The Research On Gait Signal Evaluation Method Of Patients With Parkinson's Disease

The patients suffering from Parkinson's disease(PD)is increasing recently.The physicians still confirm the Parkinson's disease by observing patient's behavior or detecting patient's response after using drugs,which exists a certain degree of subjectivity.Clinically the different physicians make different diagnosis for the same patients,making the patients confused,increasing the burden on the family of patient and even aggravating the disease.Therefore the characteristics of Parkinson's disease movement disorder is as diagnostic clues,this dissertation uses the sensor equipment to quantify the movement disorders of parkinsonian patients and employs Parkinson's disease prediction model to provide objective data support on the treatment of Parkinson's disease for physicians,which has a certain practical prospects.The main contributions in the dissertation can be summarized as follow:(1)We employ a new quantitative assessment system to quantify the gait of patients with Parkinson's disease.To verify the feasibility of the system,this dissertation utilizes the system to collect the gait characteristics of patients with Parkinson's disease and normal people and launch experiment to analyze by T-Test and correlation.The experimental results show that there are significant differences in the gait features of the two types of subjects,which are feasible for the assessment of Parkinson's disease.(2)We improve the accuracy of Parkinson's disease identification by extracting turning gait characteristics in patients with Parkinson's disease.To verify its validity,this dissertation uses a variety of classifiers to analyze contrastively.Experimental results show that the extracted turning gait characteristics are effective for the identification of Parkinson's disease.(3)We propose an approach to eliminate the effects of heighit difference on gait characteristics and implement it.To verify the validity of the method,this dissertation utilizes a variety of classification models to contrastively analyze.Results show that the proposed approach is effective for improving the accuracy of Parkinson's disease prediction model.(4)We further construct a Parkinson's disease gait prediction model for unbalanced sets and make experimental analysis of the model.In order to verify the validity of the model,this dissertation constructs the model and compares with the support vector machine.The experimental results show that the recognition accuracy of the constructed model can achieve 92.05%,which can effectively reduce the risk of misdiagnosis.