CNN Learning Based Approach for Cardiac Arrhythmia and Congestive Heart Failure Detection

An electrocardiogram (ECG) pattern classification method has been proposed to distinguish heart conditions such as arrhythmia (ARR) and congestive heart failure (CHF) from normal sinus rhythms (NSR) using deep convolutional neural networks (CNNs) by converting the ECG signals into RGB images. The results demonstrate an increase in diagnostic accuracy from 90.63% to 94.12% using a pretrained CNN model by utilising additional data from the second lead of the ECG.