ECG compression by modelling the instantaneous module/phase of its DCT

Area of Science:

• Signal Processing
• Biomedical Engineering
• Time-Frequency Analysis

Background:

• Non-linear and non-stationary data, like electrocardiograms (ECG), present challenges for traditional compression methods.
• Time-frequency analysis is crucial for understanding complex biomedical signals.
• Existing compression techniques may not adequately capture the dynamic nature of ECG signals.

Purpose of the Study:

• To develop and evaluate a novel parametric modeling technique for ECG compression.
• To improve the representation and reconstruction of individual ECG beats.
• To assess the performance of the proposed method using a standard ECG database.

Main Methods:

• Parametric modeling of instantaneous module and phase from Discrete Cosine Transform (DCT) of ECG beats.
• Direct estimation of model parameters from the DCT coefficients.
• Reconstruction of ECG beats using the estimated parameters.

Main Results:

• The proposed parametric modeling technique enables effective reconstruction of ECG beats.
• The method demonstrates potential for compressing non-linear and non-stationary ECG data.
• Performance evaluation using the MIT-BIH arrhythmia database confirmed the technique's viability.

Conclusions:

• Parametric modeling of instantaneous module and phase offers a promising approach for ECG compression.
• The DCT-based method provides accurate beat reconstruction for time-frequency analysis.
• This technique contributes to advancements in biomedical signal processing and compression.