Unobtrusive Inter-beat Interval Estimation from Multichannel Ballistocardiogram Signal Using Kalman Filter

Area of Science:

• Biomedical Engineering
• Physiological Monitoring
• Signal Processing

Background:

• Home-based cardiac monitoring is crucial for assessing heart health, with heart rate variability (HRV) being a key indicator.
• Existing methods for unobtrusive inter-beat interval (IBI) detection from ballistocardiograms (BCG) often rely on single-sensor systems with limited coverage and precision.
• Accurate IBI estimation is essential for reliable HRV calculation and effective cardiac health evaluation.

Purpose of the Study:

• To develop and validate an advanced system for unobtrusive IBI detection using a multichannel ballistocardiogram (BCG) sensor array.
• To enhance the precision and coverage of IBI estimation compared to traditional single-sensor approaches.
• To establish a robust method for multichannel BCG data fusion and signal selection for accurate heart rate variability (HRV) analysis.

Main Methods:

• Implementation of a bed-embedded 9x2 array sensor system for comprehensive BCG signal acquisition.
• Development of a mode-switch based algorithm for optimal sensor signal selection and multichannel data fusion.
• Integration of a linear regression model and Kalman filter for signal processing and IBI estimation.
• Application of a peak detection algorithm for precise IBI calculation from individual channel signals.

Main Results:

• Validation using approximately 48 hours of BCG recordings from 24 subjects in various sleeping positions.
• Achieved a mean absolute error of 31ms for IBI estimation.
• Demonstrated an average signal coverage of 83% across the multichannel sensor system.
• The proposed method proved effective in diverse sleeping conditions.

Conclusions:

• The multichannel BCG array sensor system combined with the proposed algorithm offers a promising solution for accurate and unobtrusive IBI estimation.
• This technology facilitates reliable heart rate variability (HRV) monitoring for at-home cardiac health assessment.
• The findings support the potential of this system for widespread use in personal health monitoring devices.