A low complexity on-chip ECG data compression methodology targeting remote health-care applications

Area of Science:

• Biomedical Engineering
• Signal Processing
• Integrated Circuit Design

Background:

• Remote healthcare applications require efficient data handling for electrocardiogram (ECG) signals.
• Existing on-chip ECG data compression methods often compromise reliability or complexity.
• There is a need for robust, low-complexity solutions for real-time ECG processing in wearable devices.

Purpose of the Study:

• To develop and validate a novel, low-complexity on-chip ECG data compression methodology.
• To enable reliable and efficient remote healthcare applications through advanced data compression.
• To assess the diagnostic accuracy of the reconstructed ECG data post-compression.

Main Methods:

• Implementation of a novel compression algorithm on an Application Specific Integrated Circuit (ASIC) platform (UMC 130 nm technology).
• System operates at 1 MHz with a 1.62 V supply voltage and a 16-bit system word-length.
• Validation of reconstruction fidelity using established (MIT-BIH, PTB-DB) and institutional (IITH-DB) ECG databases.

Main Results:

• Achieved an average compression ratio of approximately 90%.
• Demonstrated high reconstruction accuracy with R(2) statistics > 83%, Cross Correlation > 98%, and Regression > 99%.
• The methodology allows for storing ~47 hours of ECG data on-chip, a significant improvement over the state-of-the-art (5 hours).

Conclusions:

• The proposed on-chip ECG compression methodology offers significant data reduction with minimal loss of diagnostic information.
• This technology is suitable for resource-constrained remote healthcare systems, enhancing data storage and transmission efficiency.
• The reliable data reconstruction ensures diagnostic accuracy, paving the way for improved remote patient monitoring and telehealth services.