Low-Cost AI-Enabled Optoelectronic Wearable for Gait and Breathing Monitoring: Design, Validation, and Applications.

Area of Science:

• Biomedical Engineering
• Wearable Technology
• Rehabilitation Science

Background:

• Continuous patient monitoring is crucial for effective physical rehabilitation.
• Existing systems often lack portability, affordability, or comprehensive data analysis capabilities.
• Integrating movement and physiological monitoring can provide a holistic view of patient recovery.

Purpose of the Study:

• To develop and validate a low-cost, optoelectronic wearable sensor system for portable patient monitoring.
• To assess the system's accuracy in measuring breathing rate, acceleration, and angular velocity.
• To evaluate the system's feasibility for automating biomechanical and physical therapy assessments.

Main Methods:

• Development of a wearable sensor system integrating an inertial measurement unit (IMU) and an optical fiber-integrated chest belt.
• Utilized artificial intelligence (AI) algorithms for data clustering, classification, and regression.
• Validated sensor performance using breathing rate measurements and biomechanical tests (balance and Timed Up and Go - TUG) with 12 subjects.

Main Results:

• Achieved a low root mean squared error (RMSE) of 0.6 BPM for breathing rate monitoring.
• Obtained high accuracy for acceleration (RMSE: 0.037 m/s²) and angular velocity (RMSE: 0.27 °/s).
• Demonstrated the system's ability to detect changes in balance conditions and the dual-task effect in TUG tests.

Conclusions:

• The proposed sensor system offers a low-cost, automated solution for portable patient monitoring.
• It accurately captures physiological and movement data, suitable for physical rehabilitation.
• The system facilitates the automation and assessment of various physical therapy protocols.