Multi-feature Fusion ECG Signal Recognition Algorithm Based on VMD

Abstract

In order to improve the recognition accuracy of ECG signals, a multi-feature fusion ECG signal recognition algorithm based on Variational Mode Decomposition (VMD) and Convolutional Neural Networks (CNN) is proposed. Firstly, the VMD algorithm is used to decompose the ECG signal. Secondly, CNN is used to extract the overall feature information of the signal on the original ECG signal. At the same time, the detailed feature information is extracted on the IMF (Intrinsic Mode Function, IMF) centered on different frequencies. Finally, the extracted multiple feature information is weighted and fused to complete signal recognition. The proposed multi-feature fusion recognition signal algorithm based on VMD enhances the ability of time domain feature extraction and improves the accuracy of signal recognition. The recognition accuracy of MIT-BIH arrhythmia signal is 99.46%, which is higher than that of other algorithms. It proves the effectiveness of the proposed algorithm for arrhythmia signal recognition.