A Low Power Cardiovascular Healthcare System With Cross-Layer Optimization From Sensing Patch to Cloud Platform

Area of Science:

• Biomedical Engineering
• Computer Science
• Cardiovascular Medicine

Background:

• Cardiovascular diseases remain a leading cause of mortality, necessitating advanced remote monitoring solutions.
• Existing systems often face challenges in power consumption, data processing, and signal quality.

Purpose of the Study:

• To develop and evaluate an Internet of Medical Things (IoMT)-based cardiovascular healthcare system.
• To achieve cross-layer optimization from a wearable sensing patch to a cloud platform for improved performance and power efficiency.

Main Methods:

• Designed a wearable ECG patch with a custom System-on-Chip (SoC) for low power consumption and embedded signal processing.
• Implemented a "wearable patch-mobile-cloud" hybrid computing framework with cross-layer optimization.
• Utilized on-patch data compression and wireless connectivity to mobile devices and the cloud.

Main Results:

• Achieved an on-patch ECG signal compression ratio of 12.07 with 2.29% root mean square variation.
• Demonstrated an average signal-to-noise ratio improvement of 12.63 dB and mean square error reduction of 94.47% using the MIT-BIH database.
• Attained an average accuracy of 97% for cardiovascular disease prediction on the cloud platform.

Conclusions:

• The proposed IoMT system offers an efficient and effective solution for remote cardiovascular monitoring.
• Cross-layer optimization significantly enhances performance-power trade-offs in wearable healthcare devices.
• The system demonstrates high accuracy in disease prediction, paving the way for improved patient outcomes.