A Study Of Adaptive Interference Cancellation And Waveform Defection Of ECG Signal

As an important biological signal, ECG signal provides a valuable basis for clinical diagnosis and treatment of related diseases. However, its reference significance is based on the effective acquisition and correct recognition of ECG. In fact, this mv-level weak signal can be easily affected by various interferences caused by the power of magnetic field, patient respiratory motion or contraction and so on from sampling terminal to the receiving and display end. The interference will affect the quality of ECG waveform, leading to false detection of wave group, and thus causes misdiagnosis or faulty treatment. Therefore, the elimination of the interference and subsequent wave group identification technology has been a hot topic all over the world, and the study has important significance.Based on the above discussion, the article takes the following research and verification mainly from two aspects:The first part: the adaptive elimination of interference on ECG signals. The symbolic function and block processing concept are introduced based on the classical LMS algorithm, thus we get the normalized LMS algorithm based on sign function and the normalized block processing LMS algorithm based on sign function. The improved adaptive algorithms are applied to eliminate the ECG signal with power line interference and baseline wander. Effective simulation and detailed results are given both using the MIT-BIH database data and the ECG data from a portable ECG acquisition instrument. The results show that the improved adaptive algorithms are better than the traditional LMS algorithm in computational complexity, convergence feature and signal SNR improvement.The second part: the detection of QRS wave group of ECG signal. A group of filters based on biorthogonal spline wavelet are derived from the ideal two channel filter bank structure foundation. The identification algorithm based on biorthogonal spline wavelet transform is applied to the ECG data after interference cancellation in chapter 4 to detect the R wave, the start and the end point of QRS. To measure the overall accuracy of the algorithm, 5 groups of ECG data from MIT-BIH arrhythmia DB are further used to detect the R peak point. Statistical results show that the detection accuracy of this algorithm is as high as 97.8%.