计算机工程与应用 ›› 2018, Vol. 54 ›› Issue (1): 166-171.DOI: 10.3778/j.issn.1002-8331.1607-0209

• 模式识别与人工智能 • 上一篇    下一篇

U-型电子步道系统:帕金森病运动障碍评估

张  永1,吴  玺2,许胜强3,杨先军3,王  训4   

  1. 1.合肥工业大学 工业与装备技术研究院,合肥 230009
    2.合肥工业大学 计算机与信息学院,合肥 230009
    3.中国科学院 合肥智能机械研究所,合肥 230031
    4.安徽中医药大学 神经病学研究所附属医院,合肥 230061
  • 出版日期:2018-01-01 发布日期:2018-01-15

U-shape electronic walkway system: Dyskinesia assessment in Parkinson’s disease

ZHANG Yong1, WU Xi2, XU Shengqiang3, YANG Xianjun3, WANG Xun4   

  1. 1. Institute of Industry and Equipment Technology, Hefei University of Technology, Hefei 230009, China
    2. School of Computer and Information, Hefei University of Technology, Hefei 230009, China
    3. Institute of Intelligent Machines, Chinese Academy of Sciences, Hefei 230031, China
    4. Hospital Affiliated to Institute of Neurology, Anhui University of Chinese Medicine, Hefei 230061, China
  • Online:2018-01-01 Published:2018-01-15

摘要: 为了实现帕金森病(PD)运动障碍全面而准确地量化评估和帕金森病的辅助诊断,设计一种新的用于量化步态分析的U-型电子步道系统,该系统包括U-型电子步道和数字诊疗系统。利用U-型电子步道系统获取传感器信号,并采用数字诊疗系统提取受试者直行和转弯状态下运动障碍相关的步态运动学特征。在测试阶段,采用支持向量机(SVM)分类算法识别帕金森病患者步态模式。实验结果表明基于U-型电子步道系统提取的步态运动学特征,构建的模型能实现帕金森病患者识别,同时提取的转弯步态特征能够大幅提高帕金病患者的准确率。

关键词: 帕金森病, 步态分析, U-型电子步道系统, 支持向量机, 转弯步态特征

Abstract: To achieve the comprehensive and accurate quantitative assessment of parkinsonian dyskinesia and assistant diagnosis of Parkinson’s Disease(PD), this paper employes a new electronic walkway system used for quantifying gait analysis, the system contains the U-shape electronic walkway and the digital diagnosis system. Sensor signals are obtained from the U-shape electronic walkway. The gait kinematic characteristics related to the patient dyskinesia are extracted by the digital diagnosis system in straight and turning condition. In the testing phase, the Support Vector Machine(SVM) classification algorithm is used to identify the gait pattern of Parkinson’s disease patients. The experimental results indicate that based on the gait locomotion parameters obtained from electronic walkway, the constructed models not only achieve to recognize Parkinson’s disease patients, but also significantly improve the recognition accuracy by using the turning gait features.

Key words: Parkinson’s disease, gait analysis, U-shape electronic walkway system, Support Vector Machine(SVM), turning gait feature