Adaptive feedback system and method for Parkinson's disease

Abstract

The present invention provides an adaptive feedback system and method for Parkinson's disease. The system includes: a central coordinator, a data acquisition system, a stimulation generation system, a stimulation device and a data processing system. The data acquisition system is used to collect data from a target user, obtain user data, and send the user data to the data processing system; the data processing system is used to perform data analysis on the user data to obtain a user analysis result; the stimulation generation system is used to determine stimulation information according to the user analysis result; the central coordinator is used to determine a stimulation signal according to the stimulation information, and control the stimulation device to perform electrical stimulation on the target user according to the stimulation signal. The present invention can automatically determine the stimulation information based on the user analysis result, and can automatically generate a stimulation signal by sending the stimulation information to the central coordinator. Based on the stimulation signal, the stimulation device can automatically control the stimulation device to perform electrical stimulation on the target user, so as to achieve the effect of adaptive feedback for Parkinson's disease.

Description

Technical Field

[0001] The present invention relates to the technical field of Parkinson's disease adaptive feedback, and in particular to an adaptive feedback system and method for Parkinson's disease. Background Art

[0002] Parkinson's disease (PD) is a common degenerative neurological disease. It is a neurodegenerative disease with a long course and a prolonged course. It has a heavy medical burden and seriously affects the healthy development of the country's economy and society. Since the beginning of the disease, the symptoms gradually worsen, resulting in a sharp decline in the patient's quality of life and a heavy burden on the patient and his family. Therefore, people pay more and more attention to the information feedback on Parkinson's disease.

[0003] In the existing Parkinson's disease diagnosis information feedback process, doctors generally use manual methods to check and feedback user medical information, resulting in low efficiency in Parkinson's disease feedback. Summary of the invention

[0004] The purpose of the embodiments of the present invention is to provide an adaptive feedback system and method for Parkinson's disease, aiming to solve the problem of low efficiency of existing feedback for Parkinson's disease.

[0005] The embodiment of the present invention is implemented as follows: an adaptive feedback system for Parkinson's disease, the system comprising:

[0006] A central coordinator, a data acquisition system, a stimulation generation system and a stimulation device respectively connected to the central coordinator in communication, wherein a data output end of the data acquisition system is provided with a data processing system, and a data output end of the data processing system is connected to the stimulation generation system in communication;

[0007] The data collection system is used to collect data from target users, obtain user data, and send the user data to the data processing system;

[0008] The data processing system is used to perform data analysis on the user data to obtain user analysis results, and send the user analysis results to the stimulus generation system;

[0009] The stimulation generation system is used to determine stimulation information according to the user analysis result, and send the stimulation information to the central coordinator;

[0010] The central coordinator is used to determine a stimulation signal according to the stimulation information, and control the stimulation device to perform electrical stimulation on the target user according to the stimulation signal.

[0011] Preferably, the user data includes handwriting information, and the data processing system is further used for:

[0012] Performing burr detection on the handwriting in the handwriting information to obtain a burr detection score;

[0013] Classifying each handwriting respectively to obtain a handwriting type, and determining a handwriting writing rule according to the handwriting type of each handwriting;

[0014] Perform writing detection on the corresponding handwritings according to the handwriting writing rules to obtain writing scores;

[0015] The behavior analysis result of the target user is determined according to the writing score and the burr detection score, and the user analysis result is generated according to the behavior analysis result.

[0016] Preferably, the user data also includes physiological and biochemical data, and the data processing system is further used for:

[0017] Perform physiological data analysis on the target user according to the physiological and biochemical data to obtain a physiological analysis result, and store the physiological analysis result in the user analysis result.

[0018] Preferably, the stimulation generation system is further used to: determine the stimulation waveform, stimulation intensity, stimulation frequency and pulse width respectively according to the behavior analysis result and the physiological analysis result;

[0019] The stimulation information is generated according to the stimulation waveform, the stimulation intensity, the stimulation frequency, and the pulse width.

[0020] Preferably, the stimulation device comprises a stimulation system and an earphone device electrically connected to the stimulation system;

[0021] The stimulation system is used to transmit a constant current to the headphone device according to the stimulation signal;

[0022] The earphone device is used to output corresponding constant current pulse stimulation to the target user according to the received constant current.

[0023] Preferably, the stimulation system comprises a CPU processing unit, and a battery and a constant current control circuit electrically connected to the CPU processing unit respectively, and an output end of the constant current control circuit is electrically connected to the headphone device;

[0024] An impedance detection circuit is provided between the CPU processing unit and the headphone device;

[0025] The impedance detection circuit is used to detect the equivalent voltage applied by the constant current control circuit on the headphone device, and send the detected equivalent voltage to the CPU processing unit. The equivalent voltage is used to determine whether the headphone device is detached.

[0026] Preferably, the stimulation system also includes a waveform control circuit electrically connected to the constant current control circuit, and the waveform control circuit is used to convert the pulse waveform output by the constant current control circuit into a plurality of pulse waveforms with different output voltages, and the voltage of the latter pulse waveform is smaller than the voltage of the previous pulse waveform.

[0027] Preferably, the data processing system is also used for:

[0028] Extracting features from the physiological and biochemical data to obtain physiological features, and performing nonlinear mapping on the physiological features to obtain feature vectors;

[0029] Performing principal component analysis and linear discriminant analysis on the features to obtain statistical features and wavelet packet features;

[0030] The statistical features and the wavelet packet features are input into a classification model based on the minimum correlation entropy optimization criterion for identification and prediction to obtain the physiological analysis results.

[0031] Preferably, the data processing system is also used for:

[0032] Acquire the background area of ​​the handwriting image in the handwriting information, and perform a closing operation on the background area to obtain a corrosion area;

[0033] Acquire the complement between the background area and the corrosion area according to a preset operator to obtain a complement area, and perform an opening operation on the complement area to obtain a burr area;

[0034] Obtaining the area sum between each burr area to obtain the burr area value, and obtaining the area sum between each writing handwriting to obtain the handwriting area value;

[0035] The burr detection score is determined according to the burr area value and the handwriting area value.

[0036] Another object of an embodiment of the present invention is to provide an adaptive feedback method for Parkinson's disease, the method comprising:

[0037] Collecting data from target users to obtain user data, and performing data analysis on the user data to obtain user analysis results;

[0038] Determining stimulation information according to the user analysis result, and determining a stimulation signal according to the stimulation information;

[0039] The stimulation device is controlled according to the stimulation signal to perform electrical stimulation on the target user.

[0040] The embodiments of the present invention can quickly analyze whether the user is at risk of Parkinson's disease by sending user data to a data processing system for data analysis, and can automatically determine stimulation information based on the user analysis results. By sending the stimulation information to a central coordinator, a stimulation signal can be automatically generated. Based on the stimulation signal, the stimulation device can be automatically controlled to electrically stimulate the target user, so as to achieve the effect of adaptive feedback for Parkinson's disease. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 is a schematic structural diagram of an adaptive feedback system for Parkinson's disease provided by a first embodiment of the present invention;

[0042] Figure 2 is a schematic structural diagram of a stimulation device for Parkinson's disease provided by a second embodiment of the present invention;

[0043] Figure 3 is a flow chart of an adaptive feedback method for Parkinson's disease provided by a third embodiment of the present invention;

[0044] Figure 4 It is a schematic diagram of the structure of a terminal device provided in the fourth embodiment of the present invention. DETAILED DESCRIPTION

[0045] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0046] In order to illustrate the technical solution of the present invention, a specific embodiment is provided below for illustration.

[0047]

[0048]

[0049] Embodiment 1

[0050] See also Figure 1 , is a schematic diagram of the structure of the adaptive feedback system for Parkinson's disease provided by the first embodiment of the present invention, comprising: a central coordinator 10, a data acquisition system 11, a stimulation generation system 12 and a stimulation device 13 which are respectively connected to the central coordinator 10 for communication, a data output end of the data acquisition system 11 is provided with a data processing system 14, and a data output end of the data processing system 14 is connected to the stimulation generation system 12 for communication.

[0051] The data acquisition system 11 is used to collect data from the target user, obtain user data, and send the user data to the data processing system 14; the user data includes physiological and biochemical data, demographic data, medical imaging data, scale data, handwriting and other information, the demographic data includes age, gender, family situation, genetic history and other information, the medical imaging data includes spiral CT scan (PETCT), arterial spin labeling magnetic resonance (ASL), magnetoencephalogram, electronic computer tomography (CT) and nuclear magnetic resonance imaging (MRI) and other information, the scale data includes Parkinson's disease unified rating scale. The physiological and biochemical data includes biochemical detection information such as blood oxygen, heart rate, electromyography, electroencephalography, serum, cerebrospinal fluid, etc.

[0052] Furthermore, the data collection system 11 is provided with a plurality of data collection devices, and the data collection devices can be configured according to the needs of the user. For example, the handwriting information can be collected by using a touch screen.

[0053] In this embodiment, the data processing system 14 is used to perform data analysis on the user data to obtain the user analysis results, and send the user analysis results to the stimulation generation system 12, and evaluate whether the target user has Parkinson's disease by performing data analysis on the user data. The stimulation generation system 12 is used to determine stimulation information according to the user analysis results, and send the stimulation information to the central coordinator 10; the central coordinator 10 is used to determine the stimulation signal according to the stimulation information, and control the stimulation device 13 to perform electrical stimulation on the target user according to the stimulation signal.

[0054] Preferably, the user data includes handwriting information, and the data processing system 14 is further used for:

[0055] Performing burr detection on the handwriting in the handwriting information to obtain a burr detection score;

[0056] The handwriting information can be obtained based on the touch screen, by displaying standard information to the target user and prompting the human eye to be detected to write information on the touch screen, so as to obtain the handwriting information. The standard information includes standard handwriting and / or standard graphics, and the standard handwriting and standard graphics can be set according to the needs;

[0057] In this step, when the target user's behavior is abnormal, the target user's hand may tremble uncontrollably. Therefore, by performing burr detection on the handwriting in the handwriting information, a burr detection score is obtained. The tremor degree of the target user during the handwriting process can be effectively evaluated based on the burr detection score. Based on the tremor degree of the target user during the handwriting process, it can be effectively determined whether the target user's behavior is abnormal. For example, when it is detected that the tremor degree of the target user during the handwriting process is high, it is determined that the target user may have Parkinson's disease.

[0058] Classifying each handwriting respectively to obtain a handwriting type, and determining a handwriting writing rule according to the handwriting type of each handwriting;

[0059] Wherein, by classifying each handwriting respectively, obtaining the handwriting type, and effectively determining the handwriting rules of each handwriting based on the handwriting type, optionally, in this step, classifying each handwriting respectively, obtaining the handwriting type, and determining the handwriting rules according to the handwriting type of each handwriting, includes:

[0060] Obtaining handwriting identifications of each handwriting respectively, and determining the handwriting type of each handwriting according to the handwriting identifications;

[0061] Among them, the handwriting identifier is used to represent the position area of ​​the handwriting on the handwriting image in the handwriting information, or to represent the writing time of the handwriting. In this step, the handwriting type of each handwriting can be determined based on the position area or writing time.

[0062] For example, when the position area of ​​the handwriting on the handwriting image is in the first preset area, the handwriting type of the handwriting is a text type; when the position area of ​​the handwriting on the handwriting image is in the second preset area, the handwriting type of the handwriting is a graphic type; the area positions of the first preset area and the second preset area can be set according to demand; when the handwriting type of the handwriting is a text type, it is determined that the handwriting is text; and when the handwriting type of the handwriting is a graphic type, it is determined that the handwriting is a graphic.

[0063] For example, the writing time of each handwriting is obtained respectively, the handwriting type of the handwriting within the first preset time length is determined as a text type, and the handwriting type of the handwriting outside the first preset time length is determined as a graphic type. The time value of the first preset time length can be set according to needs.

[0064] Matching the handwriting type of each handwriting with a pre-stored rule query table to obtain the handwriting writing rule; wherein the rule query table stores the corresponding relationship between different handwriting types and corresponding handwriting writing rules;

[0065] Perform writing detection on the corresponding handwritings according to the handwriting writing rules to obtain writing scores;

[0066] Among them, the corresponding handwriting is detected according to the handwriting rules to obtain the writing score. Based on the writing score, the accuracy of the target user's handwriting can be effectively evaluated, which improves the accuracy of behavior detection. Based on the accuracy of the target user's handwriting, it can be effectively judged whether there is any abnormality in the target user's behavior.

[0067] Optionally, in this step, performing writing detection on the corresponding handwritings according to the handwriting writing rules to obtain a writing score includes:

[0068] If the handwriting type is a text type, then performing font recognition on the handwriting to obtain the handwriting font;

[0069] Among them, by performing font recognition on the handwriting, the handwriting font that represents the writing habits of the target user can be effectively recognized. For example, the detected handwriting font may be Kaiti, Songti or Fangsongti, etc.;

[0070] Acquire the standard handwriting corresponding to the written handwriting, and convert the font of the standard handwriting according to the written font;

[0071] Among them, due to the different handwriting styles of different users, there may be style differences between the handwriting and the standard handwriting. Therefore, in this step, the standard handwriting is converted into a font by using the handwriting font, which effectively improves the accuracy of the subsequent similarity recognition between the handwriting and the standard handwriting;

[0072] Performing similarity recognition on the handwriting and the standard handwriting after the font transformation to obtain handwriting similarity, and determining the writing score according to the handwriting similarity;

[0073] Among them, by performing similarity recognition between the handwriting and the standard handwriting after font conversion, the accuracy of the handwriting of the target user can be effectively determined.

[0074] Furthermore, in this step, after performing similarity recognition on the written handwriting and the standard handwriting after the font conversion to obtain the handwriting similarity, the method further includes:

[0075] Segment the bending area and the end stroke area of ​​each handwriting respectively to obtain a bending image and an end stroke image;

[0076] Among them, by separately segmenting the bending regions of each writing stroke, the images where the Chinese character strokes in each writing stroke are bent can be effectively obtained. By separately segmenting the end-stroke regions of each writing stroke, the images of the end strokes of Chinese characters in each writing stroke can be effectively obtained. For example, when the Chinese character in the writing stroke is "乙", "乙" consists of a horizontal part, a folded part, and a bent part. Then, the obtained bending image includes the image of the connection region between the corresponding strokes of the horizontal part and the folded part, and the image of the middle section region of the corresponding stroke of the bent part. The obtained tail image includes the image of the ending region of the corresponding stroke of the bent part;

[0077] Separate the bending trajectories in each bending image, and perform a bending degree detection on the bending trajectories to obtain a bending degree score;

[0078] Among them, separately obtain the part identifiers of each stroke in each bending image, and according to each, the part identifier is used to represent the stroke type of the corresponding stroke. The stroke type includes types such as horizontal (一), vertical (丨), left-falling stroke (丿), dot (丶), horizontal fold (乛), and hook (亅). When multiple strokes are stored in the bending image, then combine the part identifiers of each stroke in the same bending image to obtain a part identifier group;

[0079] In this step, when there is only one stroke in the bending image, then match the part identifier of the stroke with a pre-stored bending degree query table to obtain a target bending angle. When there are multiple strokes in the bending image, then match the part identifier group corresponding to the bending image with the bending degree query table to obtain the target bending angle. Different part identifiers, different part identifier groups, and the corresponding relationships with the corresponding target bending angles are stored in the bending degree query table;

[0080] Specifically, in this step, separately obtain the bending angles of the strokes in each bending image to obtain the trajectory bending angle, and for the same bending image, separately calculate the angle difference between the trajectory bending angle and the corresponding target bending angle, and generate a bending degree score according to the angle difference;

[0081] Separate the end-stroke trajectories in each end-stroke image, and perform a pen-stop detection on the end-stroke trajectories to obtain a writing pen-stop score;

[0082] Among them, perform a pen-stop detection on the end-stroke trajectory to detect whether there is an abnormality at the last stroke of the target user. In this step, by separately calculating the smoothness of each end-stroke trajectory, and determining the abnormality value of each end-stroke trajectory based on the smoothness, calculate the sum of the abnormality values between each end-stroke trajectory to obtain the writing pen-stop score. In this step, when the end-stroke trajectory is smoother, the abnormality value of the end-stroke trajectory is lower. When the end-stroke trajectory is less smooth, the abnormality value of the end-stroke trajectory is higher;

[0083] The writing score is determined according to the handwriting similarity, the curvature score and the writing stop score.

[0084] Furthermore, in this step, the similarity recognition between the handwriting and the standard handwriting after the font transformation is performed to obtain the handwriting similarity includes:

[0085] Respectively obtaining the graphic feature sequence of the handwriting and the standard handwriting after the font transformation, and obtaining the handwriting feature sequence and the standard feature sequence; wherein the handwriting feature sequence and the standard feature sequence are obtained by respectively obtaining the temporal features of the handwriting and the standard handwriting after the font transformation;

[0086] Inputting the writing feature sequence and the standard feature sequence into a pre-trained convolutional recurrent neural network for sequence optimization;

[0087] Calculating the distance between the writing feature sequence after sequence optimization and the standard feature sequence to obtain the handwriting similarity;

[0088] Among them, the handwriting similarity can be calculated using the Euclidean distance formula. When the handwriting similarity is greater, the target user's handwriting is more correct. When the handwriting similarity is smaller, the target user's handwriting is more incorrect, that is, the target user's behavior is more abnormal.

[0089] Preferably, in this step, the step of performing writing detection on the corresponding handwritings according to the handwriting writing rules to obtain a writing score includes:

[0090] If the handwriting type is a graphic type, obtaining a standard graphic corresponding to the handwriting, and performing similarity recognition between the handwriting and the standard graphic to obtain handwriting similarity;

[0091] Obtaining the graphic features of the standard graphic, and performing feature detection on the handwriting according to the graphic features to obtain a trajectory feature value; wherein the graphic features include the number of graphic vertex angles, the shape of the graphic, the angle of the graphic vertex angle, etc. In this step, the trajectory feature value is obtained by performing feature comparison on the number of graphic vertex angles, the shape of the graphic, the angle of the graphic vertex angle, etc. of the standard graphic with the number of graphic vertex angles, the shape of the graphic, the angle of the graphic vertex angle, etc. of the corresponding handwriting;

[0092] Determining the writing score according to the trajectory feature value and the handwriting similarity;

[0093] The smaller the trajectory feature value is, the more similar the features between the handwriting and the standard pattern are; the larger the trajectory feature value is, the less similar the features between the handwriting and the standard pattern are. In this step, the writing score is obtained by calculating the sum of the trajectory feature value and the handwriting similarity.

[0094] Determine a behavior analysis result of the target user according to the writing score and the burr detection score, and generate the user analysis result according to the behavior analysis result;

[0095] In this embodiment, by acquiring the handwriting information of the target user, burr detection and writing detection are performed on the handwriting in the handwriting information to obtain the behavior detection result of the target user. There is no need to use artificial experience to perform behavior detection, which improves the accuracy of behavior detection. By performing burr detection on the handwriting in the handwriting information, a burr detection score is obtained. Based on the burr detection score, the degree of jitter of the target user during the handwriting writing process can be effectively evaluated. Based on the degree of jitter of the target user during the handwriting writing process, it can be effectively determined whether the behavior of the target user is abnormal. According to the handwriting writing rules, writing detection is performed on the corresponding handwriting to obtain a writing score. Based on the writing score, the accuracy of the target user's handwriting can be effectively evaluated, which improves the accuracy of behavior detection. Based on the accuracy of the target user's handwriting, it can be effectively determined whether the behavior of the target user is abnormal.

[0096] Preferably, the user data also includes physiological and biochemical data, and the data processing system 14 is also used to: perform physiological data analysis on the target user based on the physiological and biochemical data to obtain physiological analysis results, and store the physiological analysis results in the user analysis results, wherein, by performing physiological data analysis on the target user, the effect of analyzing whether the target user has Parkinson's disease from a physiological perspective can be achieved.

[0097] Furthermore, the stimulation generation system 12 is also used to: determine the stimulation waveform, stimulation intensity, stimulation frequency and pulse width according to the behavior analysis results and the physiological analysis results, and generate stimulation information according to the stimulation waveform, stimulation intensity, stimulation frequency and pulse width.

[0098] Furthermore, in this embodiment, the data processing system 14 is also used to: extract features from physiological and biochemical data to obtain physiological features, and perform nonlinear mapping on the physiological features to obtain feature vectors; perform principal component analysis and linear discriminant analysis on the features to obtain statistical features and wavelet packet features; input the statistical features and wavelet packet features into a classification model based on the minimum correlation entropy optimization criterion for identification and prediction to obtain physiological analysis results. Specifically, multimodal physiological data such as heart rate, blood oxygen, EEG, and EMG are collected, and the features of these physiological signals are extracted based on deterministic learning and nonlinear dynamics. Based on an unsupervised autoencoder model, the problems of high dimensionality, strong noise, heterogeneity, and incompleteness of multimodal data may be solved, and a variety of nonlinear mappings are combined to construct a more abstract multimodal data feature learning model.

[0099] Preferably, the data processing system 14 is further used to: obtain a background area of ​​the handwriting image in the handwriting information, and perform a closing operation on the background area to obtain a corrosion area;

[0100] Acquire the complement between the background area and the corrosion area according to a preset operator to obtain a complement area, and perform an opening operation on the complement area to obtain a burr area;

[0101] Obtaining the area sum between each burr area to obtain the burr area value, and obtaining the area sum between each writing handwriting to obtain the handwriting area value;

[0102] The burr detection score is determined according to the burr area value and the handwriting area value; wherein, by performing a closing operation on the background area, the hollow area in the background area can be effectively filled, and the adjacent objects in the background area can be connected to obtain the erosion area, and by obtaining the complement between the background area and the erosion area, the area corresponding to the burr in the handwriting image can be effectively obtained, and by performing an opening operation on the complement area, the small objects and small particle noise in the complement area can be effectively eliminated.

[0103] In this embodiment, by sending user data to the data processing system 14 for data analysis, it is possible to quickly analyze whether the user is at risk of Parkinson's disease, and automatically determine stimulation information based on the user analysis results. By sending the stimulation information to the central coordinator 10, a stimulation signal can be automatically generated, and based on the stimulation signal, the stimulation device 13 can be automatically controlled to electrically stimulate the target user, so as to achieve the effect of adaptive feedback for Parkinson's disease.

[0104] Embodiment 2

[0105] See also Figure 2 , is a schematic structural diagram of a stimulation device 13 provided in a second embodiment of the present invention, wherein the stimulation device 13 comprises a stimulation system 15 and an earphone device 16 electrically connected to the stimulation system 15;

[0106] The stimulation system 15 is used to transmit a constant current to the earphone device 16 according to the stimulation signal, and the earphone device 16 is used to output a corresponding constant current pulse stimulation to the target user according to the received constant current.

[0107] Specifically, the stimulation system 15 includes a CPU processing unit and a battery and a constant current control circuit electrically connected to the CPU processing unit respectively, and the output end of the constant current control circuit is electrically connected to the headphone device 16; an impedance detection circuit is arranged between the CPU processing unit and the headphone device 16; the impedance detection circuit is used to detect the equivalent voltage applied to the headphone device 16 by the constant current control circuit, and send the detected equivalent voltage to the CPU processing unit, and the equivalent voltage is used to determine whether the headphone device 16 is detached.

[0108] Specifically, in this embodiment, the function of the impedance detection circuit is to enable the CPU processing unit to receive the detection voltage signal and determine whether the headphone device 16 has fallen off from the human body. The working principle of the impedance detection is that the impedance detection circuit periodically detects the equivalent voltage applied to the headphone device 16 by the constant current control circuit. If the equivalent voltage is 0, it means that the headphone device 16 is not connected to the human body, that is, the impedance is infinite, and it can be determined that the headphone device 16 has been detached. Otherwise, if the detected equivalent voltage is significantly not 0, the headphone device 16 has not been detached.

[0109] Preferably, an isolation input circuit is provided between the impedance detection circuit and the CPU processing unit, and the isolation input circuit is used to isolate the equivalent voltage input by the impedance detection circuit to the CPU processing unit; an isolation output circuit is provided between the constant current control circuit and the CPU processing unit, and the isolation output circuit is used to isolate the current stimulation signal output by the CPU processing unit to the constant current control circuit.

[0110] Furthermore, the stimulation system 15 also includes a waveform control circuit electrically connected to the constant current control circuit, and the waveform control circuit is used to convert the pulse waveform output by the constant current control circuit into a plurality of pulse waveforms with different output voltages, and the voltage of the latter pulse waveform is less than the voltage of the previous pulse waveform. The waveform control circuit is arranged in the waveform generating device, and the waveform controlled by the waveform control circuit has the following characteristics: by starting with a narrow high-frequency spike component, the skin impedance is quickly reduced without causing skin pain, so that the transcutaneous stimulation current can enter the subcutaneous vagus nerve at a lower voltage, so as to achieve the purpose of effectively stimulating the subcutaneous ear vagus nerve without causing pain. This method can stimulate the vagus nerve more effectively, improve the treatment effect, and reduce pain at the same time.

[0111] In this embodiment, the constant current control circuit can effectively transmit a constant current to the headphone device 16, so that the headphone device 16 can output a corresponding constant current pulse according to the received constant current, so as to achieve the effect of outputting a constant current pulse, so that the stimulation current does not change due to the different skin impedances of different people. The constant current output also suppresses the peak pulses, prevents the occurrence of tingling, and improves the user experience.

[0112] Embodiment 3

[0113] See also Figure 3 , is a flow chart of an adaptive feedback method for Parkinson's disease provided by a third embodiment of the present invention. The adaptive feedback method for Parkinson's disease can be applied to any terminal device or system. The adaptive feedback method for Parkinson's disease comprises the steps of:

[0114] Step S10, collecting data from target users to obtain user data, and performing data analysis on the user data to obtain user analysis results;

[0115] Among them, by performing physiological analysis and behavioral analysis on user data, it is possible to analyze whether the target user has Parkinson's disease from a physiological perspective and a behavioral perspective, without the need for doctors to manually review medical data;

[0116] Step S20, determining stimulation information according to the user analysis result, and determining a stimulation signal according to the stimulation information;

[0117] wherein, according to the behavioral analysis results and the physiological analysis results, the stimulation waveform, stimulation intensity, stimulation frequency and pulse width are determined respectively, and stimulation information is generated according to the stimulation waveform, stimulation intensity, stimulation frequency and pulse width;

[0118] Step S30, controlling a stimulation device to perform electrical stimulation on the target user according to the stimulation signal;

[0119] Among them, controlling the stimulation device based on the stimulation signal to electrically stimulate the target user can effectively achieve non-invasive closed-loop neuroregulation for the target user. It will not only not cause surgical trauma to the patient, but also can use neuroregulation to reduce or even delay the onset of Parkinson's disease.

[0120] In this embodiment, by sending user data to a data processing system for data analysis, it is possible to quickly analyze whether the user is at risk of Parkinson's disease, and automatically determine stimulation information based on the user analysis results. By sending the stimulation information to a central coordinator, a stimulation signal can be automatically generated, and based on the stimulation signal, the stimulation device can be automatically controlled to electrically stimulate the target user, so as to achieve the effect of adaptive feedback for Parkinson's disease.

[0121] Embodiment 4

[0122] Figure 4 2 is a block diagram of a terminal device 2 provided in the fourth embodiment of the present application. Figure 4 As shown, the terminal device 2 of this embodiment includes: a processor 20, a memory 21, and a computer program 22 stored in the memory 21 and executable on the processor 20, such as a program of an adaptive feedback method for Parkinson's disease. When the processor 20 executes the computer program 22, the steps in each embodiment of the adaptive feedback method for Parkinson's disease described above are implemented.

[0123] Exemplarily, the computer program 22 may be divided into one or more modules, which are stored in the memory 21 and executed by the processor 20 to complete the present application. The one or more modules may be a series of computer program instruction segments capable of completing specific functions, which are used to describe the execution process of the computer program 22 in the terminal device 2. The terminal device may include, but is not limited to, a processor 20 and a memory 21.

[0124] The processor 20 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0125] The memory 21 may be an internal storage unit of the terminal device 2, such as a hard disk or memory of the terminal device 2. The memory 21 may also be an external storage device of the terminal device 2, such as a plug-in hard disk, a smart memory card (Smart Med ia Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the terminal device 2. Further, the memory 21 may also include both an internal storage unit of the terminal device 2 and an external storage device. The memory 21 is used to store the computer program and other programs and data required by the terminal device. The memory 21 may also be used to temporarily store data that has been output or is to be output.

[0126] In addition, each functional module in each embodiment of the present application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The above integrated unit can be implemented in the form of hardware or in the form of software functional unit.

[0127] If the integrated module is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Among them, the computer-readable storage medium can be non-volatile or volatile. Based on this understanding, the present application implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable storage medium may include: any entity or device capable of carrying computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-On lyMemory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the content contained in computer-readable storage media can be appropriately increased or decreased according to the requirements of legislation and patent practices in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practices, computer-readable storage media do not include electrical carrier signals and telecommunication signals.

[0128] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. An adaptive feedback system for Parkinson's disease, characterized in that: The system comprises: A central coordinator, a data acquisition system, a stimulation generation system and a stimulation device respectively connected to the central coordinator in communication, wherein a data output end of the data acquisition system is provided with a data processing system, and a data output end of the data processing system is connected to the stimulation generation system in communication; The data collection system is used to collect data from target users, obtain user data, and send the user data to the data processing system; The data processing system is used to perform data analysis on the user data to obtain user analysis results, and send the user analysis results to the stimulus generation system; The stimulation generation system is used to determine stimulation information according to the user analysis result, and send the stimulation information to the central coordinator; The central coordinator is used to determine a stimulation signal according to the stimulation information, and control the stimulation device to perform electrical stimulation on the target user according to the stimulation signal; The user data includes handwriting information, and the data processing system is further used for: Performing burr detection on the handwriting in the handwriting information to obtain a burr detection score; Classifying each handwriting respectively to obtain a handwriting type, and determining a handwriting writing rule according to the handwriting type of each handwriting; Perform writing detection on the corresponding handwritings according to the handwriting writing rules to obtain writing scores; Determine a behavior analysis result of the target user according to the writing score and the burr detection score, and generate the user analysis result according to the behavior analysis result; The user data also includes physiological and biochemical data, and the data processing system is further used for: Performing physiological data analysis on the target user according to the physiological and biochemical data to obtain a physiological analysis result, and storing the physiological analysis result in the user analysis result; The stimulation generation system is also used to: determine the stimulation waveform, stimulation intensity, stimulation frequency and pulse width respectively according to the behavior analysis result and the physiological analysis result; generating the stimulation information according to the stimulation waveform, the stimulation intensity, the stimulation frequency and the pulse width; The stimulation device comprises a stimulation system and an earphone device electrically connected to the stimulation system; The stimulation system is used to transmit a constant current to the headphone device according to the stimulation signal; The earphone device is used to output corresponding constant current pulse stimulation to the target user according to the received constant current; The data processing system is also used for: Acquire the background area of ​​the handwriting image in the handwriting information, and perform a closing operation on the background area to obtain a corrosion area; Acquire the complement between the background area and the corrosion area according to a preset operator to obtain a complement area, and perform an opening operation on the complement area to obtain a burr area; Obtaining the area sum between each burr area to obtain the burr area value, and obtaining the area sum between each writing handwriting to obtain the handwriting area value; The burr detection score is determined according to the burr area value and the handwriting area value.

2. The adaptive feedback system for Parkinson's disease as claimed in claim 1, characterized in that: The stimulation system comprises a CPU processing unit, and a battery and a constant current control circuit electrically connected to the CPU processing unit respectively, and an output end of the constant current control circuit is electrically connected to the headphone device; An impedance detection circuit is provided between the CPU processing unit and the headphone device; The impedance detection circuit is used to detect the equivalent voltage applied by the constant current control circuit on the headphone device, and send the detected equivalent voltage to the CPU processing unit. The equivalent voltage is used to determine whether the headphone device is detached.

3. The adaptive feedback system for Parkinson's disease as claimed in claim 1, characterized in that: The stimulation system also includes a waveform control circuit electrically connected to the constant current control circuit, and the waveform control circuit is used to convert the pulse waveform output by the constant current control circuit into a plurality of pulse waveforms with different output voltages, and the voltage of the latter pulse waveform is smaller than the voltage of the previous pulse waveform.

4. The adaptive feedback system for Parkinson's disease as claimed in claim 1, characterized in that: The data processing system is also used for: Extracting features from the physiological and biochemical data to obtain physiological features, and performing nonlinear mapping on the physiological features to obtain feature vectors; Performing principal component analysis and linear discriminant analysis on the features to obtain statistical features and wavelet packet features; The statistical features and the wavelet packet features are input into a classification model based on the minimum correlation entropy optimization criterion for identification and prediction to obtain the physiological analysis results.

Images