Arrhythmia Classification Method Based on SECNN-LSTM

Abstract

Abstract In order to improve the recognition and prediction accuracy of automatic classification of ECG signals, this paper proposes an arrhythmia classification method based on SECNN-LSTM. First, ECG signal is preprocessed, and the data is resampled for the problem of data imbalance, then the SECNN-LSTM network model is built. The spatial features of the signal are extracted by SECNN model and the front and back dependencies of the feature information are captured by LSTM model. The method has been tested and verified on the MIT-BIH Arrhythmia Database, and compared with other traditional arrhythmia classification method. The accuracy, precision, sensitivity, specificity and F1 value of the model have been improved to different degrees, and the average accuracy of the model has reached 98.70%. The experimental results show that this method can efficiently and accurately identify normal beats and four common types of arrhythmia diseases.