Abstract
One out of three subjects older than 65 years falls. Despite extensive research, existing assessment tools for fall risk have been insufficient for predicting falls since the phenomenology is complex and there is no equipment on the market that allows everyday life monitoring. In this paper we present a novel approach for fall-risk on-line assessment based on: i) clinical condition of the subject, ii) environmental conditions, iii) electromyographic (EMG) co-contraction analysis and iv) electroencephalographic (EEG) analysis based on Movement Related Potentials (MRPs) and μ-rhythm event related desynchronizations (μ-ERDs) occurrence. This fall-risk assessment approach is implemented by a complete cyber-physical system made up by EEG and EMG wearable recording systems interfaced to an FPGA on-line performing the needed real-time processing for indexes extraction. The results present a fall-risk assessment case study on healthy subjects walking showing detectable fall-risk increasing (+1.5%) when obstacles are overcome. © 2015 IEEE.