Cuff-less Blood Pressure Measurement Using Supplementary ECG and PPG Features Extracted Through Wavelet...

Area of Science:

• Biomedical Engineering
• Cardiovascular Monitoring
• Signal Processing

Background:

• Continuous blood pressure (BP) monitoring is crucial for managing hypertension but is hindered by the discomfort of traditional cuff-based methods.
• Existing cuff-less approaches, like pulse transit time (PTT), often rely on single parameters.
• Multi-parameter analysis offers a more robust method for accurate BP estimation.

Purpose of the Study:

• To develop and validate a multi-parameter model for cuff-less estimation of systolic blood pressure (SBP) and diastolic blood pressure (DBP).
• To leverage wavelet transformation and pulse wave analysis (PWA) for precise feature extraction from ECG and PPG signals.
• To assess the accuracy and reliability of the developed BP monitoring system against established standards.

Main Methods:

• Simultaneous collection of electrocardiogram (ECG), photoplethysmographic (PPG) signals, and blood pressure (BP) data.
• Application of wavelet transformation to ECG and PPG signals for accurate detection of critical waveform points, even with artifacts.
• Development of multi-parameter regression models using extracted features from pulse wave analysis (PWA) to estimate SBP and DBP.

Main Results:

• The SBP model achieved a mean error of 0.4916 mmHg and standard deviation of 6.3986 mmHg.
• The DBP model achieved a mean error of 0.2527 mmHg and standard deviation of 3.2835 mmHg.
• After outlier removal, mean absolute error improved to 3.854 mmHg for SBP and 2.144 mmHg for DBP, meeting BHS and AAMI standards.

Conclusions:

• The developed multi-parameter model accurately estimates SBP and DBP using ECG and PPG signals.
• Wavelet transformation and PWA are effective in extracting relevant features for cuff-less BP monitoring.
• The system demonstrates potential for comfortable, continuous health monitoring and improved hypertension management.