Research On The Automatic Detection Method Of Parkinson’s Disease Based On The Characteristics Of Bradykinesia

Parkinson’s disease is one of the most common neurodegenerative diseases,and about1.7% of people over the age of 65 in my country suffer from Parkinson’s disease.The main clinical symptoms of Parkinson’s disease patients are resting tremor,muscle rigidity,bradykinesia,and balance disturbance.Human motor function gradually degrades with age,which is similar to the early symptoms of Parkinson’s disease,causing many patients to miss the best treatment time.Therefore,early judgment and accurate tracking of the patient’s condition in order to adjust the patient’s treatment plan in time has become an urgent problem to be solved.The assessment of Parkinson’s disease is usually carried out through a series of assessment methods such as the Parkinson’s scale,but most of these methods rely on the clinical experience of doctors,so the results of the assessment are too subjective and cannot accurately measure the severity of the symptoms.Sensor-based kinematic parameters can provide more accurate and objective information to track the evolution of symptoms,thereby optimizing the management and treatment of Parkinson’s disease.Most of the existing sensor-based studies only focus on the movement of patients with Parkinson’s disease during exercise,ignoring the important feature of the initial reaction time of exercise,and failing to convert the kinematic parameters collected by the sensor into data.,which is to convert the acceleration collected with the sensor as the reference system into the acceleration with the ground coordinate system as the reference system to restore the real motion process.Therefore,this paper mainly studies how to realize the coordinate system transformation of acceleration data and automatic detection of Parkinson’s.The initial symptoms of Parkinson’s mostly start from the hands.Using wearable sensor devices to study the characteristics of Parkinson’s hand bradykinesia can help doctors to judge the condition as soon as possible and take timely treatment measures.Therefore,this paper proposes a method based on the characteristics of hand bradykinesia.Parkinson’s automatic detection method.In the first step,subjects completed a time-free autonomous hand movement task in a laboratory environment,and collected kinematic parameters such as triaxial acceleration and triaxial angular velocity of their hand movements through a wearable device;in the second step,A 3D acceleration coordinate transformation method is proposed,which realizes the restoration of the acceleration data collected in any sensor coordinate system to the acceleration in a unified ground coordinate system,thereby restoring the real movement process of the subjects;Combined with the characteristics of the patient’s slow hand movement,the common features were extracted and the acceleration index,angular velocity index,etc.were constructed as reference indicators for doctors’ daily clinical diagnosis;the fourth step: using convolutional neural network,XGBoost,support vector The automatic detection of Parkinson’s disease can be achieved by using various classifiers such as computers,so as to achieve the purpose of more accurately predicting whether the subject is a Parkinson’s patient,and then assisting doctors in clinical diagnosis.