From:  A systematic review on cardiovascular risk stratification in the artificial intelligence paradigm using radiogenomics

 Examples of AI in CVD research.

AuthorYearDiseasesModelResultsSummary
Lekadir et al. [56]2016Carotid atherosclerosisCNNAcc: 75%To characterize carotid plaque composition
Attia et al. [57] 2019ALVDCNNAUC: 0.93To analyze smartwatch-recorded ECG signals and predict LV dysfunction
Heo et al. [78]2019Ischemic strokeDNN, RF, LRAUC(DNN): 0.888
AUC(RF): 0.857
AUC(LR): 0.849		To predict long-term functional outcomes at 3 months in patients with ischemic stroke patients
Attia et al. [58]2022Cardiac dysfunctionAI algorithm (unspecified DL model)AUC: 0.885To identify patients with cardiac dysfunction
McGilvray et al. [62]2022Heart failure (HF)Ensemble DL modelAUC: 0.91To identify HF medical therapy non-responders
Wang et al. [59]2024*11 CVD typesVideo Swin TransformerAUC: 0.988To develop a video-based DL approach for automatic screening and diagnosis of CVDs using CMR
Saikumar et al. [60]2024CADRCNNAcc: 99.173%To detect CAD on radiology datasets
Hasan et al. [61]2025AHB, MI, HMI, NHBCNNAcc: 99.29%To predict four cardiac conditions using ECG images
Return to article view

*11 CVD types: hypertrophic cardiomyopathy (HCM), dilated cardiomyopathy (DCM), coronary artery disease (CAD), left ventricular noncompaction (LVNC), restrictive cardiomyopathy (RCM), cardiac amyloidosis (CAM), hypertensive heart disease (HHD), myocarditis, arrhythmogenic right ventricular cardiomyopathy (ARVC), pulmonary arterial hypertension (PAH), and Ebstein’s anomaly. Acc: accuracy; AHB: abnormal heartbeat; ALVD: asymptomatic left ventricular dysfunction; AUC: area under the receiver operating characteristic curve; CAD: coronary artery disease; CMR: cardiovascular magnetic resonance; CVD: cardiovascular disease; DL: deep learning; DNN: deep neural network; ECG: electrocardiogram; HMI: history of myocardial infarction; LR: logistic regression; MI: myocardial infarction; NHB: normal heartbeat; RCNN: region-based convolutional neural network; RF: random forest.