Identification of Individually Altered Gait Behavior Using an Unobtrusive IMU Sensor Setup

Area of Science:

• Biomechanics
• Wearable Technology
• Gerontology

Background:

• Gait behavior is a critical indicator of overall health and a diagnostic marker for neuro-degenerative and musculoskeletal diseases.
• Changes in gait patterns often signify health deterioration, providing valuable insights for clinicians.
• Objective gait analysis is essential for accurate health assessments.

Purpose of the Study:

• To develop and validate an unobtrusive sensor system for objective gait behavior measurement.
• To assess the system's ability to detect individual gait deterioration.
• To evaluate the use of an aging-simulation suit (GERT) for creating realistic gait impairment data.

Main Methods:

• Utilized a sensor setup with three inertial measurement units (IMUs) on the wrist, chest, and thigh.
• Collected gait data in a movement laboratory, including treadmill and level-ground walking trials.
• Trained classifiers to differentiate normal from GERT-simulated impaired walking patterns, with validation on level-ground trials.

Main Results:

• Achieved high F1-scores (0.965-0.986) during cross-validation for classifier training.
• Demonstrated promising validation results with prediction accuracies exceeding 80% for gait impairment detection.
• Successfully validated the use of the GERT suit for simulating age-related gait changes.

Conclusions:

• An unobtrusive IMU-based system can effectively monitor and identify individual gait deterioration.
• Aging-simulation suits provide a valuable tool for generating realistic gait impairment data for research.
• This technology holds significant clinical relevance for early detection and management of mobility issues.