ECG signal classification method based on improved BP neural network

Abstract

Accurate identification of ECG signals is the key to intelligent diagnosis of ECG monitoring systems. In order to improve the classification accuracy of ECG signals, an improved ECG signal classification algorithm based on BP neural network was studied. Firstly, statistical analysis was performed on the MIT-BIH Arrhythmia Database sample experts. The normal heart beat, ventricular premature beat, left bundle branch block heart beat and right bundle branch block heart beat were selected as neural network recognition targets, and extracted by principal component analysis. 25 heart beat features are used as sample vectors. The simulation results show that the improved BP neural network has better classification and recognition ability, and the accuracy of the whole sample classification is 98.4%. The algorithm has fast convergence speed and high classification accuracy, which is helpful for detecting and diagnosing heart diseases.

Keywords

• ECG signal
• improved BP neural network
• principal component analysis
• intelligent diagnosis