Automatic detection and localisation of myocardial infarction using multi-channel dense attention neural network

Yupeng Qiang, Xunde Dong, Yang Yang

Abstract: Highlights•MCDANN: a multi-channel dense attention neural network for MI detection and localisation using 12-lead ECG signals, without noise pre-processing.•MCADNN model with DACB captures discriminative & emphasises critical features for distinguishing different types of MI.•MCDANN model significantly improves inter-patient MI localisation performance, making it more feasible for clinical applications.