Edge-Enabled Real-Time Gait Assessment for Degenerative Spinal Disease Using Wearable Inertial Sensors

Area of Science:

• Biomedical Engineering
• Wearable Technology
• Clinical Biomechanics

Background:

• Degenerative Spinal Disease (DSD) diagnosis and monitoring often rely on subjective or inaccessible gait analysis methods.
• Conventional gait assessment using visual observation or motion-capture systems limits routine clinical application.

Purpose of the Study:

• To develop and validate an edge-enabled, real-time gait analysis framework for DSD using ankle-worn inertial measurement units (IMUs).
• To create an interpretable clinical index combining gait performance, variability, and asymmetry for DSD assessment.

Main Methods:

• Integration of causal gait-event detection and spatiotemporal gait estimation using two ankle-worn IMUs.
• Implementation of a cycle-wise sensor-orientation correction strategy to enhance stability, reducing reliance on Zero-Velocity Updates (ZUPT).
• Development of a composite score integrating gait metrics for a clinical index, validated against optical motion capture.

Main Results:

• The framework achieved centimeter-level spatial estimation accuracy in gait analysis.
• High accuracy and F1-scores were reported for various walking tasks (Regular-, Toe-, Heel-, Tandem-Walk), demonstrating robust performance.
• Stride length, stride frequency, swing ratio, and step-to-step variability effectively discriminated between healthy controls and DSD patients.

Conclusions:

• Ankle-mounted IMUs combined with edge computation offer a practical, point-of-care solution for gait assessment in DSD.
• The developed system enables accessible, real-time functional monitoring and supports remote patient management.
• Interpretable gait scoring enhances clinical utility for DSD diagnosis and rehabilitation tracking.