Compressed Segmented Beat Modulation Method using Discrete Cosine Transform^.

Area of Science:

• Biomedical Engineering
• Signal Processing
• Cardiology

Background:

• Current 24-hour electrocardiogram (ECG) monitors may miss infrequent arrhythmias.
• Online ECG processing and wearable sensors are increasingly used for long-term monitoring.
• Large data storage and transmission are needed for continuous cloud-based ECG analysis.

Purpose of the Study:

• To propose and evaluate a novel compression method for ambulatory ECG monitoring.
• To address the storage and transmission challenges of continuous ECG data.
• To assess the effectiveness of the proposed method in preserving signal quality and energy.

Main Methods:

• A discrete cosine transform (DCT) compressed segmented beat modulation method (SBMM) was developed.
• The method was tested using the MIT-BIH ECG Compression Test Database.
• Performance was evaluated using signal-to-noise ratio (SNR) and compression ratio (CR) at varying signal energy levels.

Main Results:

• An average SNR of 4.56 dB was achieved, with a 1.68 dB decline compared to uncompressed signals.
• 95% of signal energy was preserved.
• Quantization at 6 bits (CR=2) resulted in a 50% storage reduction compared to original 12-bit storage.

Conclusions:

• The proposed DCT-compressed SBMM effectively reduces ECG data storage size.
• The method maintains high signal energy and acceptable SNR for ambulatory monitoring.
• This technique offers a viable solution for efficient long-term ECG data management.