Muscle stiffness symptom evaluation system for Parkinson's disease patient

Abstract

The invention discloses a myotonia symptom assessment system for a Parkinson's disease patient, and relates to the field of medical equipment, and the system comprises a mechanical impedance data acquisition module which obtains pressure data, an angle and an angular velocity; the electroencephalogram signal acquisition module is used for acquiring an electroencephalogram signal of the tested object when the passive motion electromechanical device drives the arm of the tested object to passively move in a resting state; the processing module judges whether the Parkinson's disease exists or not according to the electroencephalogram signals; and if the Parkinson's disease is judged, determining a mechanical impedance value according to the pressure data, the angle and the angular velocity, and determining a myotonicity level according to the mechanical impedance value. The judgment of the electroencephalogram signals is added in the evaluation process, so that whether the tested person exerts subjective force or not can be judged, and the evaluation preciseness is improved; the mechanical impedance value is determined by using the pressure data, the angle and the angular velocity, quantitative analysis of the myotonia grade is realized, and subjective judgment of a doctor is avoided; the accuracy and objectivity of judging the myotonicity level of the Parkinson's disease are improved, the labor cost is reduced, and the diagnosis efficiency is improved.

Description

Technical Field

[0001] The present application relates to the field of medical equipment, and in particular to a system for evaluating myotonia symptoms in patients with Parkinson's disease. Background Art

[0002] Parkinson's disease ， Parkinson's Disease (PD) is a neurodegenerative disease that affects millions of people worldwide. Myotonia is one of its main symptoms, and the degree of myotonia is an important indicator for evaluating Parkinson's disease. Although there are currently a variety of methods for evaluating PD myotonia, they often rely on subjective judgment and lack accurate quantitative analysis methods.

[0003] Generally, Parkinson's myotonia needs to be diagnosed by professional doctors in the hospital. Doctors judge the degree of Parkinson's myotonia based on their experience and feelings. They are unable to judge whether subjective force is used during the mechanical impedance data collection process, resulting in low accuracy in judging the degree of Parkinson's myotonia. There are also differences in evaluations between different doctors, which lacks objectivity. In addition, the labor cost required is expensive.

[0004] Therefore, how to improve the accuracy and objectivity of Parkinson's disease myotonia grade judgment and reduce labor costs are problems that technical personnel in this field need to solve. Summary of the invention

[0005] The purpose of this application is to provide a system for evaluating myotonia symptoms in patients with Parkinson's disease, which is used to solve the problems of low accuracy and objectivity in judging the degree of myotonia in Parkinson's disease and high labor costs.

[0006] In order to solve the above technical problems, the present application provides a system for evaluating myotonia symptoms in patients with Parkinson's disease, comprising:

[0007] A mechanical impedance data acquisition module, used to acquire pressure data, angle and angular velocity detected by a passive motion electromechanical device;

[0008] An EEG signal acquisition module, used to acquire EEG signals of the subject in a resting state and when the passive motion electromechanical device drives the subject's arm to move passively;

[0009] A processing module is used to determine whether it is Parkinson's disease according to the EEG signal; if it is determined to be Parkinson's disease, a mechanical impedance value is determined according to the pressure data, the angle and the angular velocity, and a myotonia grade is determined according to the mechanical impedance value.

[0010] Optionally, the processing module includes:

[0011] A first determining submodule, configured to determine an angular acceleration according to the angular velocity;

[0012] The second determination submodule is used to determine the viscosity coefficient and the elastic coefficient in the first preset formula by using the least square method, and the first preset formula is: ;

[0013] The third determination submodule is used to determine the mechanical impedance value according to a second preset formula, where the second preset formula is: ;

[0014] in, is the mechanical impedance value, is the torque, is the moment of inertia coefficient, is the viscosity coefficient, is the elastic coefficient, is the angular acceleration, is the angular velocity, is the angle, is a constant, and the torque is calculated from the pressure data applied to the wrist of the object being measured and the distance between the force application point and the rotation axis.

[0015] Optionally, the processing module includes:

[0016] An optimization submodule, used for constructing an autoregressive model and optimizing model parameters of the autoregressive model using prediction errors;

[0017] The determination submodule is used to input the EEG signal into the optimized autoregressive model, and determine whether it is Parkinson's disease according to the output result of the optimized autoregressive model.

[0018] Optionally, the processing module further includes:

[0019] A fourth determination submodule, used to determine a number of sampling points of the EEG signal and sampling data corresponding to the sampling points;

[0020] The fifth determination submodule is used to set an extraction coefficient for extracting features, and determine a forward prediction error and a backward prediction error according to the sampled data and the extraction coefficient.

[0021] Optionally, the optimization submodule includes:

[0022] An initialization unit, used to initialize a forward prediction error and a backward prediction error;

[0023] A reflection coefficient solving unit is used to call the Burg method to recursively solve the reflection coefficient by minimizing the mean square value of the forward prediction error and the backward prediction error;

[0024] A model parameter determination unit is used to determine the model parameters of the autoregressive model according to the reflection coefficient.

[0025] Optionally, the reflection coefficient solving unit includes:

[0026] A calculation subunit, used for calculating the current order reflection coefficient according to the current order forward prediction error and the current order backward prediction error in each order iteration;

[0027] A determination subunit, configured to determine an updated forward prediction error and an updated backward prediction error using the current-order reflection coefficient;

[0028] The replacement subunit is used to replace the current-order forward prediction error with the updated forward prediction error, and replace the current-order backward prediction error with the updated backward prediction error, and enter the step of calculating the current-order reflection coefficient based on the current-order forward prediction error and the current-order backward prediction error, until the reflection coefficient of each iteration is calculated.

[0029] Optionally, the determination submodule includes:

[0030] A determination unit, used for determining whether the disease is Parkinson's disease according to the classification result of the classifier in the autoregressive model;

[0031] Wherein, the determination unit comprises:

[0032] A weight coefficient determination subunit, used to determine the weight coefficient of the linear prediction model when the total energy error is minimized in each iteration process;

[0033] A generating subunit, configured to generate a feature vector according to the weight coefficient; the feature vector includes a first feature vector of a Parkinson's disease patient research object and a second feature vector of a healthy person research object;

[0034] A classifier construction subunit performs principal component analysis on different channels, determines a hyperparameter selection of the number of principal components, and constructs a classifier according to the hyperparameter selection and the feature vector;

[0035] An unbiased eigenvector determining subunit, used to determine a scaled unbiased eigenvector according to the eigenvector, the number of research objects and a basis vector of the eigenvector;

[0036] A singular value decomposition subunit, used for performing singular value decomposition on the unbiased eigenvector to obtain a singular vector and a singular value;

[0037] A hyperplane distance determination subunit is used to select a set number of singular values ​​as hyperplane segmentation and calculate the hyperplane distance constructed by the principal component;

[0038] The classification result determination subunit is used to determine the classification result according to the hyperplane distance.

[0039] Optionally, the passive motion electromechanical device includes: a base, a motor, a rotating plate, a sliding plate, a joint fixing assembly, a main board, a flexible pressure sensor and an inertial sensor, the motor is provided on the base, the motor is connected to the bottom of the rotating plate, and is used to drive the rotating plate to rotate in a horizontal direction, a sliding groove is provided in the length direction of the rotating plate, the sliding plate is slidably connected to the sliding groove, the joint fixing assembly is provided on the sliding plate, the joint fixing assembly is used to fix the arm of the object to be measured, the flexible pressure sensor is provided on the joint fixing assembly, the inertial sensor is provided at the bottom of the rotating plate, the flexible pressure sensor is used to detect the pressure data, and the inertial sensor is used to detect the angular velocity and the angle of the rotating plate.

[0040] Optionally, the joint fixation assembly includes a joint fixation plate and an elastic band, the joint fixation plate is a U-shaped fixation plate, the bottom of the U-shaped fixation plate is connected to the sliding plate, and both ends of the U-shaped fixation plate are respectively connected to both ends of the elastic band.

[0041] Optionally, a limiting rod is further included, the sliding plate is provided with a screw hole, one end of the limiting rod is threadedly connected to the screw hole, and passes through the sliding plate through the screw hole to abut against the sliding groove.

[0042] The present application provides a system for evaluating the symptoms of myotonia in patients with Parkinson's disease, including: a mechanical impedance data acquisition module for acquiring pressure data, angle and angular velocity detected by a passive motion electromechanical device; an electroencephalogram signal acquisition module for acquiring the electroencephalogram signal of the subject under test in a resting state and when the passive motion electromechanical device drives the arm of the subject under test to move passively; a processing module for determining whether it is Parkinson's disease according to the electroencephalogram signal; if it is determined to be Parkinson's disease, the mechanical impedance value is determined according to the pressure data, angle and angular velocity, and the myotonia grade is determined according to the mechanical impedance value. By adding the judgment of the electroencephalogram signal in the evaluation process, it can be determined whether the subject is exerting force subjectively, thereby increasing the rigor of the evaluation; the mechanical impedance value is determined by using the pressure data, angle and angular velocity to realize the quantitative analysis of the myotonia grade, effectively avoiding the subjective judgment of the doctor; thereby improving the accuracy and objectivity of the judgment of the myotonia grade of Parkinson's disease, reducing the labor cost and improving the efficiency of diagnosis. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0044] Figure 1A structural diagram of a system for evaluating myotonia symptoms in patients with Parkinson's disease provided in an embodiment of the present application;

[0045] Figure 2 A structural diagram of a passive motion electromechanical device provided in an embodiment of the present application;

[0046] The figure numbers are as follows: 1-mechanical impedance data acquisition module, 2-EEG signal acquisition module, 3-processing module, 4-base, 5-motor, 6-rotating plate, 7-sliding slot, 8-sliding plate, 9-joint fixing plate. DETAILED DESCRIPTION

[0047] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0048] Electroencephalogram (EEG) technology has become one of the important tools for diagnosing Parkinson's disease (PD) due to its direct correlation with brain activity. This technology can use the brain's electrical activity measured at rest to reveal abnormalities in the cerebral cortex of Parkinson's patients, such as abnormal energy in specific frequency bands. When evaluating myotonia, in addition to the patient's resting EEG activity, changes in the patient's EEG activity when undergoing passive movement, such as changes in energy in each frequency band and changes in connectivity in each brain region, also reflect the severity of Parkinson's disease and can be used as a basis for judging Parkinson's disease.

[0049] The core of this application is to provide a myotonia symptom assessment system for Parkinson's disease patients, which is used to improve the accuracy and objectivity of Parkinson's disease myotonia grade judgment and reduce labor costs.

[0050] In order to enable those skilled in the art to better understand the present application, the present application is further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0051] Figure 1 A structural diagram of a system for evaluating myotonia symptoms in patients with Parkinson's disease provided in an embodiment of the present application, such as Figure 1 As shown, a system for evaluating myotonia symptoms in patients with Parkinson's disease comprises:

[0052] The mechanical impedance data acquisition module 1 is used to acquire the pressure data, angle and angular velocity detected by the passive motion electromechanical device.

[0053] The EEG signal acquisition module 2 is used to acquire the EEG signal of the subject in a resting state and when the passive motion electromechanical device drives the subject's arm to move passively.

[0054] Processing module 3 is used to determine whether it is Parkinson's disease based on the EEG signal; if it is Parkinson's disease, the mechanical impedance value is determined according to the pressure data, angle and angular velocity, and the myotonia grade is determined according to the mechanical impedance value.

[0055] The above modules can be applied to the host computer. Figure 2 A structural diagram of a passive motion electromechanical device 2 provided in an embodiment of the present application is shown in FIG. Figure 2 As shown, the passive motion electromechanical device 2 includes: a base 4, a motor 5, a rotating plate 6, a sliding plate 8, a joint fixing assembly, a main board, a flexible pressure sensor and an inertial sensor. The base 4 is provided with a motor 5, and the motor 5 is connected to the bottom of the rotating plate 6 for driving the rotating plate 6 to rotate in the horizontal direction. A sliding groove 7 is provided in the length direction of the rotating plate 6, and the sliding plate 8 is slidably connected to the sliding groove 7. The joint fixing assembly is arranged on the sliding plate 8, and the joint fixing assembly is used to fix the arm of the object to be measured. The main board is respectively connected to the host computer and the motor 5, and the host computer is respectively connected to the flexible pressure sensor and the inertial sensor, and the flexible pressure sensor is arranged on the joint fixing assembly, and the inertial sensor is arranged at the bottom of the rotating plate 6. The flexible pressure sensor is used to detect pressure data, and the inertial sensor is used to detect the angular velocity and angle of the rotating plate 6.

[0056] The embodiment of the present application does not specifically limit the joint fixing assembly, and any structure that can fix the arm of the object to be measured is acceptable. Specifically, the joint fixing assembly includes a joint fixing plate 9 and an elastic band, and the joint fixing plate 9 is a U-shaped fixing plate, the bottom of the U-shaped fixing plate is connected to the sliding plate 8, and the two ends of the U-shaped fixing plate are respectively connected to the two ends of the elastic band. The arm of the object to be measured is fixed by cooperating with the U-shaped fixing plate and the elastic band, and it can be suitable for arms of different thicknesses. The U-shaped fixing plate can be made of resin material, and the sliding groove 7 can be made of stainless steel, so that it will not be severely deformed due to external force.

[0057] Furthermore, it also includes a limit rod, a screw hole is provided on the sliding plate 8, one end of the limit rod is threadedly connected to the screw hole, and passes through the sliding plate 8 through the screw hole to abut against the sliding groove 7. By tightening the limit rod, the sliding plate 8 can be positioned at the preset position of the sliding groove. The embodiment of the present application does not specifically limit the number of sliding plates 8, the number of joint fixing components matches the number of sliding plates 8, two sliding plates 8 can be set, and a joint fixing component is set on each sliding plate 8. The distance between the two sliding plates 8 can be adjusted according to the arm length of the object to be measured. After determining the position of each sliding plate 8 in the sliding groove 7, tighten the limit rod to fix each sliding plate 8 at the set position of the sliding groove 7. Correspondingly, the number of flexible pressure sensors can be 2, that is, a flexible pressure sensor is set on each joint fixing plate 9.

[0058] During the test, the elbow of the subject's arm is placed flat on the rotating plate 6 and fixed by the joint fixing assembly, so that the force can be driven by the biceps and triceps. The motor 5 drives the rotating plate 6 to pull the subject's arm to rotate back and forth in the horizontal direction. The force of the flexible pressure sensor is on the joint fixing plate 9, which can directly measure the resistance of the subject during passive movement, that is, the pressure data. The inertial sensor can be a gyroscope, which can detect the angular velocity and angle of the rotating plate 6.

[0059] In the process of passively measuring mechanical impedance, traditional equipment requires external manpower to drive the arm, and the evaluation results are easily affected by the instability of human force. The present application uses motor 5 to drive to improve accuracy; the direction of passive traction of the traditional arm is a swing along the vertical direction, which will be affected by gravity during the swinging process. The present application sets the swing direction of the arm to a horizontal swing, which overcomes the above-mentioned problems.

[0060] Regarding how to detect the EEG signal of the object under test, an EEG device can be used to record the corresponding EEG signal, and the host computer is respectively connected to the EEG device and the passive motion electromechanical device to obtain the EEG signal and mechanical impedance data of the object under test while the passive motion electromechanical device 2 drives the arm of the object under test to move passively. The EEG device includes invasive or non-invasive types. The objects under test include humans and non-human primates, etc.

[0061] The host computer of the embodiment of the present application may include an interactive interface, which can be used by the user to import data, select analysis parameters and generate visualization results so that the user can intuitively understand the data.

[0062] A system for evaluating the symptoms of myotonia in patients with Parkinson's disease provided in an embodiment of the present application includes: a mechanical impedance data acquisition module for acquiring pressure data, angle and angular velocity detected by a passive motion electromechanical device; an electroencephalogram signal acquisition module for acquiring the electroencephalogram signal of the subject under test in a resting state and when the passive motion electromechanical device drives the arm of the subject under test to passively move; a processing module for determining whether it is Parkinson's disease based on the electroencephalogram signal; if it is determined to be Parkinson's disease, the mechanical impedance value is determined based on the pressure data, angle and angular velocity, and the myotonia level is determined based on the mechanical impedance value. By adding the judgment of the electroencephalogram signal in the evaluation process, it can be determined whether the subject is exerting force subjectively, thereby increasing the rigor of the evaluation; the mechanical impedance value is determined by using the pressure data, angle and angular velocity to realize the quantitative analysis of the myotonia level, effectively avoiding the subjective judgment of the doctor; thereby improving the accuracy and objectivity of the judgment of the myotonia level of Parkinson's disease, reducing the labor cost, and improving the efficiency of diagnosis.

[0063] Based on the above embodiment, the processing module of the embodiment of the present application includes:

[0064] A first determination submodule, configured to determine an angular acceleration according to an angular velocity;

[0065] The second determination submodule is used to determine the viscosity coefficient and the elastic coefficient in the first preset formula by using the least square method. The first preset formula is:

[0066] (1);

[0067] The third determination submodule is used to determine the mechanical impedance value according to a second preset formula, where the second preset formula is:

[0068] (2);

[0069] in, is the mechanical impedance value, is the torque, is the moment of inertia coefficient, is the viscosity coefficient, is the elastic coefficient, is the angular acceleration, is the angular velocity, is the angle, is a constant, and the torque is calculated from the pressure data applied to the wrist of the measured object and the distance from the force application point to the rotation axis.

[0070] because, is a constant, and the torque is calculated from the pressure data applied to the wrist of the object being measured and the distance from the force application point to the rotation axis. Based on the first preset formula, the following formula can be obtained:

[0071] (3);

[0072] is a constant, is the pressure data, is the distance between the force application point and the rotation axis. The viscosity coefficient and elastic coefficient are determined by the least square method, and then the mechanical impedance value is determined by substituting them into the second preset formula. Angular velocity detected by a passive motion electromechanical device.

[0073] Based on the above embodiments, the processing module of the embodiments of the present application includes: a fourth determination submodule, which is used to determine a number of sampling points of the EEG signal and the sampling data corresponding to the sampling points; a fifth determination submodule, which is used to set the extraction coefficient for extracting features, and determine the forward prediction error and the backward prediction error according to the sampling data and the extraction coefficient; an optimization submodule, which is used to construct an autoregressive model and optimize the model parameters of the autoregressive model using the prediction error; and a judgment submodule, which is used to input the EEG signal into the optimized autoregressive model and judge whether it is Parkinson's disease according to the output result of the optimized autoregressive model.

[0074] Among them, determining the forward prediction error and the backward prediction error specifically includes: the EEG acquisition device uses a non-invasive electroencephalogram (EEG) device, and the EEG signal classification is calculated by the following formula. The Burg method (also known as the Burg algorithm, which is a spectral analysis technique for estimating the parameters of the autoregressive (AR) model) recursively uses the data model to minimize the total energy of the forward and backward prediction errors. Assume that the N sampling points of the EEG signal are:

[0075] (4);

[0076] The first The coefficient of the feature extraction order is set to ; , in the data window The forward prediction error in the is an attempt to learn from previous data The error when predicting the sample:

[0077] (5);

[0078] The same backward prediction error is determined by the following error:

[0079] (6);

[0080] The time series The total energy of the forward and reverse prediction errors is:

[0081] (7);

[0082] Based on the above embodiments, the optimization module of the embodiment of the present application includes: an initialization unit for initializing the forward prediction error and the backward prediction error; a reflection coefficient solving unit for calling the Burg method to recursively solve the reflection coefficient by minimizing the mean square value of the forward prediction error and the backward prediction error; a model parameter determination unit for determining the model parameters of the autoregressive model according to the reflection coefficient. Among them, the reflection coefficient solving unit includes: a calculation subunit for calculating the current order reflection coefficient according to the current order forward prediction error and the current order backward prediction error in each order iteration; a determination subunit for determining the updated forward prediction error and the updated backward prediction error using the current order reflection coefficient; a replacement subunit for replacing the current order forward prediction error with the updated forward prediction error, and replacing the current order backward prediction error with the updated backward prediction error, and entering the step of calculating the current order reflection coefficient according to the current order forward prediction error and the current order backward prediction error, until the reflection coefficient of each step iteration is calculated.

[0083] In an autoregressive model (AR model), the model form is usually expressed as: .

[0084] in: is a time series at a point in time The value of is the model order, yes Parameters, and is the error term.

[0085] At a given order Take 2 to 10:

[0086] 1. Initialize forward and backward The prediction errors are the original data.

[0087] 2. For a given order p, take m = 1,2, …, p,

[0088] (8);

[0089] along with Iterative Updates and In the Burg algorithm, is the reflection coefficient (also called partial correlation coefficient) for each iteration, which is calculated by the forward and backward prediction errors of order. Here is the total length of the signal, is a time index.

[0090] The specific calculation method is as follows:

[0091] (9);

[0092] (10);

[0093] Updates are from arrive The updated value is recorded as and , then replace the original and .

[0094] use Update the AR coefficients:

[0095] ,right , represents the order of the previous step, and .

[0096] Based on the above embodiments, the determination submodule of the embodiment of the present application includes: a determination unit, which is used to determine whether it is Parkinson's disease according to the classification result of the classifier in the autoregressive model; wherein the determination unit includes: a weight coefficient determination subunit, which is used to determine the weight coefficient of the linear prediction model when the total energy error is minimized in each iteration process; a generation subunit, which is used to generate a feature vector according to the weight coefficient; the feature vector includes a first feature vector of a Parkinson's patient research object and a second feature vector of a healthy research object; a classifier construction subunit, which performs principal component analysis on different channels, determines the hyperparameter selection of the number of principal components, and constructs a classifier according to the hyperparameter selection and the feature vector; an unbiased feature vector determination subunit, which is used to determine the scaled unbiased feature vector according to the feature vector, the number of research objects and the basis vector of the feature vector; a singular value decomposition subunit, which is used to perform singular value decomposition on the unbiased feature vector to obtain a singular vector and a singular value; a hyperplane distance determination subunit, which is used to select a set number of singular values ​​as a hyperplane segmentation, and calculate the hyperplane distance constructed by the principal component; and a classification result determination subunit, which is used to determine the classification result according to the hyperplane distance.

[0097] The classification result determination process specifically includes: calculating the number of different LPC coefficients (i.e. linear predictive coding coefficients) Come in every The minimum total energy error of the order . Using LPC coefficients Generate feature vectors for classification. In addition, it is necessary to select the hyperparameter of the number of principal components n for principal component analysis according to different channels and build a classifier:

[0098] (11);

[0099] in is the feature vector of k Parkinson's disease patients, is the feature vector calculated for K healthy subjects. and The dimension is , select As a vector The basis vectors are selected As The basis vectors of , we get:

[0100] (12);

[0101] in yes No. The mean of the column;

[0102] Then calculate the scaled unbiased eigenvector:

[0103] (13);

[0104] in ; and , , and then perform singular value decomposition to obtain:

[0105] (14);

[0106] ; is an orthogonal matrix. and is a diagonal matrix. The diagonal elements are and The singular values ​​of , and then select the appropriate number of singular values ​​as the hyperplane segmentation.

[0107] Classification is performed using vector projection on principal components. Assume that the LPC feature vector for a given subject is , use the vector projection formula to determine the point Minimum distance to the hyperplane created by the set of first n principal components related to the PD topic , minimum distance For point Distance to the hyperplane constructed from the first n principal components of healthy subjects:

[0108] (15);

[0109] Finally, the classification result is obtained using the formula:

[0110] (16);

[0111] when It indicates that the subspace is closer to the healthy group, whereas it is closer to the Parkinson's group.

[0112] Once the initial data is collected using the EEG device, it is processed according to the following steps.

[0113] Assuming that the collected sequence is N data, the energy is first normalized:

[0114] (17);

[0115] The normalized sequence is:

[0116] (18);

[0117] The sixth-order Butterworth filter is used for filtering to retain the information of the relevant frequency band. The features are extracted using the previous LPC method. , principal component analysis (PCA) was applied to this data matrix using equations (6), (7), and (8), and the set of principal components associated with PD subjects was calculated. Similarly, a data matrix for control subjects was created, and the set of principal components associated with non-PD subjects was calculated. Once calculated from the training set, these two sets of principal components were fixed and used for the classification method.

[0118] The frequency band selection starts from 2.5-8Hz and gradually increases the bandwidth to 2.5-100Hz. For each filter, the LPC is changed from 2 to 10. For each channel, all possible combinations of these parameters are applied and the average accuracy of a single round of 10-fold cross validation is obtained. Using the dataset available on the Internet, the optimal hyperparameters L and n are selected, and the parameters with the highest accuracy are selected for further analysis.

[0119] The above method finally obtains the optimal filtering parameters corresponding to different channels, as well as the number of principal components n and the number of LPC coefficients L of principal component analysis. The channel with the highest single accuracy is selected as the channel for evaluation and classification.

[0120] The whole experiment requires the subject to wear an EEG collection cap first, and then put the arm flat on the rotating plate of the passive motion electromechanical device to collect the EEG signal of the subject in a resting state for two minutes, and upload it to the host computer for differentiation. Then, based on the results of the differentiation, the stm32 motherboard drives the motor to swing and drive the arm to move. After the various sensors (flexible pressure sensors and inertial sensors) collect the data, they are transmitted to the host computer to calculate the mechanical impedance value, so as to evaluate the degree of myotonia.

[0121] Traditional measurements are driven by the doctor's manual force, while the passive motion electromechanical device of the present application is driven by a motor, which is more objective; the previous arm swing direction was vertical, while the passive motion electromechanical device of the present application has a horizontal arm swing direction during detection, which can avoid the influence of gravity; the EEG signal is first used to determine whether the patient suffers from Parkinson's disease, rather than muscle rigidity caused by other diseases, thereby improving the practicality of the equipment.

[0122] The above is a detailed introduction to a system for evaluating myotonia symptoms in patients with Parkinson's disease provided by the present application. The various embodiments in the specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same and similar parts between the various embodiments can be referenced to each other. It should be pointed out that for ordinary technicians in this technical field, without departing from the principles of the present application, several improvements and modifications can be made to the present application, and these improvements and modifications also fall within the scope of protection of the claims of the present application.

[0123] It should also be noted that, in this specification, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprise", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the statement "comprises a ..." does not exclude the presence of other identical elements in the process, method, article or device including the element.

Claims

1. A system for assessing myotonia symptoms in patients with Parkinson's disease. It is characterized in that include: A mechanical impedance data acquisition module, used to acquire pressure data, angle and angular velocity detected by a passive motion electromechanical device; An EEG signal acquisition module, used to acquire EEG signals of the subject in a resting state and when the passive motion electromechanical device drives the subject's arm to move passively; A processing module, used for determining whether the patient has Parkinson's disease based on the EEG signal; If it is determined to be Parkinson's disease, a mechanical impedance value is determined according to the pressure data, the angle and the angular velocity, and a myotonia grade is determined according to the mechanical impedance value.

2. The system for evaluating myotonia symptoms in patients with Parkinson's disease according to claim 1, It is characterized in that The processing module comprises: A first determining submodule, configured to determine an angular acceleration according to the angular velocity; The second determination submodule is used to determine the viscosity coefficient and the elastic coefficient in the first preset formula by using the least square method, and the first preset formula is: ; The third determination submodule is used to determine the mechanical impedance value according to a second preset formula, where the second preset formula is: ; in, is the mechanical impedance value, is the torque, is the moment of inertia coefficient, is the viscosity coefficient, is the elastic coefficient, is the angular acceleration, is the angular velocity, is the angle, is a constant, and the torque is calculated from the pressure data applied to the wrist of the object being measured and the distance between the force application point and the rotation axis.

3. The system for evaluating myotonia symptoms in patients with Parkinson's disease according to claim 1, It is characterized in that The processing module comprises: An optimization submodule, used for constructing an autoregressive model and optimizing model parameters of the autoregressive model using prediction errors; The determination submodule is used to input the EEG signal into the optimized autoregressive model, and determine whether it is Parkinson's disease according to the output result of the optimized autoregressive model.

4. The system for evaluating myotonia symptoms in patients with Parkinson's disease according to claim 3, It is characterized in that The processing module also includes: A fourth determination submodule, used to determine a number of sampling points of the EEG signal and sampling data corresponding to the sampling points; The fifth determination submodule is used to set an extraction coefficient for extracting features, and determine a forward prediction error and a backward prediction error according to the sampled data and the extraction coefficient.

5. The system for evaluating myotonia symptoms in patients with Parkinson's disease according to claim 4, It is characterized in that The optimization submodule includes: An initialization unit, used to initialize a forward prediction error and a backward prediction error; A reflection coefficient solving unit is used to call the Burg method to recursively solve the reflection coefficient by minimizing the mean square value of the forward prediction error and the backward prediction error; A model parameter determination unit is used to determine the model parameters of the autoregressive model according to the reflection coefficient.

6. The system for evaluating myotonia symptoms in patients with Parkinson's disease according to claim 5, It is characterized in that The reflection coefficient solving unit comprises: A calculation subunit, used for calculating the current order reflection coefficient according to the current order forward prediction error and the current order backward prediction error in each order iteration; A determination subunit, configured to determine an updated forward prediction error and an updated backward prediction error using the current-order reflection coefficient; The replacement subunit is used to replace the current-order forward prediction error with the updated forward prediction error, and replace the current-order backward prediction error with the updated backward prediction error, and enter the step of calculating the current-order reflection coefficient based on the current-order forward prediction error and the current-order backward prediction error, until the reflection coefficient of each iteration is calculated.

7. The system for evaluating myotonia symptoms in patients with Parkinson's disease according to claim 3, It is characterized in that The determination submodule comprises: A determination unit, used for determining whether the disease is Parkinson's disease according to the classification result of the classifier in the autoregressive model; Wherein, the determination unit comprises: A weight coefficient determination subunit, used to determine the weight coefficient of the linear prediction model when the total energy error is minimized in each iteration process; A generating subunit, configured to generate a feature vector according to the weight coefficient; the feature vector includes a first feature vector of a Parkinson's disease patient research object and a second feature vector of a healthy person research object; A classifier construction subunit performs principal component analysis on different channels, determines a hyperparameter selection of the number of principal components, and constructs a classifier according to the hyperparameter selection and the feature vector; An unbiased eigenvector determining subunit, used to determine a scaled unbiased eigenvector according to the eigenvector, the number of research objects and a basis vector of the eigenvector; A singular value decomposition subunit, used for performing singular value decomposition on the unbiased eigenvector to obtain a singular vector and a singular value; A hyperplane distance determination subunit is used to select a set number of singular values ​​as hyperplane segmentation and calculate the hyperplane distance constructed by the principal component; The classification result determination subunit is used to determine the classification result according to the hyperplane distance.

8. The system for evaluating myotonia symptoms in patients with Parkinson's disease according to claim 1, It is characterized in that The passive motion electromechanical device includes: a base, a motor, a rotating plate, a sliding plate, a joint fixing assembly, a main board, a flexible pressure sensor and an inertial sensor. The motor is arranged on the base, and the motor is connected to the bottom of the rotating plate and is used to drive the rotating plate to rotate in a horizontal direction. A sliding groove is arranged in the length direction of the rotating plate, and the sliding plate is slidably connected to the sliding groove. The joint fixing assembly is arranged on the sliding plate, and the joint fixing assembly is used to fix the arm of the object to be measured. The flexible pressure sensor is arranged on the joint fixing assembly, and the inertial sensor is arranged at the bottom of the rotating plate. The flexible pressure sensor is used to detect the pressure data, and the inertial sensor is used to detect the angular velocity and the angle of the rotating plate.

9. The system for evaluating myotonia symptoms in patients with Parkinson's disease according to claim 8, It is characterized in that The joint fixation assembly comprises a joint fixation plate and an elastic band. The joint fixation plate is a U-shaped fixation plate. The bottom of the U-shaped fixation plate is connected to the sliding plate, and the two ends of the U-shaped fixation plate are respectively connected to the two ends of the elastic band.

10. The system for evaluating myotonia symptoms in patients with Parkinson's disease according to claim 8, It is characterized in that It also includes a limiting rod, the sliding plate is provided with a screw hole, one end of the limiting rod is threadedly connected to the screw hole, and passes through the sliding plate through the screw hole to abut against the sliding groove.

Images