At present, the upper limb function of stroke patients is often assessed clinically using a scale method, but this method has problems such as time-consuming, poor consistency of assessment results, and high participation of rehabilitation physicians. To overcome the shortcomings of the scale method, intelligent upper limb function assessment systems combining sensors and machine learning algorithms have become one of the hot research topics in recent years. Firstly, the commonly used clinical upper limb functional assessment methods are analyzed and summarized. Then the researches on intelligent assessment systems in recent years are reviewed, focusing on the technologies used in the data acquisition and data processing parts of intelligent assessment systems and their advantages and disadvantages. Lastly, the current challenges and future development directions of intelligent assessment systems are discussed. This review is hoped to provide valuable reference information for researchers in related fields.
目前临床上常采用量表方法评估脑卒中患者的上肢功能，但这种方法存在耗时长、评估结果一致性差、需康复医师参与度高等问题。为克服量表方法的短板，结合传感器和机器学习算法的上肢功能智能评估系统成为了近年来的研究热点之一。本文首先对常用的临床上肢功能评估方法做了分析总结，随后对近年来智能评估系统的研究进行了综述，重点对智能评估系统中数据采集和数据处理部分使用的技术及其优缺点进行了分析总结，最后对目前智能评估系统面临的挑战和未来的发展方向展开讨论，以期为相关领域的研究学者提供有价值的参考信息。.
Keywords: Function assessment; Intelligent evaluation; Machine learning; Stroke; Upper limb.