Gait velocity has been consistently shown to be an important indicator andpredictor of health status, especially in older adults. It is often assessedclinically, but the assessments occur infrequently and do not allow optimaldetection of key health changes when they occur. In this paper, we show thatthe time gap between activations of a pair of Passive Infrared (PIR) motionsensors installed in the consecutively visited room pair carry rich latentinformation about a person's gait velocity. We name this time gap transitiontime and show that despite a six second refractory period of the PIR sensors,transition time can be used to obtain an accurate representation of gaitvelocity. Using a Support Vector Regression (SVR) approach to model the relationshipbetween transition time and gait velocity, we show that gait velocity can beestimated with an average error less than 2.5 cm/sec. This is demonstrated withdata collected over a 5 year period from 74 older adults monitored in their ownhomes. This method is simple and cost effective and has advantages over competingapproaches such as: obtaining 20 to 100x more gait velocity measurements perday and offering the fusion of location-specific information with time stampedgait estimates. These advantages allow stable estimates of gait parameters(maximum or average speed, variability) at shorter time scales than currentapproaches. This also provides a pervasive in-home method for context-awaregait velocity sensing that allows for monitoring of gait trajectories in spaceand time.
展开▼
