Parkinson's motion symptom quantitative evaluation system based on deep learning

Abstract

The invention provides a Parkinson's motion symptom quantitative evaluation system based on deep learning, and particularly relates to the field of medical monitoring, the Parkinson's motion symptom quantitative evaluation system comprises an evaluation unit, the evaluation unit comprises a sensing module used for patient motion sensing recording, the sensing module is electrically connected with an acquisition module used for data acquisition, and the acquisition module is electrically connected with the evaluation unit. And the acquisition module is electrically connected with an environment recording module used for recording and digitizing environment data. According to the invention, due to different environments and different motion modes formed by different environments, effective motion data generated by the motion is different, so that data analysis and data arrangement are carried out, and specific effective motion data generated by the motion made by the environment of the patient can be specifically analyzed; the effective exercise data represents the amount of exercise of the datamation patient, and the recorded and analyzed data is uploaded to the database end for storage through the feature uploading module, so that the learning unit can learn and record corresponding specific data according to different environments.

Description

Technical Field

[0001] The present application relates to the field of medical monitoring, and in particular to a Parkinson's motor symptom quantitative assessment system based on deep learning. Background Art

[0002] Parkinson's disease is a chronic progressive neurological disease that causes the patient to gradually lose some of his or her body functions. The current effective treatment method uses drugs and exercise monitoring for treatment, but the amount of drugs used needs to be matched with the exercise status. When setting the drug dosage for the patient, it is necessary to monitor and evaluate the patient's motor ability and the body's ability to move. However, in the existing technology, patients cannot be specifically diagnosed in clinical diagnosis due to different environments and different exercise methods. At the same time, there is no specific data support in the diagnosis, which leads to deviations in the setting of drug dosage. As a result, only professional physicians can set accurate drug dosages based on experience. However, due to the current increase in Parkinson's patients and the lack of professional physicians, many patients cannot receive effective treatment when seeking medical treatment, and ultimately the progression of the disease cannot be effectively controlled. At this time, a Parkinson's motor symptom quantitative assessment system based on deep learning is needed to solve the above problems. Summary of the invention

[0003] The purpose of the present invention is to solve the shortcomings existing in the prior art. In the prior art, patients cannot be specifically diagnosed in clinical diagnosis due to different environments and different movement methods. At the same time, there is no specific data support in the diagnosis, which leads to deviations in the setting of drug dosage. As a result, only professional physicians can set accurate drug dosage based on experience. However, due to the increase in the number of Parkinson's patients and the small number of professional physicians, many patients cannot receive effective treatment when seeking medical treatment, and ultimately the progression of the disease cannot be effectively controlled. A Parkinson's motor symptom quantitative evaluation system based on deep learning is proposed.

[0004] In order to achieve the above object, the present invention adopts the following technical solutions: The present application provides a Parkinson's motor symptom quantitative assessment system based on deep learning, including an assessment unit, the assessment unit including a sensor module for sensing and recording patient movements, the sensor module being electrically connected to a collection module for data collection, the collection module being electrically connected to an environment collection module for recording and digitizing environmental data, the collection module being electrically connected to a simulation module for performing three-dimensional dynamic simulation according to environmental data and motion data, the simulation module being electrically connected to a learning unit for recording learning and a discrimination module for analyzing and discriminating effective movements, the discrimination module being electrically connected to the learning module, the discrimination module being electrically connected to a data analysis module for analyzing and discriminating specific values ​​of the data, and the data analysis module being electrically connected to an assessment module for evaluating various Parkinson's motor symptom scores.

[0005] In one implementation, the environment collection module collects the environmental data of the motion area, records the motion area by video recording, and converts the recorded environmental data into three-dimensional simulation values. The three-dimensional simulation values ​​in the environment collection module are then collected by the acquisition module, and the simulation module creates a three-dimensional environment model.

[0006] In one implementation, the sensing module receives and processes data through sensors installed at various parts of the patient's body, monitors and records motion data and motion time data at various parts, and the acquisition module collects the various data monitored and recorded by the sensing module. At the same time, the simulation module generates a corresponding dynamic stickman based on the various patient data collected by the acquisition module, and matches it with the three-dimensional environmental model, so that the stickman and the three-dimensional environmental model can perform various actions corresponding to the patient to form a three-dimensional dynamic model.

[0007] In one implementation, the learning unit includes a simulation analysis module for performing data analysis and matching on a three-dimensional environment model, the simulation analysis module is electrically connected to a database end for recording, storing and querying data, the simulation analysis module is electrically connected to a record analysis module for data recording and data analysis and a feature extraction module for extracting data vital signs, the record analysis module is electrically connected to a feature upload module for uploading feature data, the feature upload module is electrically connected to the database end, the feature extraction module is electrically connected to a data sorting module for data sorting and data processing, the data sorting module is electrically connected to the feature upload module, the data sorting module is electrically connected to the identification module, and the simulation analysis module is electrically connected to the simulation module.

[0008] In one implementation, the simulation analysis module extracts the three-dimensional environment model in the simulation module, matches it according to the data stored in the database, and analyzes whether the database has corresponding three-dimensional environment model data. When the database has matching three-dimensional environment model data, the feature extraction module extracts the data stored in the database, and the data sorting module sorts and analyzes the three-dimensional environment simulation data to obtain specific effective motion data that can be generated by each patient when exercising in various places.

[0009] In one implementation, when the database end does not have matching three-dimensional environment model data, the recording and analysis module records and analyzes the three-dimensional environment model data, analyzes the effective motion data generated by the patient during exercise, and the feature uploading module uploads the recorded and analyzed data to the database end for storage, and the data sorting and analysis module analyzes it at the same time.

[0010] In one implementation, the identification module identifies and records the effective motion data sorted by the data sorting module and the specific amount of motion at the patient's location, and the data analysis module specifically analyzes and sorts the data, and the evaluation module evaluates the scores of various Parkinson's motor symptoms.

[0011] In one implementation, the evaluation unit and the learning unit specifically operate by: S1. Use sensors to attach to various parts of the patient's body to obtain the patient's movement data, and at the same time record and convert the patient's movement environment, perform dynamic simulation through the simulation module, and evaluate and score the patient's movement data; S2. The learning unit records features and provides valid motion data, identifies valid motion features through the identification module, and organizes them through the data analysis module.

[0012] In one implementation, in step S1: 1) Use sensors to attach to various parts of the patient's body, and connect data through the sensor module and the sensor. The sensor module records the patient's movement trajectory and specific movement time in real time; 2) Convert the patient's movement environment video and perform dynamic simulation through the simulation module, classify and simulate the three-dimensional environment model of the patient's movement area and all the movement trajectories of the dynamic stickman, and implant the dynamic stickman into the three-dimensional environment model to move in accordance with the patient's movement trajectory; 3) The identification module analyzes and organizes the data provided by the identification module based on the effective motion data provided by the learning unit and the effective trajectory motion performed by the dynamic stickman. The evaluation module evaluates the data analyzed and organized by the data analysis module to obtain the Parkinson's motor symptom score.

[0013] In one implementation, in step S2: 1) The simulation analysis module obtains the 3D environment data in the simulation module and matches it with the database to see if there is corresponding 3D environment data. When there is no matching 3D environment model data on the database, the 3D environment model data is analyzed by the recording and analysis module to analyze the effective motion data that can be generated when the patient moves at various positions. The feature upload module uploads the recorded and analyzed data to the database for storage; 2) When the database has matching 3D environment model data, the feature extraction module will extract the matching data stored in the database, which contains valid motion data that can be generated by corresponding motions at various locations; 3) The data sorting module receives the valid motion data generated by the feature uploading module and the feature extraction module, and the identification data is processed according to the data sorted by the data sorting module.

[0014] The present invention can collect on-site environmental data through the environmental collection module, and digitize corresponding data for different environments and different movement modes formed by different environments. At the same time, the simulation module can generate a corresponding three-dimensional environmental model according to the recorded environmental data, thereby simplifying the specific movement video of the patient's movement, which can be viewed through the model simulation record, and then the simulation module can enable other medical staff to view and draw conclusions more conveniently and quickly.

[0015] The sensing module can monitor the various movements of the patient, and the dynamic stickman can be used to quickly generate the corresponding human shadow. The stickman is easy to generate in motion and easy to simulate the various movements of the patient. At the same time, the dynamic stickman is used to facilitate the related linkage with the three-dimensional dynamic model, which can better generate the corresponding three-dimensional dynamic model.

[0016] The three-dimensional environment model in the simulation module is extracted through the simulation analysis module, and the data stored in the database end is matched with the data obtained by the simulation analysis module to analyze whether the database end has the corresponding three-dimensional environment model data. When the database end has matching three-dimensional environment model data, the feature extraction module extracts the data stored in the database end. At this time, the feature extraction module can obtain the corresponding data in the database end through the simulation analysis module, and then quickly match the effective motion data that can be generated by the corresponding motion in each place in the data, thereby reducing the computing power of the system.

[0017] When there is no matching three-dimensional environmental model data on the database side, the three-dimensional environmental model data is recorded and analyzed through the recording and analysis module to analyze the effective motion data that can be generated when the patient moves at various positions. Due to different environments and different movement methods formed by different environments, the effective motion data generated by the exercise performed are also different. For example, push-ups on flat ground and push-ups on inclined surfaces. Data analysis and data collation are performed to specifically analyze the specific effective motion data that can be generated by the exercise performed by the patient in the environment. The effective motion data represents the digitized patient exercise volume, and the recorded and analyzed data is uploaded to the database side for storage through the feature upload module. At this point, the learning unit can learn and record the corresponding specific data according to different environments. The data collation and analysis module can analyze and organize the data in the feature extraction module and the feature upload module to provide data support to the identification module. The identification module can identify according to the exercise duration and effective motion data, eliminate the effective exercise proportion of the exercise duration and rest duration, and also analyze according to the exercise time and rest time within the specified time of the patient. The best effective motion data is obtained through the data analysis module, and the various Parkinson's motor symptom scores are evaluated through the evaluation module. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a system flow diagram of an evaluation unit of a Parkinson's motor symptom quantitative evaluation system based on deep learning in the present invention; Figure 2 This is a system flow diagram of the learning unit of the Parkinson's motor symptom quantitative assessment system based on deep learning in the present invention. DETAILED DESCRIPTION

[0019] The technical solutions in the embodiments of the present application will be described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments.

[0020] In the following, the terms "first", "second", etc. are used for descriptive purposes only and are not to be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first", "second", etc. may explicitly or implicitly include one or more of the features. In the description of this application, unless otherwise specified, "plurality" means two or more.

[0021] In addition, in the present application, directional terms such as "up", "down", "left" and "right" may be defined including but not limited to the orientation relative to the schematic placement of the components in the drawings. It should be understood that these directional terms may be relative concepts, which are used for relative description and clarification, and may change accordingly according to the change in the orientation of the components in the drawings.

[0022] In this application, unless otherwise specified or limited, the term "connection" should be understood in a broad sense. For example, "connection" can be a fixed connection, a detachable connection, or an integral connection; it can be a direct connection or an indirect connection through an intermediate medium. In addition, the term "coupling" can be a way of achieving electrical connection for signal transmission.

[0023] As used herein, “about,” “substantially,” or “approximately” includes the stated value and reference values ​​that are within an acceptable range of deviation from the particular value, where the acceptable range of deviation is as determined by one of ordinary skill in the art taking into account the measurements in question and the errors associated with the measurement of the particular quantity (i.e., the limitations of the measurement system).

[0024] In the embodiments of the present application, the words "exemplarily" or "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "exemplarily" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplarily" or "for example" is intended to present related concepts in a specific way.

[0025] like Figure 1-2 As shown, a Parkinson's motor symptom quantitative assessment system based on deep learning includes an assessment unit, the assessment unit includes a sensor module for patient movement sensing and recording, the sensor module is electrically connected to a collection module for data collection, the collection module is electrically connected to an environment collection module for recording and digitizing environmental data, the collection module is electrically connected to a simulation module for performing three-dimensional dynamic simulation according to environmental data and motion data, the simulation module is electrically connected to a learning unit for recording learning and a discrimination module for analyzing and identifying effective movements, the discrimination module is electrically connected to the learning module, the discrimination module is electrically connected to a data analysis module for analyzing and identifying specific values ​​of the data, and the data analysis module is electrically connected to an assessment module for evaluating various Parkinson's motor symptom scores; The environment collection module collects the environmental data of the sports area, records the sports area by video recording, and converts the recorded environmental data into three-dimensional simulation values. Then, the acquisition module collects the three-dimensional simulation values ​​in the environment collection module, and the simulation module creates a three-dimensional environment model. Through the above embodiments, the environment collection module can collect on-site environment data, and corresponding data can be generated for different environments and different movement modes formed by different environments. At the same time, the simulation module can generate a corresponding three-dimensional environment model according to the recorded environment data, thereby simplifying the specific movement video of the patient's movement, and can view it through the model simulation record, and then the simulation module can enable other medical staff to watch and draw conclusions more conveniently and quickly; The sensing module receives and processes data through sensors installed at various parts of the patient's body, monitors and records the motion data and motion time data at various parts, and the acquisition module collects the various data monitored and recorded by the sensing module. At the same time, the simulation module generates a corresponding dynamic stickman based on the various data of the patient collected by the acquisition module, and matches it with the three-dimensional environmental model, so that the stickman and the three-dimensional environmental model can perform various actions corresponding to the patient to form a three-dimensional dynamic model; Through the above embodiments, the various movements of the patient can be monitored by the sensor module, and the dynamic stickman is used to quickly generate the corresponding human figure, and the stickman is easy to generate in motion and easy to simulate the various movements of the patient. At the same time, the dynamic stickman is used to facilitate the relevant linkage with the three-dimensional dynamic model, so that the corresponding three-dimensional dynamic model can be better generated; The learning unit includes a simulation analysis module for performing data analysis and matching on a three-dimensional environment model, the simulation analysis module is electrically connected to a database end for recording, storing and querying data, the simulation analysis module is electrically connected to a recording analysis module for data recording and data analysis and a feature extraction module for extracting data vital signs, the recording analysis module is electrically connected to a feature upload module for uploading feature data, the feature upload module is electrically connected to the database end, the feature extraction module is electrically connected to a data sorting module for data sorting and data processing, the data sorting module is electrically connected to the feature upload module, the data sorting module is electrically connected to the identification module, and the simulation analysis module is electrically connected to the simulation module; The simulation analysis module extracts the three-dimensional environment model in the simulation module, matches it with the data stored in the database, and analyzes whether the database has the corresponding three-dimensional environment model data. When the database has the matching three-dimensional environment model data, the feature extraction module extracts the data stored in the database, and the data sorting module sorts and analyzes the three-dimensional environment simulation data to obtain the specific effective movement data that can be generated by each patient when exercising in various places. Through the above embodiment, the three-dimensional environment model in the simulation module is extracted by the simulation analysis module, and the data stored in the database end is matched with the data obtained by the simulation analysis module to analyze whether the database end has the corresponding three-dimensional environment model data. When the database end has the matching three-dimensional environment model data, the feature extraction module extracts the data stored in the database end. At this time, the feature extraction module can obtain the corresponding data in the database end through the simulation analysis module, and then quickly match the effective motion data that can be generated by the corresponding motion in each place in the data, thereby reducing the computing power of the system; When there is no matching 3D environment model data on the database side, the recording and analysis module records and analyzes the 3D environment model data, analyzes the effective motion data generated by the patient during exercise, and the feature uploading module uploads the recorded and analyzed data to the database side for storage, and the data sorting and analysis module analyzes it at the same time; Through the above embodiment, when there is no matching three-dimensional environment model data on the database side, the three-dimensional environment model data is recorded and analyzed through the recording and analysis module, and the effective motion data that can be generated when the patient moves at various positions is analyzed. Due to different environments and different motion modes formed by different environments, the effective motion data generated by the exercises performed are also different. For example, push-ups on flat ground and push-ups on inclined surfaces are performed. Therefore, data analysis and data sorting are performed, and the specific effective motion data that can be generated by the exercises performed by the patient in the environment can be specifically analyzed. The effective motion data represents the digitized patient exercise volume, and the recorded and analyzed data is uploaded to the database side for storage through the feature uploading module. At this point, the learning unit can learn and record corresponding specific data according to different environments, and the data in the feature extraction module and the feature uploading module can be analyzed and sorted through the data sorting and analysis module, so as to provide data support to the identification module; The identification module will identify and record the effective movement data sorted by the data sorting module and the specific movement amount of the patient's location, and the data analysis module will specifically analyze and sort the data, and the evaluation module will evaluate the scores of various Parkinson's movement symptoms; Through the above embodiment, the identification module can identify according to the exercise duration and effective exercise data, eliminate the effective exercise ratio of the exercise duration and rest duration, and also analyze according to the exercise time and rest time within the prescribed duration of the patient, obtain the best effective exercise data through the data analysis module, and evaluate various Parkinson's motor symptom scores through the evaluation module; The specific operations of the assessment unit and learning unit include the following steps: S1. Use sensors to attach to various parts of the patient's body to obtain the patient's movement data, and at the same time record and convert the patient's movement environment, perform dynamic simulation through the simulation module, and evaluate and score the patient's movement data; S2. The learning unit records features and provides valid motion data, identifies valid motion features through the identification module, and organizes them through the data analysis module.

[0026] In step S1: 1) Use sensors to attach to various parts of the patient's body, and connect data through the sensor module and the sensor. The sensor module records the patient's movement trajectory and specific movement time in real time; 2) Convert the patient's movement environment video and perform dynamic simulation through the simulation module, classify and simulate the three-dimensional environment model of the patient's movement area and all the movement trajectories of the dynamic stickman, and implant the dynamic stickman into the three-dimensional environment model to move in accordance with the patient's movement trajectory; 3) The identification module analyzes and organizes the data provided by the identification module based on the effective motion data provided by the learning unit and the effective trajectory motion performed by the dynamic stickman. The evaluation module evaluates the data analyzed and organized by the data analysis module to obtain the Parkinson's motor symptom score.

[0027] In step S2: 1) The simulation analysis module obtains the 3D environment data in the simulation module and matches it with the database to see if there is corresponding 3D environment data. When there is no matching 3D environment model data on the database, the 3D environment model data is analyzed by the recording and analysis module to analyze the effective motion data that can be generated when the patient moves at various positions. The feature upload module uploads the recorded and analyzed data to the database for storage; 2) When the database has matching 3D environment model data, the feature extraction module will extract the matching data stored in the database, which contains valid motion data that can be generated by corresponding motions at various locations; 3) The data sorting module receives the valid motion data generated by the feature uploading module and the feature extraction module, and the identification data is processed according to the data sorted by the data sorting module.

[0028] Working principle: When the present invention is in use, sensors are attached to various parts of the patient's body, and data is communicated through the sensor module and the sensor. The sensor module records the patient's movement trajectory and specific movement time in real time, and converts the patient's movement environment. The simulation module performs dynamic simulation, and classifies and simulates the three-dimensional environment model of the patient's movement area and all the movement trajectories of the dynamic stickman. The dynamic stickman is implanted in the three-dimensional environment model to move along the corresponding movement trajectory of the patient. The simulation analysis module obtains the three-dimensional environment data in the simulation module, and matches it according to the database end to see if there is corresponding three-dimensional environment data. When the database end does not have matching three-dimensional environment model data, the three-dimensional environment model data is analyzed by the recording and analysis module to analyze the patient's movement trajectory. When the person moves at various positions, the effective motion data that can be generated will be uploaded to the database by the feature uploading module. The recorded and analyzed data will be stored in the database. When the database has matching three-dimensional environmental model data, the feature extraction module will extract the matching data stored in the database. The matching data has the effective motion data that can be generated by the corresponding motion at various locations. The data sorting module will receive the effective motion data generated by the feature uploading module and the feature extraction module. The identification module will analyze and sort the data provided by the identification module based on the effective motion data provided by the learning unit and the effective trajectory motion performed by the dynamic stickman. The evaluation module will evaluate the data analyzed and sorted by the data analysis module to obtain the Parkinson's motor symptom score.

[0029] Through the description of the above implementation methods, technicians in the relevant field can clearly understand that the diagnostic method in the above embodiment can be implemented by software plus a necessary general hardware platform, and of course it can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the relevant technology can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), including a number of instructions for a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods of various embodiments of the present invention.

[0030] Through the description of the above implementation methods, technical personnel in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional modules is used as an example. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

[0031] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of modules or units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0032] The units described as separate components may or may not be physically separated, and the components shown as units may be one physical unit or multiple physical units, that is, they may be located in one place or distributed in multiple different places. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0033] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above integrated units may be implemented in the form of hardware.

[0034] The above contents are only specific implementation methods of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions within the technical scope disclosed in the present application shall be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims

1. A Parkinson's motor symptom quantitative assessment system based on deep learning, It is characterized in that The evaluation unit includes an evaluation unit, which includes a sensor module for sensing and recording patient movements, the sensor module is electrically connected to a collection module for data collection, the collection module is electrically connected to an environment collection module for recording and digitizing environment data, the collection module is electrically connected to a simulation module for performing three-dimensional dynamic simulation according to environment data and movement data, the simulation module is electrically connected to a learning unit for recording learning and a discrimination module for analyzing and distinguishing effective movements, the discrimination module is electrically connected to the learning module, the discrimination module is electrically connected to a data analysis module for analyzing and distinguishing specific values ​​of the data, and the data analysis module is electrically connected to an evaluation module for evaluating various Parkinson's movement symptom scores.

2. The Parkinson's motor symptom quantitative assessment system based on deep learning according to claim 1, It is characterized in that The environment collection module collects the environmental data of the motion area, records the motion area by video recording, and converts the recorded environmental data into three-dimensional simulation values. The three-dimensional simulation values ​​in the environment collection module are then collected by the acquisition module, and the simulation module creates a three-dimensional environment model.

3. The Parkinson's motor symptom quantitative assessment system based on deep learning according to claim 2, It is characterized in that The sensing module receives and processes data through sensors installed in various parts of the patient's body, monitors and records the movement data and movement time data of various parts, and the acquisition module collects the various data monitored and recorded in the sensing module. At the same time, the simulation module generates a corresponding dynamic Stickman based on the various patient data collected by the acquisition module, and matches it with the three-dimensional environmental model, so that the Stickman and the three-dimensional environmental model can perform various actions corresponding to the patient to form a three-dimensional dynamic model.

4. The Parkinson's motor symptom quantitative assessment system based on deep learning according to claim 1, It is characterized in that The learning unit includes a simulation analysis module for performing data analysis and matching on a three-dimensional environment model, the simulation analysis module is electrically connected to a database end for recording, storing and querying data, the simulation analysis module is electrically connected to a record analysis module for data recording and data analysis and a feature extraction module for extracting data vital signs, the record analysis module is electrically connected to a feature upload module for uploading feature data, the feature upload module is electrically connected to the database end, the feature extraction module is electrically connected to a data sorting module for data sorting and data processing, the data sorting module is electrically connected to the feature upload module, the data sorting module is electrically connected to the identification module, and the simulation analysis module is electrically connected to the simulation module.

5. The Parkinson's motor symptom quantitative assessment system based on deep learning according to claim 4, It is characterized in that The simulation analysis module will extract the three-dimensional environment model in the simulation module, match it according to the data stored in the database, and analyze whether the database has the corresponding three-dimensional environment model data. When the database has matching three-dimensional environment model data, the feature extraction module will extract the data stored in the database, and the data sorting module will sort and analyze the three-dimensional environment simulation data to obtain the specific effective motion data that can be generated by each patient when exercising in various places.

6. The Parkinson's motor symptom quantitative assessment system based on deep learning according to claim 5, It is characterized in that When there is no matching three-dimensional environment model data on the database side, the recording and analysis module records and analyzes the three-dimensional environment model data, analyzes the effective motion data generated by the patient during exercise, and the feature uploading module uploads the recorded and analyzed data to the database side for storage, and the data sorting and analysis module analyzes it at the same time.

7. The Parkinson's motor symptom quantitative assessment system based on deep learning according to claim 6, It is characterized in that The identification module will identify and record the effective movement data sorted by the data sorting module and the specific movement amount of the patient's location, and the data analysis module will specifically analyze and sort the data, and the evaluation module will evaluate the scores of various Parkinson's movement symptoms.

8. The Parkinson's motor symptom quantitative assessment system based on deep learning according to claim 1, It is characterized in that The specific operations of the evaluation unit and the learning unit include the following steps: S1. Use sensors to attach to various parts of the patient's body to obtain the patient's movement data, and at the same time record and convert the patient's movement environment, perform dynamic simulation through the simulation module, and evaluate and score the patient's movement data; S2. The learning unit records features and provides valid motion data, identifies valid motion features through the identification module, and organizes them through the data analysis module.

9. The Parkinson's motor symptom quantitative assessment system based on deep learning according to claim 8, It is characterized in that In step S1: 1) Use sensors to attach to various parts of the patient's body, and connect data through the sensor module and the sensor. The sensor module records the patient's movement trajectory and specific movement time in real time; 2) Convert the patient's movement environment video and perform dynamic simulation through the simulation module, classify and simulate the three-dimensional environment model of the patient's movement area and all the movement trajectories of the dynamic stickman, and implant the dynamic stickman into the three-dimensional environment model to move in accordance with the patient's movement trajectory; 3) The identification module analyzes and organizes the data provided by the identification module based on the effective motion data provided by the learning unit and the effective trajectory motion performed by the dynamic stickman. The evaluation module evaluates the data analyzed and organized by the data analysis module to obtain the Parkinson's motor symptom score.

10. The Parkinson's motor symptom quantitative assessment system based on deep learning according to claim 8, It is characterized in that In step S2: 1) The simulation analysis module obtains the 3D environment data in the simulation module and matches it with the database to see if there is corresponding 3D environment data. When there is no matching 3D environment model data on the database, the 3D environment model data is analyzed by the recording and analysis module to analyze the effective motion data that can be generated when the patient moves at various positions. The feature upload module uploads the recorded and analyzed data to the database for storage; 2) When the database has matching 3D environment model data, the feature extraction module will extract the matching data stored in the database, which contains valid motion data that can be generated by corresponding motions at various locations; 3) The data sorting module receives the valid motion data generated by the feature uploading module and the feature extraction module, and the identification data is processed according to the data sorted by the data sorting module.

Images