基于HI-SVM的老年人跌倒风险评估  

Risk assessment of falls of elderly based on HI-SVM

作　　者：徐方超[1] 韩兰侠 孙凤[1] 郭辉[1] 周冉 张琪[1] XU Fangchao;HAN Lanxia;SUN Feng;GUO Hui;ZHOU Ran;ZHANG Qi(School of Mechanical Engineering,Shenyang University of Technology,Shenyang 110870,China)

机构地区：[1]沈阳工业大学机械工程学院,辽宁沈阳110870

出　　处：《实验技术与管理》2023年第10期36-41,共6页Experimental Technology and Management

基　　金：国家重点研发计划项目(2020YFC2006701,2020YFC2006704);辽宁省教育厅项目(LJKMZ20220460,LJKMZ20220506);辽宁省“揭榜挂帅”科技重大专项(2022JH1/10400027)。

摘　　要：跌倒是影响老年人健康的重要潜在风险因素,准确判断老年人的跌倒风险等级对于提高老年人健康水平至关重要。该文提出了一种基于健康指数(health index,HI)和支持向量机(support vector machine,SVM)模型的新方法,用于判断老年人(60~79岁)的跌倒风险等级。首先,通过HI获得老年人跌倒风险等级的具体分类指标,用于训练HI-SVM模型并建立模型库;然后,利用建立的HI-SVM模型对测试样本进行识别,根据HI确定的风险等级值判断老年人的跌倒风险等级;最后,分别使用网格搜索(grid search,GS)、粒子群优化算法(particle swarm optimization,PSO)和灰狼算法(grey wolf optimizer,GWO)优化SVM参数后,模型的预测准确率可达97.1429%。该方法能够准确识别老年人的跌倒风险等级,为预防老年人跌倒提供了科学的依据。Fall is one of the important potential risk factors affecting the health of the elderly.Accurately judging the risk level of falls in the elderly is essential to improve the health level of the elderly.This paper presents a new method for determining the fall risk level of the elderly based on the health index(HI)and a support vector machine(SVM)model.Firstly,the HI is used to obtain the specific classification index of the fall risk level of the elderly,and the HI-SVM model is trained to establish a model library.Then,the established HI-SVM model is used to identify the test samples,and the fall risk level of the elderly is judged according to the risk level value determined by HI.Finally,after using the SVM parameters optimized by grid search(GS),particle swarm optimization(PSO),and grey wolf optimizer(GWO)to optimize SVM parameters,the prediction accuracy of the model can reach 97.1429%.This method can accurately identify the risk level of falls of the elderly and provide a scientific basis for preventing falls of the elderly.

关 键 词：健康指数 支持向量机 跌倒风险 老年人 加速度传感器 

分 类 号：X913[环境科学与工程—安全科学]