Using gaze-directed vision to identify focus of attention in pervasive healthcare systems

Since the number of devices and people involved in a pervasive healthcare system increases drastically with a huge amount of communication links, it becomes necessary to identify the exact focus of attention of the various personnel involved and to do that in a seamless manner. In this paper, we present a robust mechanism to identify the focus of attention of the medical personnel (doctors, nurses etc.) while dealing with patients, equipment and other data. We propose to use wearable gaze-directed cameras that can capture the exact point where the person is looking to identify his focus. We exploit the fact that is human nature to “look” at the relevant object and/or person attended to. We demonstrate two sample scenarios: First, in a ward with multiple patients, we identify the patient approached by the doctor so that only his information is loaded. Second, in a setup of multiple patients where each has to be given medicines by a nurse one by one while only the medicines relevant to current patient is highlighted in turn. We argue that our approach is unobtrusive and robust, and is capable of handling scenarios involving multiple equi-probable elements of interest.