The segmented-beat modulation method for ECG estimation.

Area of Science:

• Biomedical Engineering
• Signal Processing
• Cardiology

Background:

• Electrocardiographic (ECG) tracings corrupted by noise present significant challenges for accurate signal estimation.
• Traditional linear filtering methods are often inadequate for removing noise within the ECG's frequency band.
• Existing template-based ECG estimation techniques struggle to replicate physiological heart rate and morphological variability.

Purpose of the Study:

• To introduce the segmented-beat modulation method (SBMM) for improved ECG signal estimation.
• To address the limitations of existing template-based methods in reproducing heart rate and morphological variability.
• To provide a robust technique for reconstructing clean ECG signals from noisy recordings.

Main Methods:

• The proposed SBMM segments ECG beats into QRS and TUP components.
• A modulation/demodulation process is applied to the TUP segment before beat concatenation.
• This process adjusts the duration and morphology of the estimated beat to match the original beat.

Main Results:

• The SBMM demonstrated accurate R peak location estimation with no errors.
• Low segment errors were observed: ≤65 μV for QRS and ≤30 μV for TUP.
• TUP segment errors showed a positive correlation with increasing heart rate variability (r=0.59, P<10⁻²).

Conclusions:

• The SBMM is a valuable tool for generating high-quality ECG estimations.
• The method effectively handles ECG tracings with heart rate and morphological variability.
• SBMM offers an improvement over traditional template-based techniques for noisy ECG signals.