Towards Understanding ECG Rhythm Classification Using Convolutional Neural Networks and Attention Mappings
Published in
Machine Learning in Health Care(2018)External Links:
TL
TL;DR
A deep convolutional neural network was trained to classify single lead ECG waveforms as either Normal Sinus Rhythm, Atrial Fibrillation, or Other Rhythm, and class activation mappings were extracted to help better understand which areas of the waveform the model was focusing on when making a classification.