Abstract

The use of depth-sensing computer vision to capture bodily movement and gestures is increasingly being exploited in healthcare. Yet, there are few descriptions of how real-world practices and usage influence the design and development of such applications. To this end, we present the development and empirical evaluation of ASSESS MS , a system to support the clinical assessment of Multiple Sclerosis (MS) using Kinect. A key issue for developing machine-learning based systems is the need for standardized data on which statistical inferences can be made. I will demonstrate that there are many aspects of real clinical practice that are at odds with the need to capture standardized data for a computer vision system. I will then draw out the importance of providing tools to help users to see what the camera “sees” as a way to achieve standardized data, while maintaining the social interaction of an assessment.