An improved method for ECG signal feature point detection based on wavelet transform

The electrocardiogram (ECG) signals not only provide effective diagnosis support information for physicians, but also receive growing attention in areas of human identity recognition as a new kind of biomedical signals. An improved method for ECG feature point detection based on wavelet transform has been proposed in this paper. First, the ECG signal is pre-processed to remove noises and base-line wander. Secondly, QRS complex is located by the improved method based on wavelet transform. Finally, P-wave and T-wave in ECG signal are determined through window search in the predefined range. In comparison, QRS complex is also located through differential threshold method. The proposed method has been tested on the MIT-BIH arrhythmia database. Experimental results show that QRS complex can be detected at the accuracy of up to 99% and the accuracy of the awkward P-wave and T-wave location can be 95% or more. The proposed method is effective and lay foundation for human identification using ECG signal.