Superiority analysis of MLMS over adaptive filtering methods for hearth arrhythmias detection

This paper shows a comparative analysis among adaptive filtering methods in case of QRS detection from ECG signal. It is found that modified LMS (Least Mean Squares algorithm) method inherits robustness and provides a high degree of detection performance even in cases of very noisy electrocardiographic signals. The excellent performance of modified LMS algorithm is confirmed by a sensitivity of 99.93% (75 false negatives) and a positive predictivity of 99.92% (82 false positives) against the MIT-BIH arrhythmia database.