Optimizing pressure sensor array data for a smart-shoe fall monitoring system

Micro-sensors are now integral part of many technologically advanced healthcare systems used in monitoring elderly people who have high risk of fall and other mobility problems. Some theories have been established that relate inconsistency in the gait phases of a person to the possibility of an imminent fall. Using these theories, a number of monitoring systems have been developed to detect and predict falls. Smart shoe is one such solution presented here. It consists of an array of pressure sensors in the shape of a foot. Available sensors in the market do not have pressure sensors customizable to specific requirements such as a fall study. In this research, an optimized layout of pressure sensors is developed for a smart-shoe fall monitoring application. The data from the foot pressure arrays for different activities such as walking, falling forward etc. are collected. Data mining algorithms are then used to classify the fall types accurately and their performances are reported here.