Advance QRS Detection Technique Incorporating Time and Amplitude Thresholds with Statistical False Peak Elimination

Abstract

In order to improve peak detection effectiveness, this work offers a brand-new peak identification technique that can reduce noise and adjust to variations in ECG signal shape. Segmentation, time and amplitude thresholds, statistical false peak elimination, median and moving average (MA) filtering, and statistical false peak elimination are the pillars upon which the suggested technique rests. An extended median filter is also used to minimize noise at an even deeper level during preprocessing, where the filters are initially employed to reduce undesired noise and interference. Using a time axis (x-axis) and an amplitude (y-axis) threshold, each segment of the data is analyzed after it has been separated into smaller sections. Next, the erroneous peaks caused by any leftover noise are removed using the average peak-to-peak interval. Any peak that is identified twice is removed, and a post- processing stage is added to check for low-amplitude peaks that were missed. The suggested methodology outperforms a number of state-of-the-art approaches in the field when tested utilizing the MIT-BIH arrhythmia and Fantasia data bases.