Evaluation and Integration of an Automatic Fall Prediction System

Document Type


Publication Date



Purpose: The objectives were to evaluate the relationship of standardized gait and balance assessments used to estimate fall risk and the Fractal dimension (Fractal D) measure of voluntary movement path variability gathered during everyday activities; evaluate the hypothesis that Fractal D would increase prior to an elderly person’s fall; and create a practical system for automatically and u nobtrusively detecting changes in Fractal D and movement path distance duration and speed. Scope: Falls and fear of falling are about two times more frequent in elderly persons living in nursing homes and assisted living facilities (ALF) than in communit y- dwelling aged peers. The research determined if Fractal D would increase fall predictability beyond estimates provided by fall history, gait and balance assessments, cognitive impairments, and psychoactive medication use. Methods: T he voluntary movements of 53 elderly (M=76.6, SD=11.1 years, 83% female) residents in two ALFs were monitored for one year; their falls were recorded . Results: Mean Fractal D during the week preceding a fall in 23 fallers was statistically significantly higher than for 30 non-fallers whose week represented the midpoint of their observation period. Logistic regression analysis showed only Fractal D and a history of one or more previous falls were significant fall predictors, results supporting the hypotheses tied to the first two objectives of the research. The system for measuring movement path variability was awarded a patent in 2011.

Was this content written or created while at USF?


Citation / Publisher Attribution

Evaluation and Integration of an Automatic Fall Prediction System