Grooten Wilhelmus Johannes Andreas, Sandberg Lisa, Ressman John, Diamantoglou Nicolas, Johansson Elin, Rasmussen-Barr Eva
Karolinska Institutet, Department of Neurobiology, Care Sciences and Society, Division of Physiotherapy, Huddinge, Sweden.
Functional Area Occupational Therapy & Physiotherapy, Allied Health Professionals Function, Karolinska University Hospital, 171 76, Stockholm, Sweden.
BMC Musculoskelet Disord. 2018 Jan 8;19(1):6. doi: 10.1186/s12891-017-1927-0.
Clinical examinations are subjective and often show a low validity and reliability. Objective and highly reliable quantitative assessments are available in laboratory settings using 3D motion analysis, but these systems are too expensive to use for simple clinical examinations. Qinematic™ is an interactive movement analyses system based on the Kinect camera and is an easy-to-use clinical measurement system for assessing posture, balance and side-bending. The aim of the study was to test the test-retest the reliability and construct validity of Qinematic™ in a healthy population, and to calculate the minimal clinical differences for the variables of interest. A further aim was to identify the discriminative validity of Qinematic™ in people with low-back pain (LBP).
We performed a test-retest reliability study (n = 37) with around 1 week between the occasions, a construct validity study (n = 30) in which Qinematic™ was tested against a 3D motion capture system, and a discriminative validity study, in which a group of people with LBP (n = 20) was compared to healthy controls (n = 17). We tested a large range of psychometric properties of 18 variables in three sections: posture (head and pelvic position, weight distribution), balance (sway area and velocity in single- and double-leg stance), and side-bending.
The majority of the variables in the posture and balance sections, showed poor/fair reliability (ICC < 0.4) and poor/fair validity (Spearman <0.4), with significant differences between occasions, between Qinematic™ and the 3D-motion capture system. In the clinical study, Qinematic™ did not differ between people with LPB and healthy for these variables. For one variable, side-bending to the left, there was excellent reliability (ICC =0.898), excellent validity (r = 0.943), and Qinematic™ could differentiate between LPB and healthy individuals (p = 0.012).
This paper shows that a novel software program (Qinematic™) based on the Kinect camera for measuring balance, posture and side-bending has poor psychometric properties, indicating that the variables on balance and posture should not be used for monitoring individual changes over time or in research. Future research on the dynamic tasks of Qinematic™ is warranted.
临床检查具有主观性，其有效性和可靠性往往较低。在实验室环境中，可通过三维运动分析进行客观且高度可靠的定量评估，但这些系统成本过高，无法用于简单的临床检查。Qinematic™是一种基于Kinect摄像头的交互式运动分析系统，是一种用于评估姿势、平衡和侧弯的易于使用的临床测量系统。本研究的目的是测试Qinematic™在健康人群中的重测信度和结构效度，并计算感兴趣变量的最小临床差异。另一个目的是确定Qinematic™在腰痛（LBP）患者中的区分效度。
我们进行了一项重测信度研究（n = 37），两次测量之间间隔约1周；一项结构效度研究（n = 30），将Qinematic™与三维运动捕捉系统进行对比测试；以及一项区分效度研究，将一组腰痛患者（n = 20）与健康对照组（n = 17）进行比较。我们在三个部分测试了18个变量的一系列心理测量特性：姿势（头部和骨盆位置、体重分布）、平衡（单腿和双腿站立时的摆动面积和速度）以及侧弯。
姿势和平衡部分的大多数变量显示出较差/一般的信度（组内相关系数<0.4）和较差/一般的效度（斯皮尔曼相关系数<0.4），在不同测量之间、Qinematic™与三维运动捕捉系统之间存在显著差异。在临床研究中，对于这些变量，Qinematic™在腰痛患者和健康人群之间没有差异。对于一个变量，即向左侧弯，具有出色的信度（组内相关系数 = 0.898）、出色的效度（r = 0.943），并且Qinematic™能够区分腰痛患者和健康个体（p = 0.012）。
本文表明，一种基于Kinect摄像头用于测量平衡、姿势和侧弯的新型软件程序（Qinematic™）具有较差的心理测量特性，这表明平衡和姿势变量不应被用于监测个体随时间的变化或在研究中使用。有必要对Qinematic™的动态任务进行进一步研究。
