Assessment of Sit-to-Stand Movements Using a Single Kinect Sensor

Abstract

The aim of this article is to show that the absolute measurement difference between the Kinect-based system and the marker-based system for the sit-to-stand exercise is less important provided that the difference is caused by systematic errors in the Kinect measurement. Three healthy individuals participated in this study. Marker trajectories and Kinect skeletal joint data were collected while each subject performed the sit-to-stand exercise. Three key parameters, the initial, minimum, and maximum hip angles obtained from the two systems are compared. This preliminary study shows that the Kinect-based system has good agreement with the marker-based system, and it has comparable between-trial reliability and variability to those of the marker-based system. This study shows that Kinect can be used to reliably assess the quality of the sit-to-stand exercise provided that the systematic measurement errors are compensated. This study establishes the foundation to implement a Kinect-based system to automatically monitor, assess, and provide realtime feedback to a patient who carries out the sit-to-stand exercise with or without the supervision of a clinician.