Faculty of Arts and Education, Research for Educational Impact (REDI), Deakin University , Geelong, Australia.
Institute Intelligent Systems Research & Innovation, Deakin University , Geelong, Australia.
J Sports Sci. 2020 Jul;38(13):1539-1549. doi: 10.1080/02640414.2020.1747743. Epub 2020 Apr 6.
The study purpose was to use Inertial Measurement Units (IMUs) to objectively assess children's motor competence in seven movement skills. Fourteen children aged from seven to 12 years (M = 9.64) participated. Children were asked to perform up to 10 trials of each skill. Children performed the skills, which were captured by XSENS MVN Awinda wireless motion capture, and video. Skills were assessed from video as per the criteria from the Test of Gross Motor Development 3. Initially, 17 IMU sensors were used for signal processing, but this was restricted to four sensors (wrists and ankles) to be more feasible for field assessment. Results of the signal testing against its modelled "Good" signal, showed the skip was classified correctly each time, as was the sidestep. Accuracy % rates for each skill were: kick (95.2), catch (95.0), throw (80.5), jump (78.9), and hop (76.9). Using signal processing-based methods via four sensors was a reliable and feasible way to assess seven motor skills in children. This approach means monitoring and assessment of children's skills can be objective, which will potentially reduce the time involved in motor skill assessment and analysis for research, clinical, sport and education purposes.
本研究旨在使用惯性测量单元（IMU）客观评估儿童在 7 项运动技能中的运动能力。共有 14 名 7 至 12 岁的儿童（M=9.64）参与了研究。要求儿童对每个技能进行最多 10 次尝试。儿童进行了技能测试，这些测试由 XSENS MVN Awinda 无线运动捕捉和视频捕捉完成。根据《儿童运动发育测试 3》的标准，从视频中评估技能。最初，使用了 17 个 IMU 传感器进行信号处理，但为了更便于现场评估，将其限制为四个传感器（手腕和脚踝）。对信号与模拟“良好”信号的测试结果表明，每次跳跃和侧步都被正确分类。每个技能的准确率分别为：踢（95.2%）、接球（95.0%）、投掷（80.5%）、跳跃（78.9%）和单脚跳（76.9%）。使用基于四个传感器的信号处理方法是评估儿童 7 项运动技能的可靠且可行的方法。这种方法意味着对儿童技能的监测和评估可以是客观的，这将有可能减少研究、临床、运动和教育领域中运动技能评估和分析所需的时间。
