Reliability and validity of a novel Kinect-based software program for measuring a single leg squat

摘要

The Single leg squat (SLS) is a movement screening test widely used in clinical settings. The SLS is highly subjective in its nature. Objective measures, such as 3D-motion analyses, are seldom used in daily clinical work. An interactive, Kinect-based 3D-movement analysis system, the Qinematic?, is proposed to be easily used in clinical settings to assess the SLS. The aim of this study was to establish the test-retest reliability and construct validity of Qinematic? for assessing the SLS. A further aim was to identify angles of medial knee displacement, to summarise the discriminative ability of the SLS measured by Qinematic?. We performed a test-retest reliability study (n?=?37) of the SLS using Qinematic? and a construct validity study, in which Qinematic? data were compared with visual assessment of video-recorded SLS. Three variables (left knee down, right knee up and down) reached “substantial reliability” (ICC?=?0.64–0.69). One variable, “left knee up”, showed a significant difference between the two test occasions (T1–6.34°, T2 0.66°, p?=?0.013, ICC?=?0.50), and “poor absolute reliability” was seen for all variables (SEM?=?9.04–10.66, SDC?=?25.06–29.55). A moderate agreement between the visual assessment and Qinematic? data for various knee angles was shown (Kappa?=?0.45–0.58). The best discriminative ability of the SLS was found at a knee angle of 6° (AUC?=?0.82, sensitivity?=?0.86, specificity?=?0.78, PPV?=?0.58, NPV?=?0.94). Qinematic? shows a poor absolute reliability, and a substantial relative reliability, in measuring a SLS at the way down. This indicates that Qinematic? should not be recommended for the use on an individual level, but it can possibly be used on a group level. The merged results of the construct validity study indicate that Qinematic? at 6° of medial displacement can identify subjects with a knee over foot position. In summary, the use of the Qinematic? net trajectory angle, which estimates the “line of best fit” cannot be recommended to assess a knee medial to foot position and should be reconsidered.