上肢康复训练智能决策支持系统研究进展

Research Progress on Intelligent Decision-Making Support System for Upper Limb Rehabilitation Training

摘要: 据统计80%的脑卒中运动功能障碍患者患有上肢功能障碍，由于上肢承担了许多精细活动，故其功能恢复难度大。上肢康复训练周期长，很大程度上依赖于治疗师自身的主观经验，而现有用于训练的康复机器人普遍智能性不足，导致其临床效果欠佳。为了减轻治疗师和患者的负担，实现上肢康复训练智能化，智能决策系统成为了康复领域的研究热点之一。该文对近年来上肢康复训练智能决策系统的研究进行了综述，重点对系统知识库构建、特征处理、决策模型搭建方法的优缺点和应用场景进行了分析总结，最后对当前智能决策系统存在的问题和未来发展的趋势展开讨论，以期为相关领域学者提供一定的参考。

Abstract: According to statistics, 80% of stroke patients with motor dysfunction suffer from upper limb dysfunction. Because the upper limb undertakes many fine activities, it is difficult to recover its function. The period of upper limb rehabilitation training is long, which largely depends on the subjective experience of therapists. However, the existing rehabilitation robots used for training is generally lack of intelligence, which leads to poor clinical effect. In order to reduce the burden on therapists and patients and realize the intelligence of upper limb rehabilitation training, an intelligent decision-making support system has become one of the research hotspots in medical rehabilitation. In this paper, research on intelligent decision-making support system for upper limb rehabilitation training in recent years is reviewed, focusing on the advantages, disadvantages and application range of methods used in knowledge base building, feature processing and model building. Finally, current challenges and future development trends are discussed. It is expected that this paper can provide a reference for researchers in related fields.