AN AFFORDABLE GAIT AND POSTURAL BALANCE ANALYSIS SYSTEM USING KINECT FOR REHABILITATION

摘要

This paper presents a Kinect based system with a novel approachrnto find human gait parameters. This system hasrnhuge impact in rehabilitation where patient’s gait patternrnanalysis is monitored after stroke. Our method is the first ofrnits own kind where gait as well as postural stability can bernmonitored using a single system, and can be used at homesrnby patients. This system can act as an affordable alternativernto the costly GAITRite which is a standard gait monitoringrndevice used in clinical practices. We use temporal skeletonrninformation obtained from Kinect to evaluate all parameters.rnSegmenting the skeleton data in time is one of the keyrnsteps in our method. We employ robust curvature detectionrnalgorithm based on eigenvector decomposition, whichrnworks significantly better than standard second derivativernbased approach. The similar technology is also employedrnto understand the characteristics of patient’s postural controlrnin Single Limb Standing (SLS) exercise. Experimentalrnresults demonstrate that our method is comparable asrnwell as compatible with the corresponding parameters ofrnGAITRite and can also be used to calculate SLS durationrnreliably.