QRS Peak Detection Using Statistical False Peak Elimination and Derivative Search Back

ECG data must be used to accurately diagnose heart attacks, also known as myocardial infarctions. Noise makes it difficult to identify R-peaks, which are important indicators of cardiac arrhythmias. The technique statistically removes false peaks and recurring peaks, and then segments the data according to time axis requirements. The stages of implementation include segmenting data, removing spurious peaks, and applying filters to eliminate baseline drift and noise. For precise QRS peak identification in ECG signals, a MATLAB-based derivative search-back method was created. It guarantees strong performance in both arrhythmic and normal circumstances, exhibiting resilience and dependability across a range of ECG signal scenarios. It was tested on the MIT-BIH Arrhythmia and Fantasia Databases. The outcomes show the effectiveness of the approach for both raw and pre-processed signals as well as identified peaks. Positive predictivity, sensitivity, and detection error rate are used in MATLAB to assess the efficacy of the QRS peak detection algorithm across the MIT-BIH Arrhythmia and Fantasia databases. For MIT-BIH Arrhythmia Database’s key performance metrics include a sensitivity of 99.82%, positive predictivity of 99.93%, and detection error rate of 0.23%, and for Fantasia Database the values are 99.94%, 99.92%, and 0.12% respectively. These findings demonstrate the effectiveness of the QRS peak identification technique in several databases.