Parkinson's disease early diagnosis system based on handwriting behavior feature recognition

Abstract

The invention discloses an early diagnosis system for Parkinson's disease based on handwriting behavior feature recognition, which belongs to the technical field of Parkinson's disease and comprises a data acquisition module, a data processing module, a model construction module, a test optimization module, a prediction diagnosis module and an intelligent management and control module. The problems that the Parkinson's disease cannot be diagnosed in an early stage based on handwriting behavior feature recognition in the prior art, early intervention and treatment suggestions cannot be provided for patients, and the treatment effect is reduced are solved. Based on the optimal Parkinson's disease early diagnosis model, handwriting behavior real-time data is predicted, Parkinson's disease early diagnosis is carried out, the Parkinson's disease early diagnosis result is determined, a Parkinson's disease early management and control scheme is formulated, Parkinson's disease early patients are intelligently managed and controlled, and the Parkinson's disease early diagnosis and control efficiency is improved. The Parkinson's disease can be diagnosed early based on handwriting behavior feature recognition, early intervention and treatment suggestions can be provided for patients, and the treatment effect can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of Parkinson's disease, and in particular to an early diagnosis system for Parkinson's disease based on handwriting behavior feature recognition. Background Art

[0002] Parkinson's disease, also often referred to as "Parkinson's paralysis," is a neurodegenerative disease. The main cause of this disease is the degeneration and death of dopaminergic neurons in the substantia nigra, which may be related to multiple factors such as genetics, environmental factors, and aging of the nervous system. It is generally recognized that aging is the most important factor in the occurrence of Parkinson's disease. The disease has a significant high incidence in the elderly. At present, there are still great difficulties in the early diagnosis of Parkinson's disease, resulting in many patients not receiving timely treatment in the early stages of the disease, affecting the treatment effect and prognosis.

[0003] Chinese patent with publication number CN117222355A discloses a method for treating, preventing or ameliorating Parkinson's disease with putinizumab, wherein a mobile device is provided to a patient, the mobile device is programmed to receive and transmit data obtained to measure the patient's passive and / or active movement; data transmitted from the mobile device is collected; the data obtained from the patient is compared with control data to assess the presence or extent of motor deficits in the individual, and / or the data obtained from the patient is monitored for a period of time sufficient to identify changes in the patient's active or passive motor function; however, the patent has the following defects:

[0004] The existing methods cannot perform early diagnosis of Parkinson's disease based on handwriting behavior feature recognition, cannot provide early intervention and treatment suggestions for patients, and thus reduce the treatment effect. Summary of the invention

[0005] The purpose of the present invention is to provide an early diagnosis system for Parkinson's disease based on handwriting behavior feature recognition, which can perform early diagnosis of Parkinson's disease based on handwriting behavior feature recognition, provide early intervention and treatment suggestions for patients, improve treatment effects, and solve the problems raised in the above-mentioned background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] The Parkinson's disease early diagnosis system based on handwriting behavior feature recognition includes:

[0008] A data collection module, used to collect the user's brain electrical activity change data, the user's handwriting data and the user's handwriting drawing data, and determine the handwriting behavior perception data based on the non-invasive electroencephalogram;

[0009] A data processing module, used for processing the handwriting behavior perception data based on the non-invasive electroencephalogram to determine the handwriting behavior feature data based on the non-invasive electroencephalogram;

[0010] A model building module is used to build an early diagnosis model for Parkinson's disease based on handwriting behavior feature recognition according to the needs of early diagnosis of Parkinson's disease;

[0011] The test optimization module is used to test, evaluate and optimize the Parkinson's disease early diagnosis model to determine the optimal Parkinson's disease early diagnosis model based on handwriting behavior feature recognition;

[0012] The predictive diagnosis module is used to predict the real-time data of handwriting behavior and diagnose Parkinson's disease in the early stage, and determine the early diagnosis results of Parkinson's disease based on the recognition of handwriting behavior characteristics;

[0013] The intelligent management and control module is used to formulate an early management plan for Parkinson's disease based on handwriting behavior feature recognition and to provide intelligent management and control for early Parkinson's disease patients.

[0014] Preferably, the data acquisition module includes:

[0015] An EEG acquisition unit is used to monitor and collect the user's brain function and cognitive state during the writing process based on a non-invasive EEG, and obtain the user's EEG activity change data;

[0016] A handwriting acquisition unit, used to monitor and collect the handwriting of the user during the writing process based on computer vision, and obtain the user's handwriting data;

[0017] A drawing acquisition unit, used to monitor and collect the handwritten drawings of the user during the writing process based on computer vision, and obtain the handwritten drawing data of the user;

[0018] Among them, based on the user's brain electrical activity change data, the user's handwriting data and the user's handwriting drawing data, the handwriting behavior perception data based on the non-invasive electroencephalogram is determined.

[0019] Preferably, the data acquisition module further includes:

[0020] The word count real-time collection module is used to collect the number of complete words completed per unit time during the user's writing process in real time; wherein the value range of the unit time is 2min-8min;

[0021] A module for obtaining the area occupied by characters, for obtaining the area occupied by each character completed in each unit time after each unit time is completed;

[0022] A writing index parameter acquisition module, used to acquire the writing index parameter corresponding to a unit time according to the number of complete characters of the user in each unit time and the character area occupied by each character completed in each unit time;

[0023] The writing index parameter is obtained by the following formula:

[0024]

[0025] Where R represents the writing index parameter; n represents the number of unit time experienced by the user in writing; S zi represents the total writing area corresponding to the i-th unit time; S zi-1 represents the total writing area corresponding to the i-1th unit time; N i N represents the number of complete characters written in the i-th unit time; i-1 represents the number of complete written characters corresponding to the i-1th unit time; f n N represents the standard deviation of the number of complete characters written in n units of time; p S represents the average number of complete characters written in n units of time; p S represents the average value of the total writing area corresponding to n unit time; dij represents the area occupied by the jth complete word in the i-th unit time;

[0026] The adjustment module is used to adjust the time interval for collecting handwriting using the difference value of the writing index parameter between every two adjacent unit times.

[0027] Preferably, the adjustment module includes:

[0028] A writing index parameter extraction module, used to extract writing index parameters of every two adjacent unit times;

[0029] A difference value acquisition module, used for acquiring a difference value of the writing index parameters of every two adjacent unit times by using the writing index parameters of every two adjacent unit times;

[0030] A first group of writing index parameter acquisition module is used to extract the writing index parameters of two unit times corresponding to the maximum value of the difference between the difference values ​​of the writing index parameters of each two adjacent unit times as the first group of writing index parameters;

[0031] A second group of writing index parameter acquisition module is used to extract the writing index parameters of two unit times corresponding to the minimum difference value among the difference values ​​of the writing index parameters of each two adjacent unit times as the second group of writing index parameters;

[0032] The acquisition module for adjusting the collection time interval is used to obtain the collection time interval adjustment coefficient by using the first group of writing index parameters and the second group of writing index parameters, wherein the collection time interval adjustment coefficient is obtained by the following formula:

[0033]

[0034] Wherein, K represents the collection time interval adjustment coefficient; k represents the number of writing index parameters other than the first group of writing index parameters and the second group of writing index parameters; R i represents the value corresponding to the i-th writing index parameter except the first group of writing index parameters and the second group of writing index parameters; C max and C min Respectively represent the maximum and minimum difference values ​​of the difference values ​​of the writing index parameters between every two adjacent unit times; R max01 and R max02 They represent the writing index parameters of the first group of writing index parameters corresponding to two unit times; R min01 and R min02 Respectively represent the writing index parameters of two units of time corresponding to the second group of writing index parameters;

[0035] A comparison module, used for comparing the acquisition time interval adjustment coefficient with a preset adjustment coefficient threshold;

[0036] An adjustment execution module, configured to adjust the current handwriting collection time interval using the collection time interval adjustment coefficient when the collection time interval adjustment coefficient exceeds a preset adjustment coefficient threshold, so as to obtain an adjusted handwriting collection time interval;

[0037] The adjusted time interval for collecting handwriting is obtained by the following formula:

[0038] T=[1+log 10 (1+K)]·T0

[0039] Wherein, T represents the adjusted time interval for collecting handwriting; T0 represents the current time interval for collecting handwriting; and K represents the collection time interval adjustment coefficient.

[0040] Preferably, the data processing module includes:

[0041] A data cleaning unit, used for cleaning the handwriting behavior sensing data;

[0042] Acquire handwriting behavior perception data based on non-invasive EEG;

[0043] Cleaning of handwriting behavior perception data based on non-invasive EEG, including:

[0044] To conduct consistency checks on non-invasive EEG-based handwriting behavior perception data;

[0045] According to the reasonable value range and mutual relationship of each variable in the handwriting behavior perception data based on non-invasive EEG, check whether the handwriting behavior perception data based on non-invasive EEG meets the requirements;

[0046] Remove inconsistent data that is beyond the normal range, logically unreasonable or contradictory in the handwriting behavior perception data based on non-invasive EEG;

[0047] Process invalid and missing values ​​of handwriting behavior perception data based on non-invasive EEG;

[0048] Remove invalid and missing data from handwriting behavior perception data based on non-invasive EEG that are not valuable for early diagnosis of Parkinson's disease;

[0049] Identify handwriting behavior perception data that is valuable for early diagnosis of Parkinson's disease.

[0050] Preferably, the data processing module further includes:

[0051] A data conversion unit, used for converting the cleaned handwriting behavior perception data;

[0052] Obtain handwriting behavior perception data that is valuable for early diagnosis of Parkinson's disease after cleaning;

[0053] Convert handwriting behavior perception data that is valuable for early diagnosis of Parkinson's disease;

[0054] Eliminate dimensional differences between handwriting behavior perception data;

[0055] Determine handwriting behavior perception data with data standardization;

[0056] A feature extraction unit, used for extracting features from the converted handwriting behavior perception data;

[0057] Acquire the handwriting behavior perception data with data standardization after conversion;

[0058] Feature extraction of handwriting behavior perception data with data standardization;

[0059] Extract features that can reflect the early diagnosis of Parkinson's disease;

[0060] Determine handwriting behavior characteristic data based on non-invasive electroencephalogram.

[0061] Preferably, the model building module includes:

[0062] A data division unit, used for dividing the handwriting behavior feature data;

[0063] Obtain handwriting behavior feature data based on non-invasive EEG;

[0064] To classify the handwriting behavior feature data based on non-invasive EEG;

[0065] Determine the training data set and the test data set;

[0066] Model training unit, used to build an early diagnosis model for Parkinson's disease;

[0067] According to the demand for early diagnosis of Parkinson's disease based on handwriting behavior feature recognition, a neural network model framework suitable for early diagnosis of Parkinson's disease is selected;

[0068] Based on the training data set, the selected neural network model framework suitable for early diagnosis of Parkinson's disease is trained;

[0069] An early diagnosis model for Parkinson's disease based on handwriting behavior feature recognition was determined.

[0070] Preferably, the test optimization module includes:

[0071] A test evaluation unit, used to test and evaluate the early diagnosis model of Parkinson's disease;

[0072] Obtain an early diagnosis model for Parkinson's disease based on handwriting behavior feature recognition;

[0073] Based on the test data set, the early diagnosis model of Parkinson's disease was tested and evaluated;

[0074] Determine the test evaluation results based on the Parkinson's disease early diagnosis model;

[0075] An optimization and adjustment unit, used for optimizing and adjusting the early diagnosis model of Parkinson's disease;

[0076] Obtain test evaluation results based on the Parkinson's disease early diagnosis model;

[0077] Mining and analyzing the test evaluation results based on the Parkinson's disease early diagnosis model;

[0078] Determine the optimal adjustment plan based on the early diagnosis model of Parkinson's disease;

[0079] Optimizing and adjusting the Parkinson's disease early diagnosis model according to the optimization and adjustment plan based on the Parkinson's disease early diagnosis model;

[0080] Determine the optimal early diagnosis model for Parkinson's disease based on handwriting behavior feature recognition.

[0081] Preferably, the predictive diagnosis module includes:

[0082] A data acquisition unit, used for acquiring real-time data of handwriting behavior based on non-invasive electroencephalogram;

[0083] According to the need for early diagnosis of Parkinson's disease based on handwriting behavior feature recognition, the user's EEG activity changes, handwriting and handwriting drawings are monitored and collected in real time to obtain real-time handwriting behavior data based on non-invasive EEG;

[0084] Predictive diagnosis unit, used to predict real-time data of handwriting behavior and make early diagnosis of Parkinson's disease;

[0085] Obtain the optimal early diagnosis model for Parkinson's disease based on handwriting behavior feature recognition;

[0086] Input the real-time data of handwriting behavior based on non-invasive EEG into the optimal early diagnosis model of Parkinson's disease based on handwriting behavior feature recognition;

[0087] Based on the optimal Parkinson's disease early diagnosis model based on handwriting behavior feature recognition, the real-time data of handwriting behavior based on non-invasive EEG is predicted and Parkinson's disease is diagnosed early;

[0088] Determine the early diagnosis results of Parkinson's disease based on handwriting behavior feature recognition;

[0089] Among them, the early diagnosis result of Parkinson's disease based on handwriting behavior feature recognition includes: the user is an early Parkinson's disease patient or the user is not an early Parkinson's disease patient;

[0090] When the user is an early-stage Parkinson's disease patient, an early Parkinson's disease management and control plan is formulated to carry out intelligent management and control of early-stage Parkinson's disease patients.

[0091] Preferably, the intelligent management and control module includes:

[0092] a strategy development unit to develop early management plans for Parkinson’s disease;

[0093] Obtain early diagnosis results of Parkinson's disease based on handwriting behavior feature recognition;

[0094] Mining and analyzing the early diagnosis results of Parkinson's disease based on handwriting behavior feature recognition;

[0095] Determine an early control plan for Parkinson's disease based on handwriting behavior feature recognition;

[0096] Intelligent management and control unit, used for intelligent management and control of early-stage Parkinson's disease patients;

[0097] Obtain an early management plan for Parkinson's disease based on handwriting behavior feature recognition;

[0098] Based on the Parkinson's disease early management and control program, intelligent management and control of early Parkinson's disease patients are carried out, timely warnings are given to early Parkinson's disease patients, and personalized management and treatment measures are formulated for early Parkinson's disease patients.

[0099] Compared with the prior art, the present invention has the following beneficial effects:

[0100] The present invention collects handwriting behavior perception data based on non-invasive electroencephalogram, processes the handwriting behavior perception data, determines handwriting behavior feature data, constructs a Parkinson's disease early diagnosis model according to the needs of Parkinson's disease early diagnosis, tests, evaluates and optimizes the Parkinson's disease early diagnosis model, determines the optimal Parkinson's disease early diagnosis model, predicts handwriting behavior real-time data and performs early diagnosis of Parkinson's disease based on the optimal Parkinson's disease early diagnosis model, determines the results of Parkinson's disease early diagnosis, formulates an early management and control plan for Parkinson's disease, performs intelligent management and control of early Parkinson's disease patients, can perform early diagnosis of Parkinson's disease based on handwriting behavior feature recognition, can provide patients with early intervention and treatment suggestions, and can improve treatment effects. BRIEF DESCRIPTION OF THE DRAWINGS

[0101] Figure 1 This is a structural block diagram of the Parkinson's disease early diagnosis system based on handwriting behavior feature recognition of the present invention. DETAILED DESCRIPTION

[0102] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0103] In order to solve the existing problem that Parkinson's disease cannot be diagnosed early based on handwriting behavior feature recognition, early intervention and treatment suggestions cannot be provided to patients, and the treatment effect is reduced, please refer to Figure 1 , this embodiment provides the following technical solutions:

[0104] The early diagnosis system of Parkinson's disease based on handwriting behavior feature recognition includes: data acquisition module, data processing module, model building module, test optimization module, prediction diagnosis module and intelligent management and control module.

[0105] It should be noted that through the interactive communication between the data acquisition module, data processing module, model building module, test optimization module, predictive diagnosis module and intelligent management and control module, early diagnosis of Parkinson's disease can be made based on handwriting behavior feature recognition, and early intervention and treatment suggestions can be provided to patients, which can improve the treatment effect.

[0106] Among them, the data acquisition module is used to collect the user's brain electrical activity change data, the user's handwriting data and the user's handwriting drawing data, and determine the handwriting behavior perception data based on the non-invasive electroencephalogram;

[0107] In this embodiment, the data acquisition module includes:

[0108] An EEG acquisition unit is used to monitor and collect the user's brain function and cognitive state during the writing process based on a non-invasive EEG, and obtain the user's EEG activity change data;

[0109] It should be noted that the electroencephalogram is a graph obtained by amplifying and recording the spontaneous biopotential of the brain from the scalp through sophisticated electronic instruments. It is the spontaneous and rhythmic electrical activity of groups of brain cells recorded by electrodes. Real-time monitoring of the patient's brain function helps to understand the development and changes of the disease and guide the adjustment of treatment plans.

[0110] A handwriting acquisition unit, used to monitor and collect the handwriting of the user during the writing process based on computer vision, and obtain the user's handwriting data;

[0111] A drawing acquisition unit, used to monitor and collect the handwritten drawings of the user during the writing process based on computer vision, and obtain the handwritten drawing data of the user;

[0112] Among them, based on the user's brain electrical activity change data, the user's handwriting data and the user's handwriting drawing data, the handwriting behavior perception data based on the non-invasive electroencephalogram is determined.

[0113] The data processing module is used to process the handwriting behavior perception data based on the non-invasive electroencephalogram to determine the handwriting behavior feature data based on the non-invasive electroencephalogram;

[0114] Specifically, the data acquisition module further includes:

[0115] The word count real-time collection module is used to collect the number of complete words completed per unit time during the user's writing process in real time; wherein the value range of the unit time is 2min-8min;

[0116] A module for obtaining the area occupied by characters, for obtaining the area occupied by each character completed in each unit time after each unit time is completed;

[0117] A writing index parameter acquisition module, used to acquire the writing index parameter corresponding to a unit time according to the number of complete characters of the user in each unit time and the character area occupied by each character completed in each unit time;

[0118] The writing index parameter is obtained by the following formula:

[0119]

[0120] Where R represents the writing index parameter; n represents the number of unit time experienced by the user in writing; S zi represents the total writing area corresponding to the i-th unit time; S zi-1 represents the total writing area corresponding to the i-1th unit time; N i N represents the number of complete characters written in the i-th unit time; i-1 represents the number of complete written characters corresponding to the i-1th unit time; f n N represents the standard deviation of the number of complete characters written in n units of time; p S represents the average number of complete characters written in n units of time; p S represents the average value of the total writing area corresponding to n unit time; dij represents the area occupied by the jth complete word in the i-th unit time;

[0121] The adjustment module is used to adjust the time interval for collecting handwriting using the difference value of the writing index parameter between every two adjacent unit times.

[0122] The technical effect of the above technical solution is: through the real-time word count acquisition module, the system can monitor in real time the number of complete words completed by the user in each unit time during the writing process, thereby providing the user with instant writing progress feedback. The word area acquisition module can accurately calculate the area occupied by each word after each unit time, which provides a detailed data basis for subsequent writing analysis. The writing index parameter acquisition module uses a complex formula to comprehensively consider multiple factors (such as writing area, number of words, standard deviation, average value, etc.) to calculate the writing index parameters, making the analysis of writing behavior more comprehensive and in-depth.

[0123] The adjustment module uses the difference in writing index parameters between two adjacent units of time to dynamically adjust the time interval for handwriting collection. This means that the system can automatically optimize the collection strategy according to the user's writing speed and habits to improve the accuracy and efficiency of data collection. Through real-time monitoring and dynamic adjustment, this technical solution can better adapt to the user's writing rhythm, reduce writing interruptions or data loss caused by improper collection time intervals, and thus enhance the user's writing experience. The detailed writing data collected by this technical solution can provide valuable empirical materials for writing research and help to gain a deeper understanding of the various factors in the writing process and their interrelationships. At the same time, these data can also be used for teaching evaluation, helping teachers to better understand students' writing habits and levels, so as to develop more targeted teaching plans.

[0124] To sum up, the above-mentioned technical solution not only improves the accuracy and efficiency of handwriting collection through real-time monitoring, precise analysis, dynamic adjustment and other means, but also provides users with a better writing experience and provides strong support for writing research and teaching.

[0125] Specifically, the adjustment module includes:

[0126] A writing index parameter extraction module, used to extract writing index parameters of every two adjacent unit times;

[0127] A difference value acquisition module, used for acquiring a difference value of the writing index parameters of every two adjacent unit times by using the writing index parameters of every two adjacent unit times;

[0128] A first group of writing index parameter acquisition module is used to extract the writing index parameters of two unit times corresponding to the maximum value of the difference between the difference values ​​of the writing index parameters of each two adjacent unit times as the first group of writing index parameters;

[0129] A second group of writing index parameter acquisition module is used to extract the writing index parameters of two unit times corresponding to the minimum difference value among the difference values ​​of the writing index parameters of each two adjacent unit times as the second group of writing index parameters;

[0130] The acquisition module for adjusting the collection time interval is used to obtain the collection time interval adjustment coefficient by using the first group of writing index parameters and the second group of writing index parameters, wherein the collection time interval adjustment coefficient is obtained by the following formula:

[0131]

[0132] Wherein, K represents the collection time interval adjustment coefficient; k represents the number of writing index parameters other than the first group of writing index parameters and the second group of writing index parameters; R i represents the value corresponding to the i-th writing index parameter except the first group of writing index parameters and the second group of writing index parameters; C max and C min Respectively represent the maximum and minimum difference values ​​of the difference values ​​of the writing index parameters between every two adjacent unit times; R max01 and R max02 They represent the writing index parameters of the first group of writing index parameters corresponding to two unit times; R min01 and R min02 Respectively represent the writing index parameters of two units of time corresponding to the second group of writing index parameters;

[0133] A comparison module, used for comparing the acquisition time interval adjustment coefficient with a preset adjustment coefficient threshold;

[0134] An adjustment execution module, configured to adjust the current handwriting collection time interval using the collection time interval adjustment coefficient when the collection time interval adjustment coefficient exceeds a preset adjustment coefficient threshold, so as to obtain an adjusted handwriting collection time interval;

[0135] The adjusted time interval for collecting handwriting is obtained by the following formula:

[0136] T=[1+log 10 (1+K)]·T0

[0137] Wherein, T represents the adjusted time interval for collecting handwriting; T0 represents the current time interval for collecting handwriting; and K represents the collection time interval adjustment coefficient.

[0138] The technical effect of the above technical solution is that the writing index parameter extraction module and the difference value acquisition module can accurately extract the writing index parameters of each two adjacent unit time from a large amount of data and calculate the difference value between them. This provides key data support for the subsequent adjustment of the collection time interval. The first group of writing index parameter acquisition modules and the second group of writing index parameter acquisition modules can capture extreme situations in the writing process by identifying the writing index parameters corresponding to the maximum difference and the minimum difference. These extreme situations often reflect sudden changes in writing speed or style, and have important reference value for adjusting the collection time interval.

[0139] The acquisition module of the collection time interval adjustment coefficient uses a complex formula to comprehensively consider multiple writing index parameters and difference values, and scientifically calculates the collection time interval adjustment coefficient. This coefficient takes into account both the overall trend of writing speed and the impact of extreme situations, ensuring the rationality and effectiveness of the adjustment. The comparison module and the adjustment execution module can automatically decide whether to adjust the time interval of the current handwriting collection based on the comparison result of the collection time interval adjustment coefficient and the preset adjustment coefficient threshold. This intelligent adjustment mechanism can flexibly respond to the user's actual writing situation and improve the adaptability and accuracy of data collection.

[0140] Through the above technical solution, the system can dynamically adjust the time interval for handwriting collection to ensure that the number of unnecessary collection times is reduced as much as possible without affecting the accuracy of data collection, thereby improving the collection efficiency. At the same time, since the collection time interval is more in line with the user's actual writing habits, the user's writing experience will also be improved accordingly. This technical solution also embodies the characteristics of personalization and precision. By independently analyzing and adjusting the writing data of each user, the system can better adapt to the writing habits and needs of different users, and provide more accurate data support for subsequent writing analysis, evaluation and teaching.

[0141] In summary, the above technical solution effectively improves the efficiency and accuracy of handwriting collection through a series of sophisticated module designs and algorithm implementations, while enhancing user experience and personalized service capabilities.

[0142] In this embodiment, the data processing module includes:

[0143] A data cleaning unit, used for cleaning the handwriting behavior sensing data;

[0144] Acquire handwriting behavior perception data based on non-invasive EEG;

[0145] Cleaning of handwriting behavior perception data based on non-invasive EEG, including:

[0146] To conduct consistency checks on non-invasive EEG-based handwriting behavior perception data;

[0147] According to the reasonable value range and mutual relationship of each variable in the handwriting behavior perception data based on non-invasive EEG, check whether the handwriting behavior perception data based on non-invasive EEG meets the requirements;

[0148] Remove inconsistent data that is beyond the normal range, logically unreasonable or contradictory in the handwriting behavior perception data based on non-invasive EEG;

[0149] Process invalid and missing values ​​of handwriting behavior perception data based on non-invasive EEG;

[0150] Remove invalid and missing data from handwriting behavior perception data based on non-invasive EEG that are not valuable for early diagnosis of Parkinson's disease;

[0151] Identify handwriting behavior perception data that is valuable for early diagnosis of Parkinson's disease;

[0152] A data conversion unit, used for converting the cleaned handwriting behavior perception data;

[0153] Obtain handwriting behavior perception data that is valuable for early diagnosis of Parkinson's disease after cleaning;

[0154] Convert handwriting behavior perception data that is valuable for early diagnosis of Parkinson's disease;

[0155] Eliminate dimensional differences between handwriting behavior perception data;

[0156] Determine handwriting behavior perception data with data standardization;

[0157] A feature extraction unit, used for extracting features from the converted handwriting behavior perception data;

[0158] Acquire the handwriting behavior perception data with data standardization after conversion;

[0159] Feature extraction of handwriting behavior perception data with data standardization;

[0160] Extract features that can reflect the early diagnosis of Parkinson's disease;

[0161] Determine handwriting behavior characteristic data based on non-invasive electroencephalogram.

[0162] Among them, the model building module is used to build an early diagnosis model for Parkinson's disease based on handwriting behavior feature recognition according to the needs of early diagnosis of Parkinson's disease;

[0163] In this embodiment, the model building module includes:

[0164] A data division unit, used for dividing the handwriting behavior feature data;

[0165] Obtain handwriting behavior feature data based on non-invasive EEG;

[0166] To classify the handwriting behavior feature data based on non-invasive EEG;

[0167] Determine the training data set and the test data set;

[0168] Model training unit, used to build an early diagnosis model for Parkinson's disease;

[0169] According to the demand for early diagnosis of Parkinson's disease based on handwriting behavior feature recognition, a neural network model framework suitable for early diagnosis of Parkinson's disease is selected;

[0170] Based on the training data set, the selected neural network model framework suitable for early diagnosis of Parkinson's disease is trained;

[0171] An early diagnosis model for Parkinson's disease based on handwriting behavior feature recognition was determined.

[0172] Among them, the test optimization module is used to test, evaluate and optimize the Parkinson's disease early diagnosis model to determine the optimal Parkinson's disease early diagnosis model based on handwriting behavior feature recognition;

[0173] In this embodiment, the test optimization module includes:

[0174] A test evaluation unit, used to test and evaluate the early diagnosis model of Parkinson's disease;

[0175] Obtain an early diagnosis model for Parkinson's disease based on handwriting behavior feature recognition;

[0176] Based on the test data set, the early diagnosis model of Parkinson's disease was tested and evaluated;

[0177] Determine the test evaluation results based on the Parkinson's disease early diagnosis model;

[0178] An optimization and adjustment unit, used for optimizing and adjusting the early diagnosis model of Parkinson's disease;

[0179] Obtain test evaluation results based on the Parkinson's disease early diagnosis model;

[0180] Mining and analyzing the test evaluation results based on the Parkinson's disease early diagnosis model;

[0181] Determine the optimal adjustment plan based on the early diagnosis model of Parkinson's disease;

[0182] Optimizing and adjusting the Parkinson's disease early diagnosis model according to the optimization and adjustment plan based on the Parkinson's disease early diagnosis model;

[0183] Determine the optimal early diagnosis model for Parkinson's disease based on handwriting behavior feature recognition.

[0184] Among them, the predictive diagnosis module is used to predict the real-time data of handwriting behavior and early diagnosis of Parkinson's disease, and determine the early diagnosis results of Parkinson's disease based on handwriting behavior feature recognition;

[0185] In this embodiment, the predictive diagnosis module includes:

[0186] A data acquisition unit, used for acquiring real-time data of handwriting behavior based on non-invasive electroencephalogram;

[0187] According to the need for early diagnosis of Parkinson's disease based on handwriting behavior feature recognition, the user's EEG activity changes, handwriting and handwriting drawings are monitored and collected in real time to obtain real-time handwriting behavior data based on non-invasive EEG;

[0188] Predictive diagnosis unit, used to predict real-time data of handwriting behavior and make early diagnosis of Parkinson's disease;

[0189] Obtain the optimal early diagnosis model for Parkinson's disease based on handwriting behavior feature recognition;

[0190] Input the real-time data of handwriting behavior based on non-invasive EEG into the optimal early diagnosis model of Parkinson's disease based on handwriting behavior feature recognition;

[0191] Based on the optimal Parkinson's disease early diagnosis model based on handwriting behavior feature recognition, the real-time data of handwriting behavior based on non-invasive EEG is predicted and Parkinson's disease is diagnosed early;

[0192] Determine the early diagnosis results of Parkinson's disease based on handwriting behavior feature recognition;

[0193] Among them, the early diagnosis result of Parkinson's disease based on handwriting behavior feature recognition includes: the user is an early Parkinson's disease patient or the user is not an early Parkinson's disease patient;

[0194] When the user is an early-stage Parkinson's disease patient, an early Parkinson's disease management and control plan is formulated to carry out intelligent management and control of early-stage Parkinson's disease patients.

[0195] Among them, the intelligent management and control module is used to formulate an early management plan for Parkinson's disease based on handwriting behavior feature recognition, and to carry out intelligent management and control of early Parkinson's disease patients.

[0196] In this embodiment, the intelligent management and control module includes:

[0197] a strategy development unit to develop early management plans for Parkinson’s disease;

[0198] Obtain early diagnosis results of Parkinson's disease based on handwriting behavior feature recognition;

[0199] Mining and analyzing the early diagnosis results of Parkinson's disease based on handwriting behavior feature recognition;

[0200] Determine an early control plan for Parkinson's disease based on handwriting behavior feature recognition;

[0201] Intelligent management and control unit, used for intelligent management and control of early-stage Parkinson's disease patients;

[0202] Obtain an early management plan for Parkinson's disease based on handwriting behavior feature recognition;

[0203] Based on the Parkinson's disease early management and control program, intelligent management and control of early Parkinson's disease patients are carried out, timely warnings are given to early Parkinson's disease patients, and personalized management and treatment measures are formulated for early Parkinson's disease patients.

[0204] It should be noted that, in this article, 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 "include", "comprise" 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.

[0205] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. The Parkinson's disease early diagnosis system based on handwriting behavior feature recognition is characterized by: include: A data collection module, used to collect the user's brain electrical activity change data, the user's handwriting data and the user's handwriting drawing data, and determine the handwriting behavior perception data based on the non-invasive electroencephalogram; A data processing module, used for processing the handwriting behavior perception data based on the non-invasive electroencephalogram to determine the handwriting behavior feature data based on the non-invasive electroencephalogram; A model building module is used to build an early diagnosis model for Parkinson's disease based on handwriting behavior feature recognition according to the needs of early diagnosis of Parkinson's disease; The test optimization module is used to test, evaluate and optimize the Parkinson's disease early diagnosis model to determine the optimal Parkinson's disease early diagnosis model based on handwriting behavior feature recognition; The predictive diagnosis module is used to predict the real-time data of handwriting behavior and diagnose Parkinson's disease in the early stage, and determine the early diagnosis results of Parkinson's disease based on the recognition of handwriting behavior characteristics; The intelligent management and control module is used to formulate an early management plan for Parkinson's disease based on handwriting behavior feature recognition and to provide intelligent management and control for early Parkinson's disease patients.

2. The Parkinson's disease early diagnosis system based on handwriting behavior feature recognition according to claim 1 is characterized in that: The data acquisition module comprises: An EEG acquisition unit is used to monitor and collect the user's brain function and cognitive state during the writing process based on a non-invasive EEG, and obtain the user's EEG activity change data; A handwriting acquisition unit, used to monitor and collect the handwriting of the user during the writing process based on computer vision, and obtain the user's handwriting data; A drawing acquisition unit, used to monitor and collect the handwritten drawings of the user during the writing process based on computer vision, and obtain the handwritten drawing data of the user; Among them, based on the user's brain electrical activity change data, the user's handwriting data and the user's handwriting drawing data, the handwriting behavior perception data based on the non-invasive electroencephalogram is determined.

3. According to the Parkinson's disease early diagnosis system based on handwriting behavior feature recognition in claim 2, the data acquisition module further comprises: The word count real-time collection module is used to collect the number of complete words completed per unit time during the user's writing process in real time; wherein the value range of the unit time is 2min-8min; A module for acquiring the area occupied by characters, for acquiring the area occupied by each character completed in each unit time after each unit time is completed; A writing index parameter acquisition module, used to acquire the writing index parameter corresponding to a unit time according to the number of complete characters of the user in each unit time and the character area occupied by each character completed in each unit time; The writing index parameter is obtained by the following formula: Where R represents the writing index parameter; n represents the number of unit time experienced by the user in writing; S zi represents the total writing area corresponding to the i-th unit time; S zi-1 represents the total writing area corresponding to the i-1th unit time; N i N represents the number of complete characters written in the i-th unit time; i-1 represents the number of complete written characters corresponding to the i-1th unit time; f n N represents the standard deviation of the number of complete characters written in n units of time; p S represents the average number of complete characters written in n units of time; p S represents the average value of the total writing area corresponding to n unit time; dij represents the area occupied by the jth complete word in the i-th unit time; The adjustment module is used to adjust the time interval for collecting handwriting using the difference value of the writing index parameter between every two adjacent unit times.

4. According to the Parkinson's disease early diagnosis system based on handwriting behavior feature recognition in claim 3, the adjustment module comprises: A writing index parameter extraction module, used to extract writing index parameters of every two adjacent unit times; A difference value acquisition module, used for acquiring a difference value of the writing index parameters of every two adjacent unit times by using the writing index parameters of every two adjacent unit times; A first group of writing index parameter acquisition module is used to extract the writing index parameters of two unit times corresponding to the maximum value of the difference between the difference values ​​of the writing index parameters of each two adjacent unit times as the first group of writing index parameters; A second group of writing index parameter acquisition module is used to extract the writing index parameters of two unit times corresponding to the minimum difference value among the difference values ​​of the writing index parameters of each two adjacent unit times as the second group of writing index parameters; The acquisition module for adjusting the collection time interval is used to obtain the collection time interval adjustment coefficient by using the first group of writing index parameters and the second group of writing index parameters, wherein the collection time interval adjustment coefficient is obtained by the following formula: Wherein, K represents the collection time interval adjustment coefficient; k represents the number of writing index parameters other than the first group of writing index parameters and the second group of writing index parameters; R i represents the value corresponding to the i-th writing index parameter except the first group of writing index parameters and the second group of writing index parameters; C max and C min Respectively represent the maximum and minimum difference values ​​of the difference values ​​of the writing index parameters between every two adjacent unit times; R max01 and R max02 They represent the writing index parameters of the first group of writing index parameters corresponding to two unit times; R min01 and R min02 Respectively represent the writing index parameters of two units of time corresponding to the second group of writing index parameters; A comparison module, used for comparing the acquisition time interval adjustment coefficient with a preset adjustment coefficient threshold; An adjustment execution module, configured to adjust the current handwriting collection time interval using the collection time interval adjustment coefficient when the collection time interval adjustment coefficient exceeds a preset adjustment coefficient threshold, so as to obtain an adjusted handwriting collection time interval; The adjusted time interval for collecting handwriting is obtained by the following formula: T=[1+log 10 (1+K)]·T0 Wherein, T represents the adjusted time interval for collecting handwriting; T0 represents the current time interval for collecting handwriting; and K represents the collection time interval adjustment coefficient.

5. The Parkinson's disease early diagnosis system based on handwriting behavior feature recognition according to claim 2 is characterized in that: The data processing module comprises: A data cleaning unit, used for cleaning handwriting behavior sensing data; Acquire handwriting behavior perception data based on non-invasive EEG; Cleaning of handwriting behavior perception data based on non-invasive EEG, including: To conduct consistency checks on non-invasive EEG-based handwriting behavior perception data; According to the reasonable value range and mutual relationship of each variable in the handwriting behavior perception data based on non-invasive EEG, check whether the handwriting behavior perception data based on non-invasive EEG meets the requirements; Remove inconsistent data that is beyond the normal range, logically unreasonable or contradictory in the handwriting behavior perception data based on non-invasive EEG; Process invalid and missing values ​​of handwriting behavior perception data based on non-invasive EEG; Remove invalid and missing data from handwriting behavior perception data based on non-invasive EEG that are not valuable for early diagnosis of Parkinson's disease; Identify handwriting behavior perception data that is valuable for early diagnosis of Parkinson's disease.

6. The Parkinson's disease early diagnosis system based on handwriting behavior feature recognition according to claim 5 is characterized in that: The data processing module also includes: A data conversion unit, used for converting the cleaned handwriting behavior perception data; Obtain handwriting behavior perception data that is valuable for early diagnosis of Parkinson's disease after cleaning; Convert handwriting behavior perception data that is valuable for early diagnosis of Parkinson's disease; Eliminate dimensional differences between handwriting behavior perception data; Determine handwriting behavior perception data with data standardization; A feature extraction unit, used for extracting features from the converted handwriting behavior perception data; Acquire the handwriting behavior perception data with data standardization after conversion; Feature extraction of handwriting behavior perception data with data standardization; Extract features that can reflect the early diagnosis of Parkinson's disease; Determine handwriting behavior characteristic data based on non-invasive electroencephalogram.

7. The Parkinson's disease early diagnosis system based on handwriting behavior feature recognition according to claim 6 is characterized in that: The model building module includes: A data division unit, used for dividing the handwriting behavior feature data; Obtain handwriting behavior feature data based on non-invasive EEG; To classify the handwriting behavior feature data based on non-invasive EEG; Determine the training data set and the test data set; Model training unit, used to build an early diagnosis model for Parkinson's disease; According to the demand for early diagnosis of Parkinson's disease based on handwriting behavior feature recognition, a neural network model framework suitable for early diagnosis of Parkinson's disease is selected; Based on the training data set, the selected neural network model framework suitable for early diagnosis of Parkinson's disease is trained; An early diagnosis model for Parkinson's disease based on handwriting behavior feature recognition was determined.

8. The Parkinson's disease early diagnosis system based on handwriting behavior feature recognition according to claim 7 is characterized in that: The test optimization module includes: A test evaluation unit, used to test and evaluate the early diagnosis model of Parkinson's disease; Obtain an early diagnosis model for Parkinson's disease based on handwriting behavior feature recognition; Based on the test data set, the early diagnosis model of Parkinson's disease was tested and evaluated; Determine the test evaluation results based on the Parkinson's disease early diagnosis model; An optimization and adjustment unit, used for optimizing and adjusting the early diagnosis model of Parkinson's disease; Obtain test evaluation results based on the Parkinson's disease early diagnosis model; Mining and analyzing the test evaluation results based on the Parkinson's disease early diagnosis model; Determine the optimal adjustment plan based on the early diagnosis model of Parkinson's disease; Optimizing and adjusting the Parkinson's disease early diagnosis model according to the optimization and adjustment plan based on the Parkinson's disease early diagnosis model; Determine the optimal early diagnosis model for Parkinson's disease based on handwriting behavior feature recognition.

9. The Parkinson's disease early diagnosis system based on handwriting behavior feature recognition according to claim 8, characterized in that: The predictive diagnosis module comprises: A data acquisition unit, used for acquiring real-time data of handwriting behavior based on non-invasive electroencephalogram; According to the need for early diagnosis of Parkinson's disease based on handwriting behavior feature recognition, the user's EEG activity changes, handwriting and handwriting drawings are monitored and collected in real time to obtain real-time handwriting behavior data based on non-invasive EEG; Predictive diagnosis unit, used to predict real-time data of handwriting behavior and make early diagnosis of Parkinson's disease; Obtain the optimal early diagnosis model for Parkinson's disease based on handwriting behavior feature recognition; Input the real-time data of handwriting behavior based on non-invasive EEG into the optimal early diagnosis model of Parkinson's disease based on handwriting behavior feature recognition; Based on the optimal Parkinson's disease early diagnosis model based on handwriting behavior feature recognition, the real-time data of handwriting behavior based on non-invasive EEG is predicted and Parkinson's disease is diagnosed early; Determine the early diagnosis results of Parkinson's disease based on handwriting behavior feature recognition; Among them, the early diagnosis result of Parkinson's disease based on handwriting behavior feature recognition includes: the user is an early Parkinson's disease patient or the user is not an early Parkinson's disease patient; When the user is an early-stage Parkinson's disease patient, an early Parkinson's disease management and control plan is formulated to carry out intelligent management and control of early-stage Parkinson's disease patients.

10. The Parkinson's disease early diagnosis system based on handwriting behavior feature recognition according to claim 9, characterized in that: The intelligent management and control module includes: a strategy development unit to develop early management plans for Parkinson’s disease; Obtain early diagnosis results of Parkinson's disease based on handwriting behavior feature recognition; Mining and analyzing the early diagnosis results of Parkinson's disease based on handwriting behavior feature recognition; Determine an early control plan for Parkinson's disease based on handwriting behavior feature recognition; Intelligent management and control unit, used for intelligent management and control of early-stage Parkinson's disease patients; Obtain an early management plan for Parkinson's disease based on handwriting behavior feature recognition; Based on the Parkinson's disease early management and control program, intelligent management and control of early Parkinson's disease patients are carried out, timely warnings are given to early Parkinson's disease patients, and personalized management and treatment measures are formulated for early Parkinson's disease patients.

Images