A statistical approach for determination of time plane features from digitized ECG

Area of Science:

• Biomedical Engineering
• Signal Processing
• Cardiology

Background:

• Electrocardiogram (ECG) analysis is crucial for diagnosing cardiac conditions.
• Accurate feature extraction from ECG signals is essential for reliable interpretation.
• Existing methods may face challenges with baseline wander and precise wave delineation.

Purpose of the Study:

• To develop and validate a novel statistical method for time-plane feature extraction from digitized ECG signals.
• To accurately detect the position and magnitude of QRS complex, P wave, and T wave in single-lead ECG data.
• To assess the algorithm's performance in R-peak detection and baseline modulation removal.

Main Methods:

• A statistical approach is employed for feature extraction from ECG samples.
• The algorithm utilizes relative comparison of magnitude and slopes of ECG signal segments.
• Baseline modulation is identified and removed from the ECG dataset.

Main Results:

• The algorithm successfully detects the position and magnitude of key ECG waves (QRS, P, T).
• R-peak detection and baseline modulation removal were tested on the MIT-BIH arrhythmia and PTBDB databases.
• The implemented method achieved an overall accuracy exceeding 99%.

Conclusions:

• The proposed statistical method offers a highly accurate approach for ECG feature extraction.
• The algorithm demonstrates robust performance in identifying critical waveform components and handling signal artifacts.
• This technique holds promise for enhancing automated ECG analysis and diagnostic tools.