Fall risk assessment using new sEMG-based smart socks

The electromyography signals (EMG) are widely used for the joint movements and muscles contractions monitoring in several healthcare applications. The recent progresses in surface EMG (sEMG) technologies have allowed for the development of low invasive and reliable sEMG-based wearable devices with this aim. These devices promote long-term monitoring, however they are often very expensive and not easy to be appropriately positioned. Moreover they employ mono-use pre-gelled electrodes that can cause skin redness. To overcome these issues, a prototype of a new smart sock has been realized. It is equipped with reusable stretchable and non-adhesive hybrid polymer electrolytes-based electrodes and can send sEMG data through a low energy wireless transmission connection. The developed device detects EMG signals coming from the Gastrocnemius-Tibialis muscles of the legs and it is suitable for lower-limb related pathology assessment, such as age-related changes in gait, sarcopenia pathology, fall risk, etc. In the paper it has been described, as a case study, the use of the socks to detect the risk of falling. A Machine Learning scheme has been chosen in order to overcome the well-known drawbacks of threshold approaches widely used in pre-fall systems, in which the algorithm parameters have to be set according to the users' specific physical characteristics. The supervised classification phase has been obtained through a low computational cost and a high classification accuracy level Linear Discriminant Analysis. The developed system shows high performance in terms of sensitivity and specificity (about 80%) in controlled conditions, with a mean lead-time before the impact of about 700 ms.