BME Frontiers / 2022 / Article / Tab 3

Research Article

Automatic Detection of Atrial Fibrillation from Single-Lead ECG Using Deep Learning of the Cardiac Cycle

Table 3

Comparison of performance of different AF detection algorithms evaluated on the AFDB data. For our approaches, average -score, sensitivity, specificity, PPV, area under the ROC curve (AUC), and accuracy (ACC) of 10-fold cross-validation are presented with the standard deviation. 30 heartbeat windows are considered an input. Each window is labeled according to the majority of the heartbeat annotations of the window. values marked with are estimated from the corresponding sensitivity and PPV reported in the literature.

MethodSensitivitySpecificityPPVAUCACC

DHB+RNN
P-DHB+RNN
S-DHB+RNN
DHB+avg
P-DHB+avg
S-DHB+avg
Asgari et al. [27]97.097.1
Islam et al. [23]94.995.895.594.195.6
Andersen et al. [25]97.3