动态AI肢体功能评估在高校学员体能训练中的应用

Application of dynamic AI limb function assessment in physical training of college students

目的探讨人工智能技术实现肢体运动功能评价的可行性。方法采集某高校2020—2022级学员连续单腿跳跃和功能性动作筛查(FMS)过头深蹲测试的全程影像学资料,通过录像分析进行统计与描述。通过计算机视觉技术实现测试过程的肢体动作捕捉与分析。结果51.7%以上的学员存在肢体运动功能异常,以踝关节灵活性受限为主(46.7%);且女学员的比例高于男学员。连续单脚跳跃测试筛查出的踝关节力线异常比例超过65%,其中15%以上学员在深蹲测试评估中结果为正常。结论计算机视觉技术能够有效识别抗阻运动和动态运动过程中的肢体异常角度和位置信息。动态肢体动作识别与分析系统将为大规模筛查和评估提供支撑。

Objective To explore the feasibility of artificial intelligence technology to achieve limb motor function evaluation.Methods The imaging data of continuous one-leg jump and Functional motion screening(FMS)over-squat test of students from 2020-2022 in a college were collected and statistically described by video analysis.The body motion capture and analysis of the test process are realized by computer vision technology.Results More than 51.7%of the students had abnormal limb motor function,mainly limited ankle flexibility(46.7%).The proportion of female students is higher than that of male students.The proportion of abnormal ankle joint force line screened by the continuous one-foot jump test was more than 65%,and more than 15%of the students were normal in the squat test evaluation.Conclusion Computer vision technology can effectively identify the abnormal angle and position information of limbs during resistance movement and dynamic movement.The dynamic limb movement recognition and analysis system will support mass screening and evaluation.