Mark R. Hodges, Mark W. Newman, Martha E. Pollack
A significant amount of prior work has addressed the question of whether wireless sensor networks can collect sufficient data about how people interact with everyday objects to support accurate interpretation of their behavior. One commonly discussed application for object-use-based activity monitoring is the development of systems to support people with cognitive impairment. However, most prior studies have involved only cognitively intact subjects. In this paper, we report on an experiment in which we attempt to replicate an earlier study that used object-use activity monitoring to identify individuals, but we used subjects with traumatic brain injuries (TBI). We describe the work, which is currently ongoing, and provide preliminary answers to the key questions of whether this approach to activity monitoring is feasible for people with TBI, and whether the data collected provides a level of identification accuracy equivalent to that of the unimpaired subjects.