Abstract:

Developmental Coordination Disorder（DCD） is a high-incidence problem in the growth of children, which seriously affects the physical and mental health of children and causes many adverse effects on the long-term health of adults. With the development of artificial intelligence, human pose estimation and action recognition technology in the field of computer vision can be used to assist in the diagnosis of such diseases, which is conducive to improving the degree of medical inclusiveness and medical efficiency, and is of great significance to alleviating the shortage of medical resources. The review reports the current research status of the AI diagnosis system for DCD, introducs the clinical diagnosis methods of DCD, and based on this, proposes the diagnosis technology route of computer-aided diagnosis of DCD, and summarizes the human pose estimation and action recognition deep learning method, analyzes the currently mainstream assessment methods and datasets of human pose estimation and action recognition, discusses the problems and challenges of its application in computer-aided diagnosis.

Key words: developmental coordination disorder, human pose estimation, action recognition

摘要：

运动协调障碍是儿童生长发育过程中的一种高发问题，严重影响儿童身心健康并对成年后的远期健康造成多种不良影响。随着人工智能的发展，可利用计算机视觉领域的人体姿态估计和动作识别技术辅助诊断这类疾病，有利于提升医疗普惠程度和医疗效率，对缓解医疗资源不足有着重要意义。报告了儿童运动协调障碍AI诊断系统的研究现状，介绍了儿童运动协调障碍的临床诊断方式，并基于此提出了计算机辅助诊断该类疾病的诊断技术路线，总结了人体姿态估计和动作识别任务的深度学习方法，分析了目前主流使用的人体姿态估计与动作识别任务评估指标与其对应数据集，讨论了其应用于计算机辅助诊断的问题及挑战。

关键词: 儿童运动协调障碍, 人体姿态估计, 动作识别