Comparing Microsoft Kinect and Observational Gait Analysis in Assessing Gait Parameters of Apparently Healthy Adults

Abstract

Objectives: Although the Microsoft Kinect has compelling potential for gait analysis in medicine, data available to compare it with observational gait analysis (OGA) is scarce. This study compared the Microsoft Kinect and the OGA in assessing the gait parameters of apparently healthy adults. Methods: Ninety-seven apparently healthy young male adults participated in this comparative study. First, the participant’s age, height, weight, and body mass index were obtained. Afterward, gait parameters involving the number of steps, cadence, stride length, and step length were assessed concurrently following OGA standard procedures and the Microsoft Kinect during a 6-m walk down the hallway. The obtained data were analyzed using descriptive and inferential statistics. The significance level was set at P<0.05. Results: The Mean±SD walk time, steps, cadence, velocity, and stride length were 8.07±1.39 s, 14.0±2.96 counts, 72.9±11.9 steps/min, 0.8±0.13 m/s, and 0.77±0.13m, respectively. Step length was significantly higher (P<0.05) with Microsoft Kinect than OGA, whereas stride length and walk speed values were significantly (P<0.05) lower with Microsoft Kinect. A moderate but significant (P=0.001) positive correlation existed between Microsoft Kinect and OGA regarding walk speed. In contrast, regarding the step length, a weak but significant (P<0.05) positive correlation was found between Microsoft Kinect and OGA. Discussion: Step length values of Microsoft Kinect were significantly higher than OGA values, whereas stride length and walk speed values of Microsoft Kinect were significantly lower than OGA values. Walk speed and step length measured by Microsoft Kinect and OGA were positively correlated.