Morphological Recognition Method of S-T Segment Based on Interval Fuzzy Localization and Feature Comparison

Sudden heart disease is one of the diseases with fast onset and high mortality rate. Sudden cardiac disease is highly correlated with the S-T segment morphology of ECG signals. In this paper, a morphological recognition algorithm of S-T segment is proposed. Firstly, the wavelet transform is used to denoise the ECG signal waveform, and the information features of the R wave, S wave and T wave of the ECG signal are obtained by extraction methods. Then, between the S wave trough and the T wave peak, the interval fuzzy positioning method is used to obtain the approximate position of the medically defined ST segment, and the ST segment morphological identification features are obtained by comparing the fuzzy ST segment with the original ECG signal: ST segment offset level, ST segment assistance,discriminant line slope G,ST segment signals and the discriminant line Euclidean distance sum. The simulation experiments were carried out using ECG signal samples from the Massachusetts Institute of Technology ECG Signal Database (MIT-BIH). Experiments show that the proposed method has an average accuracy of 94.3% in S-T segment morphological recognition.So it has high recognition performance for the S-T segment of arrhythmia ECG signals.