ECG signal classification based on BPNN

摘要

This paper focus on the ECG signal classification based on the BP Neuron Network. The AR coefficients and relative errors were used to represent the ECG segments in current research. The data in the paper obtained from MIT-BIH database. It included Normal Sinus Rhythm (NSR), premature ventricular contraction (PVC), Ventricular Tachycardia (VT), and Ventricular Fibrillation (VF). The back propagation neural network (BPNN) was utilized to classify the classes. The training and testing data was 100 and 100 samples,. The results show that overall accuracy is 95.72%–97.36%.