Research On Freezing Of Gait Detection Method For Patients With Parkinson's Disease Based On IMU

Freezing of gait(FOG)is a gait disorder characterized by recurrent episodes of transient gait lag,often occurring in the late stages of Parkinson's disease(PD),and the cause is unknown.The occurrence of FOG seriously affects the mobility of patients with Parkinson's disease and reduces the quality of life.At present,the clinical professional physician multi-meter is used to judge and severity of the gait,and the objective quantitative evaluation method is less used.Therefore,in clinical work,different doctors have different opinions on the patient's condition,which will result.Seriously affect the improvement of the patient's condition.Therefore,it is particularly important to determine the objectivity of the FOG and the quantitative assessment of its severity,and to better develop a rehabilitation treatment plan for patients with Parkinson's disease.To this end,this paper proposes an IMU-based freezing of gait detection for patients with Parkinson's disease,method.The main work of this paper is as follows:(1)The research background and significance of freezing of gait detection in patients with Parkinson's disease were analyzed.The research status of freezing of gait detection methods in patients with Parkinson's disease at home and abroad was investigated,and the advantages and disadvantages of freezing of gait assessment methods at home and abroad were summarized.(2)The functional indicators and design principles of freezing of gait data acquisition system for patients with Parkinson's disease were explored,and the implementation scheme of freezing of gait data acquisition system for patients with Parkinson's disease based on inertial measurement unit(IMU)was constructed.(3)With reference to the internationally recognized scale,Unified Parkinson Disease Rating Scale(UPDRS),a set of experimental program for freezing of gait detection in patients with Parkinson's disease was designed and experimental data were collected.In addition,a preprocessing method of the motion data collected by the Inertial Measurement Unit is performed.(4)A new method for extracting the freezing threshold and the peak index is proposed.The data characteristics of the dataset and the experimental data are compared to compare the data characteristics in the frequency domain.The validity of the new freezing threshold is verified.Based on the public data set,only the freezing index is used.The detection method and the comparison of the freezing index and the peak index detection method,the experimental results prove the validity of the peak index,and thus the characteristic parameters are obtained.(5)A freezing of gait detection method for Parkinson's disease patients based on support vector machine is proposed.The simple categorization application of support vector machine is used to identify the patient's normal gait and freezing of gait,and compared with the freezing index detection method.The results confirmed the recognition rate of the method.