面向ADHD的利用HTC Vive采集手部运动数据的可行性研究

Feasibility Study of Data Acquisition of Hand Movement with HTC Vive for ADHD

摘要:

虚拟现实技术在辅助注意力缺陷伴多动障碍（Attention Deficit Hyperactivity Disorder，ADHD）的客观诊断中取得了一定成效。为了分析ADHD患者与正常儿童在虚拟环境中手部交互的运动差异，需要追踪被试的手部运动并对其进行测量。消费级虚拟现实系统HTC Vive配有手持控制器，具有运动捕捉功能。然而目前临床上还未将手持控制器用于人体手部运动追踪，因此尝试探讨手持控制器用于追踪人体手部运动的可行性，并设计了两个实验。实验一评估HTC Vive系统追踪手持控制器静置时的位置和旋转精度，分析系统的随机误差；实验二评估系统追踪手持控制器的平移和旋转精度。实验一测量得出：手持控制器静置时在X、Y、Z轴的位置抖动误差均小于0.25 mm，旋转抖动误差均小于0.05°；进而采用艾伦方差方法对实测数据进行分析，得到系统量化噪声、零偏不稳定性、速率斜坡等主要误差项系数。实验二测量得出：手持控制器绕X、Y、Z轴的旋转误差均小于0.35°，沿X、Y、Z轴的平移误差均小于3 mm。实验结果表明：HTC Vive系统可用于采集人体的手部运动数据，为辅助ADHD患者的客观诊断提供了可靠的基础支持

Abstract:

Virtual reality technology has made strong progress in helping to objectively diagnose Attention Deficit Hyperactivity Disorder (ADHD). In order to analyze the differences in hand interaction between patients of ADHD and normal children in a virtual environment, it's required to track and measure the hand movement of the subjects. While HTC Vive, a consumer virtual reality system, is equipped with a hand-held controller with motion capture function, its hand-held controller has not been used to track the human hand movement in clinic. So we try to explore the feasibility of using the handheld controller to track the human hand movement and conduct two experiments. Experiment 1 evaluates its position and rotation accuracy, when the handheld controller of the HTC Vive system is in static position; the random error of the system is analyzed. Experiment 2 evaluates the handheld controller's translation and rotation accuracy. Experiment 1 shows that, when the handheld controller is stationary, the location jitter errors of the handheld controller in the X, Y, and Z axes are all less than 0.25 mm and the rotational jitter errors are all less than 0.05°; the Allan variance method is further used to evaluate the measured data and the system’s main error coefficients are obtained, such as quantization noise, bias instability and rate ramp. Experiment 2 shows that the handheld controller's rotation errors around the X, Y, and Z axes are all less than 0.35° and the translation errors are all less than 3 mm along the X, Y , and Z axes. These experimental results indicate that HTC Vive can be adopted to acquire data of human hand movement, which provides a reliable basis for the objective diagnosis of patients with ADHD