An Adaptive Median Filter Based on Sampling Rate for R-Peak Detection and Major-Arrhythmia Analysis.

Area of Science:

• Biomedical Engineering
• Signal Processing
• Medical Device Technology

Background:

• The proliferation of Internet of Medical Things (IoMT) devices necessitates robust vital sign monitoring.
• Portable electrocardiogram (ECG) devices are rapidly advancing, but varying sampling rates pose a challenge for accurate analysis.
• Developing adaptive algorithms is crucial for ECG analysis across diverse hardware capabilities.

Purpose of the Study:

• To propose an R-point detection method adaptable to different sampling rates in ECG signals.
• To develop a technique for analyzing major arrhythmias based on signal characteristics derived from the detected R-point.
• To validate the proposed method's effectiveness using diverse ECG datasets.

Main Methods:

• An adaptive median filter approach was developed, with filter size dynamically adjusted based on the ECG's sampling rate.
• A sliding window technique was employed, applying a wider median filter to high-variance QRS complex sections.
• R-point detection was achieved by subtracting the filtered signal from the original ECG signal, followed by arrhythmia analysis.

Main Results:

• The proposed R-point detection method demonstrated effectiveness across various sampling rates.
• The technique successfully identified major arrhythmias using the characteristics of the detected R-points.
• Simulations using diverse databases (MIT-BIH arrhythmia, atrial fibrillation) and real-world signals confirmed the method's efficacy.

Conclusions:

• The adaptive median filter-based R-point detection method is a reliable approach for ECG analysis, particularly for portable devices.
• The proposed arrhythmia analysis technique, leveraging the detected R-point, offers a valuable tool for clinical diagnostics.
• This method addresses the challenge of varying sampling rates, improving the versatility of ECG analysis in medical IoT applications.