Fall risk estimation based on co-contraction of lower limb during walking

Monitoring biosignals has become more people-friendly since the development of wearable electro-conductive cloth. This paper proposes an application using wearable EMG sensors that estimates the fall risk of elderly people based on co-contraction of the lower limb during walking. Fifty-two healthy elderly people participated in a gait test with continuous EMG of the thigh (rectus femoris and biceps femoris) and cruris (tibialis anterior and gastrocnemius). The participants were categorized into a "faller group" and a "non-faller group" based on their experience of falls in the previous year. Co-contraction of thigh and cruris were determined based on EMG, and were then used to estimate the experience of falling using a linear discrimination method. The results showed that thigh co-contraction during the stance phase can be predictive of falling experience with 65% accuracy.