Artificial Intelligence-Driven Freezing Of Gait Detection in the Home: Investigating How Free-living Activities Affect the Algorithm

What this Trial Is About

Researchers are evaluating an artificial intelligence (AI) algorithm designed to detect freezing of gait (FOG) episodes in people with Parkinson's disease, a symptom that increases the risk of falling. The study aims to test this AI algorithm in a home environment, which is less controlled than laboratory settings, to see if it can accurately identify FOG episodes. The collected data will also be used to improve the AI's ability to detect FOG automatically and explore real-time detection capabilities. Participants include people with Parkinson's disease who experience daily FOG, those who do not experience FOG, and healthy older adults as controls. The study involves free-living gait assessments over two days, each lasting 5 hours, followed by a standardized gait assessment lasting 4 hours on the third day. The AI algorithm's performance will be compared to expert video analysis, considered the gold standard, across these different walking tests. During the study, participants will wear inertial measurement unit (IMU) sensors to collect movement data while walking in their home environment and during standardized tests. Researchers will measure the percentage of time spent freezing, the number of FOG episodes, and the algorithm's ability to distinguish different types of FOG and medication states. The study also evaluates the consistency of detection between assessments and the occurrence of false detections. Participation includes multiple days of monitoring, and safety and adherence will be closely tracked throughout the study period.

AID-FOG: Artificial Intelligence-Driven Freezing of Gait Detection in the Home