TABLE 3.
Overview of the use of AI in cardiology.
| Application/Task | Model | Data | Training set | Testing set | Accuracy/specificity/ sensitivity |
References |
| Identification of arrythmia (SR, LBBB, RBBB, PVC, PAC) | SVM | ECG | - | - | 100% (SR), 98.66% (LBBB), 100% (RBBB), 99.66% (PVC), and 100% (PAC) accuracy | (15) |
| Discrimination of HCM from ATH | Ensemble ML (SVM, RF, ANNs) | Clinical, echocardiography | - | - | 87% sensitivity and 82% specificity | (32) |
| Prediction of ACM in patients with suspected CAD undergoing CTA | Boosted ensemble algorithm | Clinical, CTA | - | 10,030 subjects | AUC 0.79 | (40) |
| Arrythmia detection (prediction of 12 types of arrythmia compared to cardiologist) | DNN | ECG | - | - | 99% accuracy | (13) |
| Detection of subclinical AF | CNN | ECG | 454,789 images | 130,801 images | AUC 0.90, sensitivity 82.3%, specificity 83.4%, accuracy 83.3% | (17) |
| Identification of ventricular dysfunction (EF 35%) | CNN | ECG, echocardiography | 44,959 subjects | 52,870 subjects | AUC 0.93, sensitivity 86.3%, specificity 85.7%, and accuracy 85.7% | (52) |
| Phenogroup HF patients and identification of responders to CRT implantation | Unsupervised ML (Multiple Kernel Learning and K-means clustering) | Clinical, echocardiography | - | 1,106 subjects | - | (60) |
| Automated analysis of cardiac structure and function (left ventricular chamber volumes, mass and EF) | CNN | CMR | 599 subjects | 110 subjects | - | (47) |
| Prediction of CAD on CTA | Boosted ensemble algorithm | Clinical, CTA (CACS) | - | 13,054 subjects | AUC 0.881 | (38) |
| Prediction of ACM for 1, 2-, 3-, 4-, and 5-years post CRT implantation | RF | Clinical, ECG, echocardiography | 2,282 subjects | 1,510 subjects | AUC 0.768 (1 year), 0.793 (2 years), 0.785 (3 years), 0.776 (4 years), 0.803 (5 years) | (59) |
| Prediction of early coronary revascularisation within 90 days after SPECT MPI | Ensemble LogitBoost algorithm | Clinical, SPECT | - | 1,980 subjects | AUC 0.81 | (44) |
| Identification of patients with PAH | Tensor based ML algorithm (multilinear subspace learning) | CMR | 200 subjects | 1,122 subjects | AUC 0.92 | (48) |
ACM, all-cause mortality; AF, atrial fibrillation; ANNs, artificial neural networks, ATH, athlete’s heart; AUC, area under the curve [integral of the ROC (receiver operator characteristic) curve]; CACS, coronary artery calcium score; CAD, coronary artery disease; CMR, cardiovascular magnetic resonance imaging; CNN, convolutional neural network; CRT, cardiac resynchronisation therapy; CTA, Cardiac Computed Tomography Angiography; DNN, deep neural network; ECG, electrocardiogram; EF, ejection fraction; HCM, hypertrophic cardiomyopathy; HF, heart failure; LBBB, left bundle branch block; ML, machine learning; MPI, myocardial perfusion imaging; PAC, premature atrial contraction; PAH, pulmonary arterial hypertension; PVC, premature ventricular contraction; RBBB, right bundle branch block; RF, random forest; SPECT, single-proton emission computerised tomography; SR, sinus rhythm; SVM, support vector machine.