., & Yoon, S.-J. (2019). Osteoporosis detection in panoramic radiographs using a deep convolutional neural network-based computer- assisted diagnosis system: a preliminary study. Dento Maxillo Facial Radiology, 48(1), 20170344. https://doi.org/10.1259/ dmfr.20170344

    Lee, Y. H. (2018). Efficiency Improvement in a Busy Radiology Practice: Determination of Musculoskeletal Magnetic Resonance Imaging Protocol Using Deep-Learning Convolutional Neural Networks. Journal of Digital Imaging, 31(5), 604–610. https://doi.org/10.1007/s10278-018-0066-y

    Leiner, T., Bennink, E., Mol, C. P., Kuijf, H. J., & Veldhuis, W. B. (2021). Bringing AI to the clinic: blueprint for a vendor-neutral AI deployment infrastructure. Insights into Imaging, 12(1), 11. https://doi.org/10.1186/s13244-020-00931-1

    Lekadir, K., Osuala, R., Gallin, C., Lazrak, N., Kushibar, K., Tsakou, G., Aussó, S., Alberich, L. C., Marias, K., Tsiknakis, M., Colantonio, S., Papanikolaou, N., Salahuddin, Z., Woodruff, H. C., Lambin, P., & Martí-Bonmatí, L. (2021). FUTURE-AI: Guiding Principles and Consensus Recommendations for Trustworthy Artificial Intelligence in Medical Imaging. In arXiv [cs.CV]. arXiv. http://arxiv.org/ abs/2109.09658

    Letourneau-Guillon, L., Camirand, D., Guilbert, F., & Forghani, R. (2020). Artificial Intelligence Applications for Workflow, Process Optimization and Predictive Analytics. Neuroimaging Clinics of North America, 30(4), e1–e15. https://doi.org/10.1016/j.nic.2020.08.008

    Levin, D. C., Parker, L., & Rao, V. M. (2017). Recent Trends in Imaging Use in Hospital Settings: Implications for Future Planning. Journal of the American College of Radiology: JACR, 14(3), 331–336. https://doi.org/10.1016/j.jacr.2016.08.025

    Lindsey, R., Daluiski, A., Chopra, S., Lachapelle, A., Mozer, M., Sicular, S., Hanel, D., Gardner, M., Gupta, A., Hotchkiss, R., & Potter, H. (2018). Deep neural network improves fracture detection by clinicians. Proceedings of the National Academy of Sciences of the United States of America, 115(45), 11591–11596. https://doi.org/10.1073/pnas.1806905115

    Liu, F., Tang, J., Ma, J., Wang, C., Ha, Q., Yu, Y., & Zhou, Z. (2021). The application of artificial intelligence to chest medical image analysis. Intelligent Medicine, 1(3), 104–117. https://doi.org/10.1016/j.imed.2021.06.004

    Liu, F., Zhou, Z., Samsonov, A., Blankenbaker, D., Larison, W., Kanarek, A., Lian, K., Kambhampati, S., & Kijowski, R. (2018). Deep Learning Approach for Evaluating Knee MR Images: Achieving High Diagnostic Performance for Cartilage Lesion Detection. Radiology, 289(1), 160–169. https://doi.org/10.1148/ radiol.2018172986

    Liu, X., Cruz Rivera, S., Moher, D., Calvert, M. J., Denniston, A. K., & SPIRIT-AI and CONSORT-AI Working Group. (2020). Reporting guidelines for clinical trial reports for interventions involving artificial intelligence: the CONSORT-AI extension. Nature Medicine, 26(9), 1364–1374. https://doi.org/10.1038/ s41591-020-1034-x

    Liu, X., Faes, L., Kale, A. U., Wagner, S. K., Fu, D. J., Bruynseels, A., Mahendiran, T., Moraes, G., Shamdas, M., Kern, C., Ledsam, J. R., Schmid, M. K., Balaskas, K., Topol, E. J., Bachmann, L. M., Keane, P. A., & Denniston, A. K. (2019). A comparison of deep learning performance against health-care professionals in detecting diseases from medical imaging: a systematic review and meta-analysis. The Lancet. Digital Health, 1(6), e271–e297. https://doi.org/10.1016/S2589- 7500(19)30123-2

    Li, X., Shen, L., Xie, X., Huang, S., Xie, Z., Hong, X., & Yu, J. (2020). Multi-resolution convolutional networks for chest X-ray radiograph based lung nodule detection. Artificial Intelligence in Medicine, 103, 101744. https://doi.org/10.1016/j. artmed.2019.101744

    Lotan, E., Tschider, C., Sodickson, D. K., Caplan, A. L., Bruno, M., Zhang, B., & Lui, Y. W. (2020). Medical Imaging and Privacy in the Era of Artificial Intelligence: Myth, Fallacy, and the Future. Journal of the American College of Radiology: JACR, 17(9), 1159–1162. https://doi.org/10.1016/j.jacr.2020.04.007

    Maegerlein, C., Fischer, J., Mönch, S., Berndt, M., Wunderlich, S., Seifert, C. L., Lehm, M., Boeckh-Behrens, T., Zimmer, C., & Friedrich, B. (2019). Automated Calculation of the Alberta Stroke Program Early CT Score: Feasibility and
    Reliability. Radiology, 291(1), 141–148. https://doi.org/10.1148/ radiol.2019181228

    Mairhöfer, D., Laufer, M., Simon, P. M., Sieren, M., Bischof, A., Käster, T., Barth, E., Barkhausen, J., & Martinetz, T. (2021). An AI-based Framework for Diagnostic Quality Assessment of Ankle Radiographs. https://openreview.net/ pdf?id=bj04hJss_xZ

    Mancio, J., Pashakhanloo, F., El-Rewaidy, H., Jang, J., Joshi, G., Csecs, I., Ngo, L., Rowin, E., Manning, W., Maron, M., & Nezafat, R. (2022). Machine learning phenotyping of scarred myocardium from cine in hypertrophic cardiomyopathy. European Heart Journal Cardiovascular Imaging, 23(4), 532–542. https://doi.org/10.1093/ehjci/jeab056

    Matsoukas, S., Morey, J., Lock, G., Chada, D., Shigematsu, T., Marayati, N. F., Delman, B. N., Doshi, A., Majidi, S.,
    De Leacy, R., Kellner, C. P., & Fifi, J. T.
    (2022). AI software detection of large vessel occlusion stroke on CT angiography: a real-world prospective diagnostic test accuracy study. Journal of Neurointerventional Surgery. https://doi.org/10.1136/ neurintsurg-2021-018391

    McKinney, S. M., Sieniek, M., Godbole, V., Godwin, J., Antropova, N., Ashrafian, H., Back, T., Chesus, M., Corrado, G. S., Darzi, A., Etemadi, M., Garcia-Vicente, F., Gilbert, F. J., Halling-Brown, M., Hassabis, D., Jansen, S., Karthikesalingam, A., Kelly, C. J., King, D., … Shetty, S. (2020). International evaluation of an AI system for breast cancer screening. Nature, 577(7788), 89–94. https://doi.org/10.1038/s41586-019-1799-6

    McLeavy, C. M., Chunara, M. H., Gravell, R. J., Rauf, A., Cushnie, A., Staley Talbot, C., & Hawkins, R. M. (2021). The future of CT: deep learning reconstruction. Clinical Radiology, 76(6), 407–415. https://doi.org/10.1016/j.crad.2021.01.010

    Medical AI evaluation. (n.d.). Retrieved June 26, 2022, from https://ericwu09.github.io/medical-ai-evaluation/

    Mlynarski, P., Delingette, H., Criminisi, A., & Ayache, N. (2019). Deep learning with mixed supervision for brain tumor segmentation. Journal of Medical Imaging (Bellingham, Wash.), 6(3), 034002. https://doi.org/10.1117/1.JMI.6.3.034002

    Mongan, J., Moy, L., & Kahn, C. E., Jr. (2020). Checklist for Artificial Intelligence in Medical Imaging (CLAIM): A Guide for Authors and Reviewers. Radiology. Artificial Intelligence, 2(2), e200029. https://doi.org/10.1148/ryai.2020200029

    Moon, H., Huo, Y., Abramson, R. G., Peters, R. A., Assad, A., Moyo, T. K., Savona, M. R., & Landman, B. A. (2019). Acceleration of spleen segmentation with end-to-end deep learning method and automated pipeline. Computers in Biology and Medicine, 107, 109–117. https://doi.org/10.1016/j.compbiomed.2019.01.018

    Morey, J. R., Zhang, X., Yaeger, K. A., Fiano, E., Marayati, N. F., Kellner, C. P., De Leacy, R. A., Doshi, A., Tuhrim, S., & Fifi, J. T. (2021). Real-World Experience with Artificial Intelligence- Based Triage in Transferred Large Vessel Occlusion Stroke Patients. Cerebrovascular Diseases, 50(4), 450–455. https://doi. org/10.1159/000515320

    Murdoch, B. (2021). Privacy and artificial intelligence: challenges for protecting health information in a new era. BMC Medical Ethics, 22(1), 122. https://doi.org/10.1186/s12910-021-00687-3

    Murray, N. M., Unberath, M., Hager, G. D., & Hui, F. K. (2020). Artificial intelligence to diagnose ischemic stroke and identify large vessel occlusions: a systematic review. Journal of Neurointerventional Surgery, 12(2), 156–164. https://doi.org/10.1136/neurintsurg-2019-015135

    Nagendran, M., Chen, Y., Lovejoy, C. A., Gordon, A. C., Komorowski, M., Harvey, H., Topol, E. J., Ioannidis, J. P. A., Collins, G. S., & Maruthappu, M. (2020). Artificial intelligence versus clinicians: systematic review of design, reporting standards, and claims of deep learning studies. BMJ, 368. https://doi.org/10.1136/bmj.m689

    Nair, T., Precup, D., Arnold, D. L., & Arbel, T. (2020). Exploring uncertainty measures in deep networks for Multiple sclerosis lesion detection and segmentation. Medical Image Analysis, 59, 101557. https://doi.org/10.1016/j.media.2019.101557

    Nakao, T., Hanaoka, S., Nomura, Y., Sato, I., Nemoto, M., Miki, S., Maeda, E., Yoshikawa, T., Hayashi, N., & Abe, O. (2018). Deep neural network-based computer-assisted detection of cerebral aneurysms in MR angiography. Journal of Magnetic Resonance Imaging: JMRI, 47(4), 948–953. https://doi.org/10.1002/jmri.25842

    Nam, J. G., Kim, M., Park, J., Hwang, E. J., Lee, J. H., Hong, J. H., Goo, J. M., & Park, C. M. (2021). Development and validation of a deep learning algorithm detecting 10 common abnormalities on chest radiographs. The European Respiratory Journal: Official Journal of the European Society for Clinical Respiratory Physiology, 57(5). https://doi.org/10.1183/13993003.03061-2020

    Narayana, P. A., Coronado, I., Sujit, S. J., Wolinsky, J. S., Lublin, F. D., & Gabr, R. E. (2020). Deep Learning for Predicting Enhancing Lesions in Multiple Sclerosis from Noncontrast. MRI. Radiology, 294(2), 398–404. https://doi.org/10.1148/ radiol.2019191061

    National Institute for Health and Care Excellence (NICE). (n.d.). Evidence standards framework for digital health technologies. Retrieved June 10, 2022, from https://www.nice.org.uk/corporate/ecd7

    Neisius, U., El-Rewaidy, H., Nakamori, S., Rodriguez, J., Manning, W. J., & Nezafat, R. (2019). Radiomic Analysis of Myocardial Native T1 Imaging Discriminates Between Hypertensive Heart Disease and Hypertrophic Cardiomyopathy. JACC. Cardiovascular Imaging, 12(10), 1946–1954. https://doi. org/10.1016/j.jcmg.2018.11.024

    Nelson, A., Herron, D., Rees, G., & Nachev, P. (2019). Predicting scheduled hospital attendance with artificial intelligence. Npj Digital Medicine, 2(1), 26. https://doi.org/10.1038/s41746-019-0103-3

    Nielsen, A., Hansen, M. B., Tietze, A., & Mouridsen, K. (2018). Prediction of Tissue Outcome and Assessment of Treatment Effect in Acute Ischemic Stroke Using Deep Learning. Stroke; a Journal of Cerebral Circulation, STROKEAHA.117.019740. https://doi.org/10.1161/STROKEAHA.117.019740

    Norori, N., Hu, Q., Aellen, F. M., Faraci, F. D., & Tzovara, A. (2021). Addressing bias in big data and AI for health care: A call for open science. Patterns (New York, N.Y.), 2(10), 100347. https://doi.org/10.1016/j.patter.2021.100347

    O’Connor, S. D., & Bhalla, M. (2021). Should Artificial Intelligence Tell Radiologists Which Study to Read Next? [Review of Should Artificial Intelligence Tell Radiologists Which Study to Read Next?]. Radiology. Artificial Intelligence, 3(2), e210009. https://doi.org/10.1148/ryai.2021210009

    Office for Civil Rights (OCR). (2012, September 7). Guidance Regarding Methods for De-identification of Protected Health Information in Accordance with the Health Insurance Portability and Accountability Act (HIPAA) Privacy Rule. HHS.gov; US Department of Health and Human Services. https://www.hhs.gov/hipaa/for-professionals/privacy/special-topics/de- identification/index.html

    Oktay, O., Schlemper, J., Le Folgoc, L., Lee, M., Heinrich, M., Misawa, K., Mori, K., McDonagh, S., Hammerla, N. Y.,
    Kainz, B., Glocker, B., & Rueckert, D.
    (2018). Attention U-Net: Learning Where to Look for the Pancreas. In arXiv [cs.CV]. arXiv. http://arxiv.org/abs/1804.03999

    Olczak, J., Fahlberg, N., Maki, A., Razavian, A. S., Jilert, A., Stark, A., Sköldenberg, O., & Gordon, M. (2017). Artificial intelligence for analyzing orthopedic trauma radiographs. Acta Orthopaedica, 88(6), 581–586. https://doi.org/10.1080/1745367
    4.2017.1344459

    Olthof, A. W., van Ooijen, P. M. A., & Rezazade Mehrizi, M. H. (2020). Promises of artificial intelligence in neuroradiology: a systematic technographic review. Neuroradiology, 62(10), 1265–1278. https://doi.org/10.1007/s00234-020-02424-w

    Omoumi, P., Ducarouge, A., Tournier, A., Harvey, H., Kahn, C. E., Jr, Louvet-de Verchère, F., Pinto Dos Santos, D., Kober, T., & Richiardi, J. (2021). To buy or not to buy-evaluating commercial AI solutions in radiology (the ECLAIR guidelines). European Radiology, 31(6), 3786–3796. https://doi.org/10.1007/ s00330-020-07684-x

    O’Neill, T. J., Xi, Y., Stehel, E., Browning, T., Ng, Y. S., Baker, C., & Peshock, R. M. (2021). Active Reprioritization of the Reading Worklist Using Artificial Intelligence Has a Beneficial Effect on the Turnaround Time for Interpretation of Head CT with Intracranial Hemorrhage. Radiology. Artificial Intelligence, 3(2), e200024. https://doi.org/10.1148/ryai.2020200024

    Ooi, S. K. G., Makmur, A., Soon, A. Y. Q., Fook-Chong, S., Liew, C., Sia, S. Y., Ting, Y. H., & Lim, C. Y. (2021). Attitudes toward artificial intelligence in radiology with learner needs assessment within radiology residency programmes: a national multi-programme survey. Singapore Medical Journal, 62(3), 126–134. https://doi.org/10.11622/smedj.2019141

    Pan, Y., Shi, D., Wang, H., Chen, T., Cui, D., Cheng, X., & Lu, Y. (2020). Automatic opportunistic osteoporosis screening using low-dose chest computed tomography scans obtained for lung cancer screening. European Radiology, 30(7), 4107–4116. https://doi.org/10.1007/s00330-020-06679-y

    Park, H. J., Kim, S. M., La Yun, B., Jang, M., Kim, B., Jang, J. Y., Lee, J. Y., & Lee, S. H. (2019). A computer-aided diagnosis system using artificial intelligence for the diagnosis and characterization of breast masses on ultrasound: Added value for the inexperienced breast radiologist. Medicine, 98(3), e14146. https://doi.org/10.1097/MD.0000000000014146

    Price, I. I., & Nicholson, W. (2019). Medical AI and Contextual Bias. https://papers.ssrn.com/abstract=3347890

    Puvanasunthararajah, S., Fontanarosa, D., Wille, M.-L., & Camps, S. M. (2021). The application of metal artifact reduction methods on computed tomography scans for radiotherapy applications: A literature review. Journal of Applied Clinical Medical Physics / American College of Medical Physics, 22(6), 198–223. https://doi.org/10.1002/acm2.13255

    Qin, Z. Z., Sander, M. S., Rai, B., Titahong, C. N., Sudrungrot, S., Laah, S. N., Adhikari, L. M., Carter, E. J., Puri, L., Codlin, A. J., & Creswell, J. (2019). Using artificial intelligence to read chest radiographs for tuberculosis detection: A multi-site evaluation of the diagnostic accuracy of three deep learning systems. Scientific Reports, 9(1), 15000. https://doi.org/10.1038/ s41598-019-51503-3

    Ramspek, C. L., Jager, K. J., Dekker, F. W., Zoccali, C., & van Diepen, M. (2021). External validation of prognostic models: what, why, how, when and where?

    Rao, B., Zohrabian, V., Cedeno, P., Saha, A., Pahade, J., & Davis, M. A. (2021). Utility of Artificial Intelligence Tool as a Prospective Radiology Peer Reviewer - Detection of Unreported Intracranial Hemorrhage. Academic Radiology, 28(1), 85–93. https://doi.org/10.1016/j.acra.2020.01.035

    Reddy, S., Allan, S., Coghlan, S., & Cooper, P. (2020). A governance model for the application of AI in health care. Journal of the American Medical Informatics Association: JAMIA, 27(3), 491–497. https://doi.org/10.1093/jamia/ocz192

    Reddy, S., Rogers, W., Makinen, V.-P., Coiera, E., Brown, P., Wenzel, M., Weicken, E., Ansari, S., Mathur, P., Casey, A., & Kelly, B. (2021). Evaluation framework to guide implementation of AI systems into healthcare settings. BMJ Health & Care Informatics, 28(1). https://doi.org/10.1136/ bmjhci-2021-100444

    Reyes, M., Meier, R., Pereira, S., Silva, C. A., Dahlweid, F.-M., von Tengg-Kobligk, H., Summers, R. M., & Wiest, R. (2020). On the Interpretability of Artificial Intelligence in Radiology: Challenges and Opportunities. Radiology. Artificial Intelligence, 2(3), e190043. https://doi.org/10.1148/ryai.2020190043

    Rezazade Mehrizi, M. H., van Ooijen, P., & Homan, M. (2021). Applications of artificial intelligence (AI) in diagnostic radiology: a technography study. European Radiology, 31(4), 1805–1811. https://doi.org/10.1007/s00330-020-07230-9

    Richardson, J. P., Smith, C., Curtis, S., Watson, S., Zhu, X., Barry, B., & Sharp, R. R. (2021). Patient apprehensions about the use of artificial intelligence in healthcare. NPJ Digital Medicine, 4(1), 140. https://doi.org/10.1038/s41746-021-00509-1

    Richardson, M. L., Garwood, E. R., Lee, Y., Li, M. D., Lo, H. S., Nagaraju, A., Nguyen, X. V., Probyn, L., Rajiah, P., Sin, J., Wasnik, A. P., & Xu, K. (2021). Noninterpretive Uses of Artificial Intelligence in Radiology. Academic Radiology, 28(9), 1225– 1235. https://doi.org/10.1016/j.acra.2020.01.012

    Rockenbach, M. A. B. (2021, June 13). Multimodal AI in healthcare: Closing the gaps. CodeX. https://medium.com/codex/ multimodal-ai-in-healthcare-1f5152e83be2

    Rodríguez-Ruiz, A., Krupinski, E., Mordang, J.-J., Schilling, K., Heywang-Köbrunner, S. H., Sechopoulos, I., & Mann, R. M. (2019). Detection of Breast Cancer with Mammography: Effect of an Artificial Intelligence Support System. Radiology, 290(2), 305–314. https://doi.org/10.1148/radiol.2018181371

    Rodriguez-Ruiz, A., Lång, K., Gubern-Merida, A., Broeders, M., Gennaro, G., Clauser, P., Helbich, T. H., Chevalier, M., Tan, T., Mertelmeier, T., Wallis, M. G., Andersson, I., Zackrisson, S., Mann, R. M., & Sechopoulos, I. (2019). Stand- Alone Artificial Intelligence for Breast Cancer Detection in Mammography: Comparison With 101 Radiologists. Journal of the National Cancer Institute, 111(9), 916–922. https://doi.org/10.1093/jnci/djy222

    Santomartino, S. M., & Yi, P. H. (2022). Systematic Review of Radiologist and Medical Student Attitudes on the Role and Impact of AI in Radiology. Academic Radiology. https://doi.org/10.1016/j.acra.2021.12.032

    Schemmel, A., Lee, M., Hanley, T., Pooler, B. D., Kennedy, T., Field, A., Wiegmann, D., & Yu, J.-P. J. (2016). Radiology Workflow Disruptors: A Detailed Analysis. Journal of the American College of Radiology: JACR, 13(10), 1210–1214. https://doi.org/10.1016/j.jacr.2016.04.009

    Schreiber-Zinaman, J., & Rosenkrantz, A. B. (2017). Frequency and reasons for extra sequences in clinical abdominal MRI examinations. Abdominal Radiology (New York), 42(1), 306–311. https://doi.org/10.1007/s00261-016-0877-6

    Scott, I. A., Carter, S. M., & Coiera, E. (2021). Exploring stakeholder attitudes towards AI in clinical practice. BMJ Health & Care Informatics, 28(1). https://doi.org/10.1136/ bmjhci-2021-100450

    Seah, J. C. Y., Tang, C. H. M., Buchlak, Q. D., Holt, X. G., Wardman, J. B., Aimoldin, A., Esmaili, N., Ahmad, H., Pham, H., Lambert, J. F., Hachey, B., Hogg, S. J. F., Johnston, B. P., Bennett, C., Oakden-Rayner, L., Brotchie, P., & Jones, C. M. (2021). Effect of a comprehensive deep-learning model on the accuracy of chest x-ray interpretation by radiologists: a retrospective, multireader multicase study. The Lancet. Digital Health, 3(8), e496–e506. https://doi.org/10.1016/S2589- 7500(21)00106-0

    Sectra Amplifier Marketplace. (2021, July 5). Sectra Medical. https://medical.sectra.com/product/sectra-amplifier- marketplace/

    Sermesant, M., Delingette, H., Cochet, H., Jaïs, P., & Ayache, N. (2021). Applications of artificial intelligence in cardiovascular imaging. Nature Reviews. Cardiology, 18(8), 600–609. https://doi.org/10.1038/s41569-021-00527-2

    Setio, A. A. A., Traverso, A., de Bel, T., Berens, M. S. N., van den Bogaard, C., Cerello, P., Chen, H., Dou, Q., Fantacci, M. E., Geurts, B., Gugten, R. van der, Heng, P. A., Jansen, B., de Kaste, M. M. J., Kotov, V., Lin, J. Y.-H., Manders, J. T. M. C., Sóñora-Mengana, A., García-Naranjo, J. C., … Jacobs, C. (2017). Validation, comparison, and combination of algorithms for automatic detection of pulmonary nodules in computed tomography images: The LUNA16 challenge. Medical Image Analysis, 42, 1–13. https://doi.org/10.1016/j.media.2017.06.015

    Seyyed-Kalantari, L., Zhang, H., McDermott, M. B. A., Chen, I. Y., & Ghassemi, M. (2021). Underdiagnosis bias of artificial intelligence algorithms applied to chest radiographs in under- served patient populations. Nature Medicine, 27(12), 2176– 2182. https://doi.org/10.1038/s41591-021-01595-0

    Shan, H., Padole, A., Homayounieh, F., Kruger, U., Khera, R. D., Nitiwarangkul, C., Kalra, M. K., & Wang, G. (2019). Competitive performance of a modularized deep neural network compared to commercial algorithms for low-dose CT image reconstruction. Nature Machine Intelligence, 1(6), 269–276. https://doi.org/10.1038/s42256-019-0057-9

    Sharma, K., Rupprecht, C., Caroli, A., Aparicio, M. C., Remuzzi, A., Baust, M., & Navab, N. (2017). Automatic Segmentation of Kidneys using Deep Learning for Total Kidney Volume Quantification in Autosomal Dominant Polycystic Kidney Disease. Scientific Reports, 7(1), 2049. https://doi.org/10.1038/s41598-017-01779-0

    Shelmerdine, S. C., Arthurs, O. J., Denniston, A., & Sebire, N. J. (2021). Review of study reporting guidelines for clinical studies using artificial intelligence in healthcare. BMJ Health & Care Informatics, 28(1). https://doi.org/10.1136/ bmjhci-2021-100385

    Shinagare, A. B., Ip, I. K., Abbett, S. K., Hanson, R., Seltzer, S. E., & Khorasani, R. (2014). Inpatient imaging utilization: trends of the past decade. AJR. American Journal of Roentgenology, 202(3), W277–W283. https://doi.org/10.2214/AJR.13.10986

    Shlobin, N. A., Baig, A. A., Waqas, M., Patel, T. R., Dossani, R. H., Wilson, M., Cappuzzo, J. M., Siddiqui, A. H., Tutino, V. M., & Levy, E. I. (2022). Artificial Intelligence for Large-Vessel Occlusion Stroke: A Systematic Review. World Neurosurgery, 159, 207–220.e1. https://doi.org/10.1016/j.wneu.2021.12.004

    Silberg, J., & Manyika, J. (2019, June 6). Tackling bias in artificial intelligence (and in humans). McKinsey & Company. https://www.mckinsey.com/featured-insights/artificial- intelligence/tackling-bias-in-artificial-intelligence-and-in- humans

    Singh, S., Kalra, M. K., Hsieh, J., Licato, P. E., Do, S., Pien, H. H., & Blake, M. A. (2010). Abdominal CT: comparison of adaptive statistical iterative and filtered back projection reconstruction techniques. Radiology, 257(2), 373–383. https://doi.org/10.1148/radiol.10092212

    Smith-Bindman, R., Kwan, M. L., Marlow, E. C., Theis, M. K., Bolch, W., Cheng, S. Y., Bowles, E. J. A., Duncan, J. R.,
    Greenlee, R. T., Kushi, L. H., Pole, J. D., Rahm, A. K., Stout, N. K., Weinmann, S., & Miglioretti, D. L.
    (2019). Trends in Use of Medical Imaging in US Health Care Systems and in Ontario, Canada, 2000-2016. JAMA: The Journal of the American Medical Association, 322(9), 843–856. https://doi.org/10.1001/ jama.2019.11456

    Sutherland, G., Russell, N., Gibbard, R., & Dobrescu, A. (n.d.). The value of radiology, part II. https://car.ca/wp-content/ uploads/2019/07/value-of-radiology-part-2-en.pdf

    Tamada, D., Kromrey, M.-L., Ichikawa, S., Onishi, H., & Motosugi, U. (2020). Motion Artifact Reduction Using a Convolutional Neural Network for Dynamic Contrast Enhanced MR Imaging of the Liver. Magnetic Resonance in Medical Sciences: MRMS: An Official Journal of Japan Society of Magnetic Resonance in Medicine, 19(1), 64–76. https://doi.org/10.2463/ mrms.mp.2018-0156

    The Medical Futurist. (n.d.). The Medical Futurist. Retrieved February 23, 2022, from https://medicalfuturist.com/fda- approved-ai-based-algorithms/

    The Nuance AI Marketplace for Diagnostic Imaging. (n.d.). https://www.nuance.com/content/dam/nuance/en_us/collateral/ healthcare/data-sheet/ds-ai-marketplace-for-diagnostic- imaging-en-us.pdf

    Thodberg, H. H., Kreiborg, S., Juul, A., & Pedersen, K. D. (2009). The BoneXpert method for automated determination of skeletal maturity. IEEE Transactions on Medical Imaging, 28(1), 52–66. https://doi.org/10.1109/TMI.2008.926067

    Thomas, K. A., Kidziński, Ł., Halilaj, E., Fleming, S. L., Venkataraman, G. R., Oei, E. H. G., Gold, G. E., & Delp, S. L. (2020). Automated Classification of Radiographic Knee Osteoarthritis Severity Using Deep Neural Networks. Radiology. Artificial Intelligence, 2(2), e190065. https://doi.org/10.1148/ ryai.2020190065

    Towards trustable machine learning. (2018). Nature Biomedical Engineering, 2(10), 709–710. https://doi.org/10.1038/ s41551-018-0315-x

    Trinidad, M. G., Platt, J., & Kardia, S. L. R. (2020). The public’s comfort with sharing health data with third-party commercial companies. Humanities and Social Sciences Communications, 7(1), 1–10. https://doi.org/10.1057/s41599-020-00641-5

    Trivedi, H., Mesterhazy, J., Laguna, B., Vu, T., & Sohn, J. H. (2018). Automatic Determination of the Need for Intravenous Contrast in Musculoskeletal MRI Examinations Using IBM Watson’s Natural Language Processing Algorithm. Journal of Digital Imaging, 31(2), 245–251. https://doi.org/10.1007/ s10278-017-0021-3

    Tsao, D. N. (2020, July 27). AI in medical diagnostics 2020- 2030: Image recognition, players, clinical applications, forecasts: IDTechEx. https://www.idtechex.com/en/research-report/ai-in- medical-diagnostics-2020-2030-image-recognition-players- clinical-applications-forecasts/766

    Tucci, V., Saary, J., & Doyle, T. E. (2022). Factors influencing trust in medical artificial intelligence for healthcare professionals: a narrative review. Journal of Medical Artificial Intelligence, 5, 4–4. https://doi.org/10.21037/jmai-21-25

    Ueda, D., Yamamoto, A., Nishimori, M., Shimono, T., Doishita, S., Shimazaki, A., Katayama, Y., Fukumoto, S., Choppin, A., Shimahara, Y., & Miki, Y. (2019). Deep Learning for MR Angiography: Automated Detection of Cerebral Aneurysms. Radiology, 290(1), 187–194. https://doi.org/10.1148/ radiol.2018180901

    Urakawa, T., Tanaka, Y., Goto, S., Matsuzawa, H., Watanabe, K., & Endo, N. (2019). Detecting intertrochanteric hip fractures with orthopedist-level accuracy using a deep convolutional neural network. Skeletal Radiology, 48(2), 239–244. https://doi.org/10.1007/s00256-018-3016-3

    van Leeuwen, K. G., Schalekamp, S., Rutten, M. J. C. M., van Ginneken, B., & de Rooij, M. (2021). Artificial intelligence in radiology: 100 commercially available products and their scientific evidence. European Radiology, 31(6), 3797–3804. https://doi.org/10.1007/s00330-021-07892-z

    Vayena, E., & Blasimme, A. (2017). Biomedical Big Data: New Models of Control Over Access, Use and Governance. Journal of Bioethical Inquiry, 14(4), 501–513. https://doi.org/10.1007/ s11673-017-9809-6

    Vayena, E., Blasimme, A., & Cohen, I. G. (2018). Machine learning in medicine: Addressing ethical challenges. PLoS Medicine, 15(11), e1002689. https://doi.org/10.1371/journal. pmed.1002689

    Wang, J., Yang, F., Liu, W., Sun, J., Han, Y., Li, D., Gkoutos, G. V., Zhu, Y., & Chen, Y. (2020). Radiomic Analysis of Native T1 Mapping Images Discriminates Between MYH7 and MYBPC3- Related Hypertrophic Cardiomyopathy. Journal of Magnetic Resonance Imaging: JMRI, 52(6), 1714–1721. https://doi.org/10.1002/jmri.27209

    Wang, S.-H., Tang, C., Sun, J., Yang, J., Huang, C., Phillips, P., & Zhang, Y.-D. (2018). Multiple Sclerosis Identification by 14-Layer Convolutional Neural Network With Batch Normalization, Dropout, and Stochastic Pooling. Frontiers in Neuroscience, 12, 818. https://doi.org/10.3389/fnins.2018.00818

    Watanabe, A. T., Lim, V., Vu, H. X., Chim, R., Weise, E., Liu, J., Bradley, W. G., & Comstock, C. E. (2019). Improved
    Cancer Detection Using Artificial Intelligence: a Retrospective Evaluation of Missed Cancers on Mammography. Journal
    of Digital Imaging, 32
    (4), 625–637. https://doi.org/10.1007/ s10278-019-00192-5

    Weikert, T., Francone, M., Abbara, S., Baessler, B., Choi, B. W., Gutberlet, M., Hecht, E. M., Loewe, C., Mousseaux, E., Natale, L., Nikolaou, K., Ordovas, K. G., Peebles, C., Prieto, C., Salgado, R., Velthuis, B., Vliegenthart, R., Bremerich, J., & Leiner, T. (2021). Machine learning in cardiovascular radiology: ESCR position statement on design requirements, quality assessment, current applications, opportunities, and challenges. European Radiology, 31(6), 3909–3922. https://doi.org/10.1007/ s00330-020-07417-0

    Whittlestone, J., Nyrup, R., Alexandrova, A., Dihal, K., & Cave, S. (2019). Ethical and societal implications of algorithms, data, and artificial intelligence: a roadmap for research.
    London: Nuffield Foundation. https://www.nuffieldfoundation. org/sites/default/files/files/Ethical-and-Societal-Implications-of- Data-and-AI-report-Nuffield-Foundat.pdf

    WHO operational handbook on tuberculosis Module 2: Screening – Systematic screening for tuberculosis disease. (n.d.). Retrieved June 19, 2022, from https://www.who.int/ publications-detail-redirect/9789240022614

    Willemink, M. J., & Noël, P. B. (2019). The evolution of image reconstruction for CT-from filtered back projection to artificial intelligence. European Radiology, 29(5), 2185–2195. https://doi.org/10.1007/s00330-018-5810-7

    Winder, M., Owczarek, A. J., Chudek, J., Pilch-Kowalczyk, J., & Baron, J. (2021). Are We Overdoing It? Changes in Diagnostic Imaging Workload during the Years 2010-2020 including the Impact of the SARS-CoV-2 Pandemic. Healthcare (Basel, Switzerland), 9(11). https://doi.org/10.3390/healthcare9111557

    Wong, T. T., Kazam, J. K., & Rasiej, M. J. (2019). Effect of Analytics-Driven Worklists on Musculoskeletal MRI
    Interpretation Times in an Academic Setting. AJR. American Journal of Roentgenology, 1–5. https://doi.org/10.2214/ AJR.18.20434

    Wu, B., Zhou, Z., Wang, J., & Wang, Y. (2018). Joint learning for pulmonary nodule segmentation, attributes and malignancy prediction. 2018 IEEE 15th International Symposium on Biomedical Imaging (ISBI 2018), 1109–1113. https://doi.org/10.1109/ISBI.2018.8363765

    Wu, E., Wu, K., Daneshjou, R., Ouyang, D., Ho, D. E., & Zou, J. (2021). How medical AI devices are evaluated: limitations and recommendations from an analysis of FDA approvals. Nature Medicine, 27(4), 582–584. https://doi.org/10.1038/s41591-021-01312-x

    Wu, G.-G., Zhou, L.-Q., Xu, J.-W., Wang, J.-Y., Wei, Q., Deng, Y.-B., Cui, X.-W., & Dietrich, C. F. (2019). Artificial intelligence in breast ultrasound. World Journal of Radiology, 11(2), 19–26. https://doi.org/10.4329/wjr.v11.i2.19

    Yala, A., Schuster, T., Miles, R., Barzilay, R., & Lehman, C. (2019). A Deep Learning Model to Triage Screening Mammograms: A Simulation Study. Radiology, 293(1), 38–46. https://doi.org/10.1148/radiol.2019182908

    Yasaka, K., Akai, H., Kunimatsu, A., Abe, O., & Kiryu, S. (2018). Liver Fibrosis: Deep Convolutional Neural Network for Staging by Using Gadoxetic Acid-enhanced Hepatobiliary Phase MR Images. Radiology, 287(1), 146–155. https://doi.org/10.1148/ radiol.2017171928

    Yeung, K. (2018). A Study of the Implications of Advanced Digital Technologies (Including AI Systems) for the Concept of
    Responsibility Within a Human Rights Framework.
    https://papers. ssrn.com/abstract=3286027

    Yoo, Y., Tang, L. Y. W., Li, D. K. B., Metz, L., Kolind, S., Traboulsee, A. L., & Tam, R. C. (2019). Deep learning of brain lesion patterns and user-defined clinical and MRI features for predicting conversion to multiple sclerosis from clinically isolated syndrome. Computer Methods in Biomechanics and Biomedical Engineering: Imaging & Visualization, 7(3), 250–259. https://doi.org/10.1080/21681163.2017.1356750

    Yu, A. C., Mohajer, B., & Eng, J. (2022). External Validation of Deep Learning Algorithms for Radiologic Diagnosis: A Systematic Review. Radiology. Artificial Intelligence, 4(3), e210064. https://doi.org/10.1148/ryai.210064

    Yu, J.-P. J., Kansagra, A. P., & Mongan, J. (2014). The radiologist’s workflow environment: evaluation of disruptors and potential implications. Journal of the American College of Radiology: JACR, 11(6), 589–593. https://doi.org/10.1016/j. jacr.2013.12.026

    Yusuf, M., Atal, I., Li, J., Smith, P., Ravaud, P., Fergie, M., Callaghan, M., & Selfe, J. (2020). Reporting quality of studies using machine learning models for medical diagnosis: a systematic review. BMJ Open, 10(3), e034568. https://doi.org/10.1136/bmjopen-2019-034568

    Yu, Y., Xie, Y., Thamm, T., Gong, E., Ouyang, J., Christensen, S., Marks, M. P., Lansberg, M. G., Albers, G. W., & Zaharchuk,
    G.
    (2021). Tissue at Risk and Ischemic Core Estimation Using Deep Learning in Acute Stroke. AJNR. American Journal of Neuroradiology, 42(6), 1030–1037. https://doi.org/10.3174/ajnr. A7081

    Yu, Y., Xie, Y., Thamm, T., Gong, E., Ouyang, J., Huang, C., Christensen, S., Marks, M. P., Lansberg, M. G., Albers, G. W., & Zaharchuk, G. (2020). Use of Deep Learning to Predict Final Ischemic Stroke Lesions From Initial Magnetic Resonance Imaging. JAMA Network Open, 3(3), e200772. https://doi.org/10.1001/jamanetworkopen.2020.0772

    Zhang, Y., & Yu, H. (2018). Convolutional Neural Network Based Metal Artifact Reduction in X-Ray Computed Tomography. IEEE Transactions on Medical Imaging, 37(6), 1370–1381. https://doi.org/10.1109/TMI.2018.2823083

    Zhao, B., Liu, Z., Ding, S., Liu, G., Cao, C., & Wu, H. (2022). Motion artifact correction for MR images based on convolutional neural network. Optoelectronics Letters, 18(1), 54–58. https://doi.org/10.1007/s11801-022-1084-z

    Zhao, J., Huang, Y., Song, Y., Xie, D., Hu, M., Qiu, H., & Chu, J. (2020). Diagnostic accuracy and potential covariates for machine learning to identify IDH mutations in glioma patients: evidence from a meta-analysis. European Radiology, 30(8), 4664–4674. https://doi.org/10.1007/s00330-020-06717-9

    Zhou, C., Ding, C., Wang, X., Lu, Z., & Tao, D. (2020). One-pass Multi-task Networks with Cross-task Guided Attention for Brain Tumor Segmentation. IEEE Transactions on Image Processing: A Publication of the IEEE Signal Processing Society. https://doi.org/10.1109/TIP.2020.2973510

    Zicari, R. V., Brodersen, J., Brusseau, J., Düdder, B., Eichhorn, T., Ivanov, T., Kararigas, G., Kringen, P., McCullough, M., Möslein, F., Mushtaq, N., Roig, G., Stürtz, N., Tolle, K., Tithi, J. J., van Halem, I., & Westerlund, M. (2021). Z-Inspection®: A Process to Assess Trustworthy AI. IEEE Transactions on Technology and Society, 1–1. https://doi.org/10.1109/TTS.2021.3066209