Estimating risk of falling in the elderly by monitoring daily activities

This project aims to develop signal processing algorithms to extract features from inertial signals which correlate with current clinical measures of fall risk. The signals are acquired during normal unsupervised movement using a small waist-worn monitor.

This project aims to develop signal processing algorithms to extract features from inertial signals which correlate with current clinical measures of fall risk. The signals are acquired during normal unsupervised movement using a small waist-worn monitor. Falls injuries cost the healthcare system around $500 million each year; hence, fall prevention is desirable. No ambulatory fall risk monitors currently exist due to the difficulty of interpreting these complex movement signals in free-living environments. The developed algorithms will help foster the future trialling and development of fall risk monitors, with obvious benefits for Australia’s biomedical engineering industry and the future well-being of Australia’s older people.

Key contact

+61-2-9385-0561
s.redmond@unsw.edu.au