The Pretreatment and Feature Data Extraction of ECG Based on Matlab

The basis of ECG analysis and diagnosis is detection of reliable QRS wave group. In this paper, filtered the EMG interference and frequency interference in the ECG, and eliminated noise using wavelet threshold denoising method, and distinguished the QRS wave group of pretreated ECG, Extracted heart rate, R-R interval and QRS interval data. Using the standard MIT-BIH arrhythmia database for algorithm verification, the results show that the denoising algorithm could effectively eliminate noise, and feature points recognition accuracy achieved 99.8%. More important, extracted feature data provided a reliable data source for the automatic diagnosis of ECG.