Multi-Signal Acquisition System for Continuous Blood Pressure Monitoring.

Area of Science:

• Biomedical Engineering
• Cardiovascular Physiology
• Signal Processing

Background:

• Continuous blood pressure (BP) monitoring is crucial for detecting and preventing cardiovascular diseases like hypertension.
• Integrating multiple physiological signals (ECG, PPG, ICG) can improve BP estimation accuracy, but system complexity and feature extraction are challenges.
• Existing research on multi-signal integration for BP monitoring is limited and technologies are in early development.

Purpose of the Study:

• To develop a compact and efficient BP monitoring system capable of synchronously acquiring and processing ECG, PPG, and ICG signals.
• To evaluate the accuracy of BP estimation using the developed multi-signal system.
• To identify optimal features and models for enhancing BP prediction accuracy.

Main Methods:

• Developed a BP monitoring system with optimized circuit design allowing shared electrodes for ECG and ICG modules, reducing complexity.
• Synchronously acquired ECG, PPG, and ICG signals from 40 healthy subjects (400 min total data, 4390 records).
• Employed pulse wave analysis features combined with the XGBoost model for BP estimation.

Main Results:

• The integrated system achieved high accuracy in BP estimation: mean absolute error of 3.76 ± 3.98 mmHg for systolic BP and 2.71 ± 2.57 mmHg for diastolic BP.
• BP estimation performance met Grade A standards under the British Hypertension Society (BHS) guidelines.
• Results are comparable to or surpass existing multi-signal BP monitoring studies.

Conclusions:

• The proposed system offers an efficient and practical solution for continuous BP monitoring.
• Combining pulse wave analysis features with the XGBoost model significantly enhances BP prediction accuracy.
• The multi-signal approach demonstrates potential for improved cardiovascular disease management.