帕金森病患者红外线数字化步态测量数据的半自动提取方法的建立

Semi-automatic Extraction Method to Establish for the PD Gait Data of Infrared Digital Measurement

很多神经性疾病和骨骼损伤性疾病都会造成运动障碍导致异常步态,帕金森症就是其中的一种,通过步态参数定量评估代替医生目测定性评估,可以更准确的对疾病进行康复评估。目前,对步态定量分析常用的方法是通过运动捕捉仪采集人体三维坐标,再通过三维坐标提取步态特征。在提取过程中,由于原始数据庞大,完全手工处理繁复,同时自动处理中临床步态分界点情况众多,完全自动选取存在困难。本研究结合多个软件的优势,利用matlab绘图手工选取分界点,再自动提取步态特征在友好界面中显示结果,实现对三维坐标的半自动处理,高效地进行步态参数的提取并准确反映了帕金森症临床步态的个体特征。

Many neurological diseases and bone-damaged diseases can cause movement disorder leading to abnormal gait,such as Parkinson＇s disease. It will be more accurate to evaluate the rehabilitation of some disease through quantitative evaluation of gait parameters instead of qualitative evaluation by doctor＇s visual inspection. At present,the common method to make quantitative analysis of gait is to collect the three-dimensional coordinates of human body through the motion capture devices,then to extract the gait characteristics through the three-dimensional coordinates. During the extraction process,it is difficult to conduct completely automatic selection due to the large masses of the original data,the complexity of completely manual processing,as well as the numerous cases of the demarcation points of clinical gait. In this case,with the combination of the advantages of multiple softwares,we utilize matlab to select demarcation points manually and then automatically extracted the displaying results of gait characteristics in friendly interface in order to realize the semi-automatic processing of three-dimensional coordinates,thus managing to extract the gait parameters efficiently as well as reflect the individual characteristics of clinical gait for Parkinson＇s disease accurately.