Existing automated fall detection devices are lacking in their ability to detect falls among wheelchair users. A new TechSAge article in Assistive Technology highlights the development of a fall detection algorithm, developed in a laboratory setting using machine learning techniques, that can accurately differentiate between wheelchair-related falls and wheelchair mobility activities. Researchers conducted a pilot study wherein 30 young, healthy, and ambulatory adults simulated 258 wheelchair falls and 220 wheelchair mobility activities in a lab with fall data retrieved from accelerometers worn on participants’ wrist, chest, and head. Findings indicate that the algorithm should be integrated into wrist-worn devices and further tested among wheelchair users to evaluate its ability to minimize consequences from falls. Request the full article by Libak Abou, Alexander Fliflet, Peter Presti, Jacob Sosnoff, Harshal Mahajan, Mikaela Frechette, and Laura Rice.

Gaby Trier is a senior majoring in Kinesiology with a minor in Disability Studies at the University of Illinois at Urbana-Champaign. Gaby is the lab coordinator for the Disability, Participation and Quality of Life (DPQoL) Lab. She works on the TechSAge Monitoring and Managing Falls project supporting community-based trials of the fall detection prototype for wheelchair users. After graduation, Gaby plans to pursue a physical therapy graduate program. In her free time, Gaby helps to train mobility service dogs.