基于虚拟现实技术的康复训练系统设计与效果评估研究  

Research on the Design and Effect Evaluation of Rehabilitation Training System Based on Virtual Reality Technology

作　　者：章蕾 张执 ZHANG Lei;ZHANG Zhi(Shengjing Hospital of China Medical University,Shenyang 110022,China;School of Mechanical Engineering and Automation,Shenyang Institute of Technology,Shenyang 113122,China)

机构地区：[1]中国医科大学附属盛京医院,沈阳110022 [2]沈阳工学院机械工程与自动化学院,沈阳113122

出　　处：《计算机测量与控制》2025年第2期317-324,共8页Computer Measurement &Control

基　　金：辽宁省科技计划联合计划(应用基础研究项目)(2023JH2/101700068)。

摘　　要：虚拟现实技术被越来越多地应用到康复训练中,但其实际的康复训练效果仍然不够理想;为此,研究结合脑电信号和虚拟现实技术设计了康复训练系统;首先,利用非入侵式电机采集运动想象脑电信号,并进行预处理;其次,利用粒子群优化算法改进支持向量机,进而对采集到的运动想象脑电信号进行分析;然后,以识别分类后的运动想象脑电信号为输入,设计一种基于虚拟现实技术的康复训练系统;最后,提出一种基于改进长短期记忆网络的康复训练效果评估模型,以验证系统的应用效果;实验结果表明,在对两名受试者的运动想象训练实验中,所提改进支持向量机分类器对于受试者的分类准确率最高,为95.54%;所提康复训练效果评估模型对于15名患者的评估准确率均在93%以上,最高可达到96.02%,对上肢功能障碍的评估精确率为95.87%;实验结果证实了所提系统和效果评估模型的性能,有利于提高康复训练过程的智能化水平,改善患者的康复训练效果。Virtual reality technology is increasingly being applied in rehabilitation training,but its actual rehabilitation training effect is still not ideal.To this end,a rehabilitation training system was designed by combining EEG signals and virtual reality technology.Firstly,non-invasive motors are used to collect motor imagery EEG signals and preprocess them.Secondly,using particle swarm optimization algorithm to improve support vector machine,and then analyzing the collected motor imagery EEG signals.Then,using the identified and classified motor imagery EEG signals as input,a rehabilitation training system based on virtual reality technology is designed.Finally,a rehabilitation training effectiveness evaluation model based on an improved long short-term memory network is proposed to verify the application effectiveness of the system.The experimental results showed that in the exercise imagination training experiment on two subjects,the proposed improved support vector machine classifier achieved the highest classification accuracy of 95.54%for the subjects.The proposed rehabilitation training effect evaluation model has an accuracy rate of over 93%for 15 patients,with the highest reaching 96.02%.The accuracy rate for evaluating upper limb dysfunction is 95.87%.The experimental results confirmed the performance of the proposed system and effect evaluation model,which is beneficial for improving the intelligence level of rehabilitation training process and enhancing the rehabilitation training effect of patients.

关 键 词：虚拟现实 康复训练 脑电信号 效果评估 循环神经网络 

分 类 号：TP274[自动化与计算机技术—检测技术与自动化装置]