Rehabilitation guidance system and method for chronic diseases

Abstract

The invention discloses a rehabilitation guidance system and method for chronic diseases, belongs to the technical field of data recognition, and aims to realize similarity reference by collecting historical data of all patients and rehabilitation schemes corresponding to the historical data, match rehabilitation medication conditions and rehabilitation training conditions most meeting own conditions for the patients, and improve the rehabilitation guidance effect. And meanwhile, the artificial intelligence technology is adopted to identify the current basic data and the current chronic disease index corresponding to the patient to be guided, so that the rehabilitation medication condition and the rehabilitation training condition suggested by a doctor can be obtained, more rehabilitation guidance schemes are provided for the patient, and the patient is assisted to take effective rehabilitation measures.

Description

Technical Field

[0001] The present invention belongs to the technical field of data recognition, and in particular relates to a rehabilitation guidance system and method for chronic diseases. Background Art

[0002] Chronic diseases, also known as chronic non-communicable diseases, are a type of disease that has an insidious onset, a long course, and is difficult to cure. The causes of these diseases are usually complex, and some have not even been fully identified.

[0003] In the rehabilitation process of existing chronic diseases, doctors in hospitals often give advice and patients go home to implement it themselves. However, patients may forget, especially the elderly, which makes it easier to forget, resulting in ineffective implementation of rehabilitation measures. In addition, at different stages of chronic diseases, the amount of medication and rehabilitation training required may vary, resulting in patients being unable to effectively implement rehabilitation measures. Summary of the invention

[0004] The present invention provides a rehabilitation guidance system and method for chronic diseases, which are used to solve the technical problems existing in the prior art.

[0005] In one aspect, the present invention provides a rehabilitation guidance system for chronic diseases, comprising: a data acquisition module, a history database module, a data matching module, a data analysis module, and a rehabilitation guidance module;

[0006] The data acquisition module is used to collect the current basic data and current chronic disease indicators corresponding to the patient to be guided, and transmit the current basic data and current chronic disease indicators to the data matching module;

[0007] The historical database module is used to store historical data corresponding to chronic diseases and rehabilitation programs corresponding to the historical data; wherein the historical data includes historical basic data and historical chronic disease indicators, and the rehabilitation program includes rehabilitation medication and rehabilitation training;

[0008] The data matching module is used to match the current basic data and the current chronic disease index in the database, determine the closest target historical data, and determine the rehabilitation program corresponding to the target historical data as the first target rehabilitation program;

[0009] The historical database module is also used to store a plurality of candidate rehabilitation programs and rehabilitation program categories that are stored in association with each other and recommended by the doctor;

[0010] The data analysis module is used to use a pre-deployed artificial intelligence model to identify the current basic data and the current chronic disease indicators, obtain the target rehabilitation program category, and determine the corresponding candidate rehabilitation program according to the target rehabilitation program category to obtain a second target rehabilitation program;

[0011] The rehabilitation guidance module is used to generate rehabilitation guidance suggestions according to the first target rehabilitation plan and the second target rehabilitation plan, and push the rehabilitation guidance suggestions to the user to achieve rehabilitation guidance for chronic diseases.

[0012] Furthermore, the current basic data and current chronic disease indicators corresponding to the patient to be guided are collected, including:

[0013] Collect the age, gender, geographical location, weight and height of the patient to be guided to obtain the current basic data;

[0014] Collect the physical indicators related to the chronic diseases suffered by the patients to be guided, and obtain the current chronic disease indicators.

[0015] Further, the historical database module includes a plurality of storage units;

[0016] Each storage unit stores a chronic disease history, which includes historical basic data at multiple time nodes acquired at a fixed data sampling frequency;

[0017] The chronic disease history also includes rehabilitation medication, rehabilitation training and corresponding rehabilitation time;

[0018] Among them, rehabilitation medication includes medication time and dosage, and rehabilitation training includes training time and dosage.

[0019] Furthermore, the current basic data and the current chronic disease index are matched in the database to determine the closest target historical data, and the rehabilitation program corresponding to the target historical data is determined as the first target rehabilitation program, including:

[0020] Convert the characters in the current basic data and the current chronic disease indicators into numerical values, and normalize them to obtain the normalized current basic data and the current chronic disease indicators;

[0021] The normalized current basic data and current chronic disease indicators form a first eigenvector;

[0022] Convert the characters of the historical data corresponding to each time node in the historical database module into numerical values, and normalize them to obtain the normalized historical data;

[0023] The normalized historical data is used to form the second eigenvector;

[0024] Obtaining the cosine similarity between the first eigenvector and the second eigenvector, and determining a second eigenvector with the largest cosine similarity;

[0025] Determine the target time node corresponding to the second eigenvector according to the second eigenvector with the largest cosine similarity;

[0026] The part of the rehabilitation plan corresponding to the second eigenvector that is located after the target time node is used as the first target rehabilitation plan.

[0027] Furthermore, a pre-deployed artificial intelligence model is used to identify the current basic data and the current chronic disease indicators, obtain the target rehabilitation program category, and determine the corresponding candidate rehabilitation program according to the target rehabilitation program category to obtain a second target rehabilitation program, including:

[0028] Pre-deploy the artificial intelligence model to obtain the pre-deployed artificial intelligence model;

[0029] Obtaining the current basic data and a first feature vector corresponding to the current chronic disease indicator, and using the first feature vector as an input of a pre-deployed artificial intelligence model to obtain a target rehabilitation program category;

[0030] According to the target rehabilitation program category, a corresponding candidate rehabilitation program is determined from a historical database module to obtain a second target rehabilitation program.

[0031] Furthermore, the artificial intelligence model is pre-deployed to obtain the pre-deployed artificial intelligence model, including:

[0032] The radial basis neural network is used to construct an artificial intelligence model, and the model parameters of the artificial intelligence model are randomly initialized;

[0033] Updating the model parameters of the artificial intelligence model to obtain an updated artificial intelligence model;

[0034] Obtain the error function value corresponding to the artificial intelligence model, and determine whether the error function value is less than a preset threshold. If so, deploy the current artificial intelligence model; otherwise, return to the step of updating the model parameters of the artificial intelligence model.

[0035] Furthermore, the error function value corresponding to the artificial intelligence model is obtained as follows:

[0036]

[0037] Where E represents the error function value, n = 1, 2, ..., N, N represents the total number of sample data, m = 1, 2, ..., M, M represents the total number of elements in the output data; ynm It means taking the nth sample data as input and obtaining the mth element in the output data; d nm represents the mth element in the expected output vector corresponding to the nth sample data, d max Indicates the maximum value of the elements in the expected output vector, d min Represents the minimum value of the elements in the expected output vector.

[0038] Furthermore, the model parameters of the artificial intelligence model are updated as follows:

[0039]

[0040] in, represents the adjusted network weight, that is, the network weight corresponding to the jth neuron in the hidden layer during the t+1th training; represents the network weight before adjustment, that is, the network weight corresponding to the jth neuron in the hidden layer during the tth training; η w represents the first learning rate; Indicates the residual signal e corresponding to the neural network model during the tth training n ;e n =∑d nm -y nm , Indicates the input sample data x of the jth neuron in the hidden layer of the neural network model during the tth training. n Response; β w represents the first operation parameter value corresponding to the network weight, and β w The value range of is [0,1]; Indicates the adjustment amount corresponding to the network weight during the tth training; represents the network weight corresponding to the jth neuron in the hidden layer during the t-1th training; when t = 1, represents the optimal network weight; j = 1, 2, ..., J, J represents the total number of neurons in the neural network model; represents the adjusted network center, that is, the network center corresponding to the jth neuron in the hidden layer during the t+1th training; represents the network center before adjustment, that is, the network center corresponding to the jth neuron in the hidden layer during the tth training; η c represents the second learning rate; β c represents the second operation parameter value corresponding to the network center, and β c The value range of is [0,1]; Indicates the adjustment amount corresponding to the network center during the tth training, Indicates the network center corresponding to the jth neuron in the hidden layer during the t-1th training; when t = 1, represents the optimal network center; represents the adjusted network width, that is, the network width corresponding to the jth neuron in the hidden layer during the t+1th training; represents the network width before adjustment, that is, the network width corresponding to the jth neuron in the hidden layer during the tth training; η σ represents the third learning rate, β c represents the third operation parameter value corresponding to the network width, and β c The value range of is [0,1]; Indicates the adjustment amount corresponding to the network width during the tth training; represents the network width corresponding to the jth neuron in the hidden layer during the t-1th training; when t = 1, represents the optimal network width.

[0041] Furthermore, the first learning rate, the second learning rate, and the third learning rate all use dynamic learning rates:

[0042]

[0043] Where η represents the first learning rate, the second learning rate or the third learning rate, g represents the gradient, and ξ represents the learning rate calculation coefficient. represents the symbol for partial derivative, express or

[0044] In another aspect, the present invention provides a rehabilitation guidance method for chronic diseases, comprising:

[0045] Collect the current basic data and current chronic disease indicators corresponding to the patients to be guided;

[0046] Obtaining pre-stored historical data corresponding to chronic diseases and rehabilitation plans corresponding to the historical data, and obtaining a plurality of associated candidate rehabilitation plans and rehabilitation plan categories recommended by pre-stored doctors; wherein the historical data includes historical basic data and historical chronic disease indicators, and the rehabilitation plan includes rehabilitation medication and rehabilitation training;

[0047] Match the current basic data and the current chronic disease indicators in the database, determine the closest target historical data, and determine the rehabilitation plan corresponding to the target historical data as the first target rehabilitation plan;

[0048] Using a pre-deployed artificial intelligence model to identify the current basic data and the current chronic disease indicators, obtain the target rehabilitation program category, and determine the corresponding candidate rehabilitation program according to the target rehabilitation program category to obtain a second target rehabilitation program;

[0049] Rehabilitation guidance suggestions are generated according to the first target rehabilitation plan and the second target rehabilitation plan, and the rehabilitation guidance suggestions are pushed to the user to achieve rehabilitation guidance for chronic diseases.

[0050] The present invention provides a rehabilitation guidance system and method for chronic diseases. By collecting historical data of all patients and the rehabilitation plans corresponding to the historical data, similarity reference is achieved, and the rehabilitation medication and rehabilitation training that best suit the patient's conditions are matched to the patient. At the same time, artificial intelligence technology is used to identify the current basic data and current chronic disease indicators corresponding to the patient to be guided, so as to obtain the rehabilitation medication and rehabilitation training recommended by the doctor, provide patients with more rehabilitation guidance plans, and assist patients in taking effective rehabilitation measures. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0052] Figure 1 A schematic diagram of the structure of a rehabilitation guidance system for chronic diseases provided by an embodiment of the present invention.

[0053] Figure 2 A flowchart of a rehabilitation guidance method for chronic diseases provided in an embodiment of the present invention.

[0054] Among them, 101-data collection module, 102-historical database module, 103-data matching module, 104-data analysis module, 105-rehabilitation guidance module.

[0055] The above drawings have shown clear embodiments of the present invention, which will be described in more detail below. These drawings and text descriptions are not intended to limit the scope of the present invention in any way, but to illustrate the concept of the present invention for those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0056] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Instead, they are merely examples of devices and methods consistent with some aspects of the present invention as detailed in the appended claims.

[0057] The embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0058] Chronic diseases, whose full name is chronic non-communicable diseases, are a type of disease with an insidious onset, a long course and lingering illness. The causes of such diseases are usually complex, and some have not even been fully confirmed. In the existing rehabilitation process of chronic diseases, hospital doctors often give advice and then patients go home to implement it themselves, but patients may forget, especially the elderly are more likely to forget, which leads to the inability to effectively implement rehabilitation measures. Moreover, if patients implement it themselves, the amount of medication and rehabilitation training that may be required may vary at different stages of chronic diseases, which leads to the inability of patients to effectively implement rehabilitation measures. Therefore, an embodiment of the present invention provides a rehabilitation guidance system and method for chronic diseases to assist patients in implementing more effective rehabilitation measures.

[0059] like Figure 1 As shown, the embodiment of the present invention provides a rehabilitation guidance system for chronic diseases, including: a data acquisition module 101, a history database module 102, a data matching module 103, a data analysis module 104 and a rehabilitation guidance module 105;

[0060] The data acquisition module 101 is used to collect the current basic data and the current chronic disease index corresponding to the patient to be guided, and transmit the current basic data and the current chronic disease index to the data matching module;

[0061] The current basic data and current chronic disease indicators can be input by the patient through human-computer interaction, and used as a basis for providing rehabilitation guidance to the patient.

[0062] The historical database module 102 is used to store historical data corresponding to chronic diseases and rehabilitation programs corresponding to the historical data; wherein the historical data includes historical basic data and historical chronic disease indicators, and the rehabilitation program includes rehabilitation medication and rehabilitation training;

[0063] The historical database module 102 stores historical data and the rehabilitation plans corresponding to the historical data. Therefore, it can be matched with the relevant data of the current patient to be guided, so as to match patients with similar conditions, and then select the rehabilitation plan with the shortest rehabilitation time from the historical data of these patients, which helps to assist the patient to be guided in rehabilitation.

[0064] The data matching module 103 is used to match the current basic data and the current chronic disease index in the database, determine the closest target historical data, and determine the rehabilitation program corresponding to the target historical data as the first target rehabilitation program;

[0065] For example, the data can be converted into vector features first, and then based on the vector features, the similarity algorithm is used to determine the most similar historical data, and then the recovery time is judged. It is worth noting that if there is no recovered case in the same medical history, the rehabilitation plan with the mildest condition in the same time period is used as the first target rehabilitation plan to provide a rehabilitation reference for the patient to be guided.

[0066] The historical database module 102 is also used to store a plurality of candidate rehabilitation programs and rehabilitation program categories that are stored in association with each other and recommended by the doctor;

[0067] In order to provide professional advice, historical data at multiple time points can be determined in advance. For each historical data, professional doctors can determine the candidate rehabilitation plans and rehabilitation plan categories. Then, after learning the correlation between historical data and rehabilitation plan categories through an artificial intelligence model, the current basic data and current chronic disease indicators corresponding to the patient to be guided can be analyzed to provide corresponding candidate rehabilitation plans and achieve rehabilitation guidance.

[0068] The data analysis module 104 is used to use a pre-deployed artificial intelligence model to identify the current basic data and the current chronic disease index, obtain the target rehabilitation program category, and determine the corresponding candidate rehabilitation program according to the target rehabilitation program category to obtain a second target rehabilitation program;

[0069] The rehabilitation guidance module 105 is used to generate rehabilitation guidance suggestions according to the first target rehabilitation plan and the second target rehabilitation plan, and push the rehabilitation guidance suggestions to the user to achieve rehabilitation guidance for chronic diseases.

[0070] By generating a first target rehabilitation plan and a second target rehabilitation plan, more guidance and suggestions can be provided to patients.

[0071] In the embodiment of the present invention, the current basic data and current chronic disease indicators corresponding to the patient to be guided are collected, including:

[0072] Collect the age, gender, geographical location, weight and height of the patient to be guided to obtain the current basic data;

[0073] Collect the physical indicators related to the chronic diseases suffered by the patients to be guided, and obtain the current chronic disease indicators.

[0074] In an embodiment of the present invention, the historical database module includes a plurality of storage units;

[0075] Each storage unit stores a chronic disease history, which includes historical basic data at multiple time nodes acquired at a fixed data sampling frequency;

[0076] The chronic disease history also includes rehabilitation medication, rehabilitation training and corresponding rehabilitation time;

[0077] Among them, rehabilitation medication includes medication time and dosage, and rehabilitation training includes training time and dosage.

[0078] In an embodiment of the present invention, the current basic data and the current chronic disease index are matched in the database, the closest target historical data is determined, and the rehabilitation program corresponding to the target historical data is determined as the first target rehabilitation program, including:

[0079] Convert the characters in the current basic data and the current chronic disease indicators into numerical values, and normalize them to obtain the normalized current basic data and the current chronic disease indicators;

[0080] The normalized current basic data and current chronic disease indicators form a first eigenvector;

[0081] Convert the characters of the historical data corresponding to each time node in the historical database module into numerical values, and normalize them to obtain the normalized historical data;

[0082] The normalized historical data is used to form the second eigenvector;

[0083] Obtaining the cosine similarity between the first eigenvector and the second eigenvector, and determining a second eigenvector with the largest cosine similarity;

[0084] Determine the target time node corresponding to the second eigenvector according to the second eigenvector with the largest cosine similarity;

[0085] The part of the rehabilitation plan corresponding to the second eigenvector that is located after the target time node is used as the first target rehabilitation plan.

[0086] Optionally, obtaining the cosine similarity between the first feature vector and the second feature vector, and determining a second feature vector with the largest cosine similarity, includes:

[0087] Obtaining cosine similarity between the first eigenvector and the second eigenvector, and determining the maximum cosine similarity;

[0088] It is determined whether the number of the largest cosine similarities is greater than 1. If so, the second eigenvector with the shortest recovery time is obtained to obtain a second eigenvector with the largest cosine similarity. Otherwise, a second eigenvector with the largest cosine similarity is directly obtained.

[0089] Through the above method, patients can quickly find rehabilitation plans similar to their own conditions, and as the data in the database increases, the data will become more and more comprehensive, so the matching accuracy will gradually increase, providing patients with more effective rehabilitation guidance.

[0090] Optionally, generating rehabilitation guidance suggestions according to the first target rehabilitation program and the second target rehabilitation program may include: if the maximum cosine similarity between the first feature vector and the second feature vector exceeds a preset first threshold, the first target rehabilitation program is used as the rehabilitation guidance suggestion; if the maximum cosine similarity between the first feature vector and the second feature vector is less than a preset second threshold, the second target rehabilitation program is used as the rehabilitation guidance suggestion; if the maximum cosine similarity between the first feature vector and the second feature vector is between the first threshold and the second threshold, the first target rehabilitation program and the second target rehabilitation program are used together as the rehabilitation guidance suggestion. Wherein, the first threshold is greater than the second threshold.

[0091] In an embodiment of the present invention, a pre-deployed artificial intelligence model is used to identify the current basic data and the current chronic disease index, obtain the target rehabilitation program category, and determine the corresponding candidate rehabilitation program according to the target rehabilitation program category to obtain a second target rehabilitation program, including:

[0092] Pre-deploy the artificial intelligence model to obtain the pre-deployed artificial intelligence model;

[0093] Obtaining the current basic data and a first feature vector corresponding to the current chronic disease indicator, and using the first feature vector as an input of a pre-deployed artificial intelligence model to obtain a target rehabilitation program category;

[0094] According to the target rehabilitation program category, a corresponding candidate rehabilitation program is determined from a historical database module to obtain a second target rehabilitation program.

[0095] In an embodiment of the present invention, the artificial intelligence model is pre-deployed to obtain the pre-deployed artificial intelligence model, including:

[0096] The radial basis neural network (BRF neural network) is used to build an artificial intelligence model, and the model parameters of the artificial intelligence model are randomly initialized;

[0097] It is worth noting that in addition to using radial basis neural networks to build artificial intelligence models, other neural networks can also be used to build artificial intelligence models.

[0098] Updating the model parameters of the artificial intelligence model to obtain an updated artificial intelligence model;

[0099] For example, an optimization algorithm such as a gradient descent method, a particle swarm algorithm, etc. may be used to update the model parameters of the artificial intelligence model.

[0100] Obtain the error function value corresponding to the artificial intelligence model, and determine whether the error function value is less than a preset threshold. If so, deploy the current artificial intelligence model; otherwise, return to the step of updating the model parameters of the artificial intelligence model.

[0101] In an embodiment of the present invention, the error function value corresponding to the artificial intelligence model is obtained:

[0102]

[0103] Where E represents the error function value, n = 1, 2, ..., N, N represents the total number of sample data, m = 1, 2, ..., M, M represents the total number of elements in the output data; y nm It means taking the nth sample data as input and obtaining the mth element in the output data; d nm represents the mth element in the expected output vector corresponding to the nth sample data, d max Indicates the maximum value of the elements in the expected output vector, d min Represents the minimum value of the elements in the expected output vector.

[0104] In an embodiment of the present invention, the model parameters of the artificial intelligence model are updated as follows:

[0105]

[0106] in, represents the adjusted network weight, that is, the network weight corresponding to the jth neuron in the hidden layer during the t+1th training; represents the network weight before adjustment, that is, the network weight corresponding to the jth neuron in the hidden layer during the tth training; η w represents the first learning rate; Indicates the residual signal e corresponding to the neural network model during the tth training n ;en =∑d nm -y nm , Indicates the input sample data x of the jth neuron in the hidden layer of the neural network model during the tth training. n Response; β w represents the first operation parameter value corresponding to the network weight, and β w The value range of is [0,1]; Indicates the adjustment amount corresponding to the network weight during the tth training; represents the network weight corresponding to the jth neuron in the hidden layer during the t-1th training; when t = 1, represents the optimal network weight; j = 1, 2, ..., J, J represents the total number of neurons in the neural network model; represents the adjusted network center, that is, the network center corresponding to the jth neuron in the hidden layer during the t+1th training; represents the network center before adjustment, that is, the network center corresponding to the jth neuron in the hidden layer during the tth training; η c represents the second learning rate; β c represents the second operation parameter value corresponding to the network center, and β c The value range of is [0,1]; Indicates the adjustment amount corresponding to the network center during the tth training, Indicates the network center corresponding to the jth neuron in the hidden layer during the t-1th training; when t = 1, represents the optimal network center; represents the adjusted network width, that is, the network width corresponding to the jth neuron in the hidden layer during the t+1th training; represents the network width before adjustment, that is, the network width corresponding to the jth neuron in the hidden layer during the tth training; η σ represents the third learning rate, β c represents the third operation parameter value corresponding to the network width, and β c The value range of is [0,1]; Indicates the adjustment amount corresponding to the network width during the tth training; represents the network width corresponding to the jth neuron in the hidden layer during the t-1th training; when t = 1, represents the optimal network width.

[0107] In the embodiment of the present invention, the first learning rate, the second learning rate and the third learning rate all adopt dynamic learning rates:

[0108]

[0109] Where η represents the first learning rate, the second learning rate or the third learning rate, g represents the gradient, and ξ represents the learning rate calculation coefficient. represents the symbol for partial derivative, express or

[0110] like Figure 2 As shown, an embodiment of the present invention provides a rehabilitation guidance method for chronic diseases, comprising:

[0111] S201, collecting current basic data and current chronic disease indicators corresponding to the patient to be guided;

[0112] S202, obtaining pre-stored historical data corresponding to chronic diseases and rehabilitation plans corresponding to the historical data, and obtaining a plurality of associated candidate rehabilitation plans and rehabilitation plan categories recommended by pre-stored doctors; wherein the historical data includes historical basic data and historical chronic disease indicators, and the rehabilitation plan includes rehabilitation medication and rehabilitation training;

[0113] S203, matching the current basic data and the current chronic disease index in the database, determining the closest target historical data, and determining the rehabilitation program corresponding to the target historical data as the first target rehabilitation program;

[0114] S204, using a pre-deployed artificial intelligence model to identify the current basic data and the current chronic disease indicators, obtain a target rehabilitation program category, and determine a corresponding candidate rehabilitation program according to the target rehabilitation program category to obtain a second target rehabilitation program;

[0115] S205: Generate rehabilitation guidance suggestions according to the first target rehabilitation plan and the second target rehabilitation plan, and push the rehabilitation guidance suggestions to the user to implement rehabilitation guidance for chronic diseases.

[0116] A rehabilitation guidance method for chronic diseases provided in an embodiment of the present invention can be executed by the above-mentioned system. Its principles and beneficial effects are similar and will not be described in detail here.

[0117] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented by software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer programs or instructions. When the computer program or instruction is loaded and executed on a computer, the process or function described in the embodiment of the present invention is executed in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, a network device, a user device or other programmable device. The computer program or instruction can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer program or instruction can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired or wireless means. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. that integrates one or more available media. The available medium can be a magnetic medium, such as a floppy disk, a hard disk, or a tape; it can also be an optical medium, such as a digital video disc (DVD); it can also be a semiconductor medium, such as a solid state drive (SSD).

[0118] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed by the present invention, and these modifications or replacements should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A rehabilitation guidance system for chronic diseases, characterized in that: include: Data collection module, historical database module, data matching module, data analysis module and rehabilitation guidance module; The data acquisition module is used to collect the current basic data and current chronic disease indicators corresponding to the patient to be guided, and transmit the current basic data and current chronic disease indicators to the data matching module; The historical database module is used to store historical data corresponding to chronic diseases and rehabilitation programs corresponding to the historical data; wherein the historical data includes historical basic data and historical chronic disease indicators, and the rehabilitation program includes rehabilitation medication and rehabilitation training; The data matching module is used to match the current basic data and the current chronic disease index in the database, determine the closest target historical data, and determine the rehabilitation program corresponding to the target historical data as the first target rehabilitation program; The historical database module is also used to store a plurality of candidate rehabilitation programs and rehabilitation program categories that are stored in association with each other and recommended by the doctor; The data analysis module is used to use a pre-deployed artificial intelligence model to identify the current basic data and the current chronic disease indicators, obtain the target rehabilitation program category, and determine the corresponding candidate rehabilitation program according to the target rehabilitation program category to obtain a second target rehabilitation program; The rehabilitation guidance module is used to generate rehabilitation guidance suggestions according to the first target rehabilitation plan and the second target rehabilitation plan, and push the rehabilitation guidance suggestions to the user to achieve rehabilitation guidance for chronic diseases.

2. The rehabilitation guidance system for chronic diseases according to claim 1, characterized in that: Collect the current basic data and current chronic disease indicators corresponding to the patients to be guided, including: Collect the age, gender, geographical location, weight and height of the patient to be guided to obtain the current basic data; Collect the physical indicators related to the chronic diseases suffered by the patients to be guided, and obtain the current chronic disease indicators.

3. The rehabilitation guidance system for chronic diseases according to claim 1, characterized in that: The historical database module includes a plurality of storage units; Each storage unit stores a chronic disease history, which includes historical basic data at multiple time nodes acquired at a fixed data sampling frequency; The chronic disease history also includes rehabilitation medication, rehabilitation training and corresponding rehabilitation time; Among them, rehabilitation medication includes medication time and dosage, and rehabilitation training includes training time and dosage.

4. The rehabilitation guidance system for chronic diseases according to claim 3, characterized in that: The current basic data and the current chronic disease indicators are matched in the database to determine the closest target historical data, and the rehabilitation program corresponding to the target historical data is determined as the first target rehabilitation program, including: Convert the characters in the current basic data and the current chronic disease indicators into numerical values, and normalize them to obtain the normalized current basic data and the current chronic disease indicators; The normalized current basic data and current chronic disease indicators form a first eigenvector; Convert the characters of the historical data corresponding to each time node in the historical database module into numerical values, and normalize them to obtain the normalized historical data; The normalized historical data is used to form the second eigenvector; Obtaining the cosine similarity between the first eigenvector and the second eigenvector, and determining a second eigenvector with the largest cosine similarity; Determine the target time node corresponding to the second eigenvector according to the second eigenvector with the largest cosine similarity; The part of the rehabilitation plan corresponding to the second eigenvector that is located after the target time node is used as the first target rehabilitation plan.

5. The rehabilitation guidance system for chronic diseases according to claim 4, characterized in that: The current basic data and the current chronic disease indicators are identified by using a pre-deployed artificial intelligence model, the target rehabilitation program category is obtained, and the corresponding candidate rehabilitation program is determined according to the target rehabilitation program category to obtain a second target rehabilitation program, including: Pre-deploy the artificial intelligence model to obtain the pre-deployed artificial intelligence model; Obtaining the current basic data and a first feature vector corresponding to the current chronic disease indicator, and using the first feature vector as an input of a pre-deployed artificial intelligence model to obtain a target rehabilitation program category; According to the target rehabilitation program category, a corresponding candidate rehabilitation program is determined from a historical database module to obtain a second target rehabilitation program.

6. The rehabilitation guidance system for chronic diseases according to claim 5, characterized in that: The artificial intelligence model is pre-deployed to obtain the pre-deployed artificial intelligence model, including: The radial basis neural network is used to construct an artificial intelligence model, and the model parameters of the artificial intelligence model are randomly initialized; Updating the model parameters of the artificial intelligence model to obtain an updated artificial intelligence model; Obtain the error function value corresponding to the artificial intelligence model, and determine whether the error function value is less than a preset threshold. If so, deploy the current artificial intelligence model; otherwise, return to the step of updating the model parameters of the artificial intelligence model.

7. The rehabilitation guidance system for chronic diseases according to claim 6, characterized in that: The error function value corresponding to the artificial intelligence model is obtained as follows: Where E represents the error function value, n = 1, 2, ..., N, N represents the total number of sample data, m = 1, 2, ..., M, M represents the total number of elements in the output data; y nm It means taking the nth sample data as input and obtaining the mth element in the output data; d nm represents the mth element in the expected output vector corresponding to the nth sample data, d max Indicates the maximum value of the elements in the expected output vector, d min Represents the minimum value of the elements in the expected output vector.

8. The rehabilitation guidance system for chronic diseases according to claim 7, characterized in that: The model parameters of the artificial intelligence model are updated as follows: in, represents the adjusted network weight, that is, the network weight corresponding to the jth neuron in the hidden layer during the t+1th training; represents the network weight before adjustment, that is, the network weight corresponding to the jth neuron in the hidden layer during the tth training; η w represents the first learning rate; Indicates the residual signal e corresponding to the neural network model during the tth training n ;e n =∑d nm -y nm , Indicates the input sample data x of the jth neuron in the hidden layer of the neural network model during the tth training. n Response; β w represents the first operation parameter value corresponding to the network weight, and β w The value range of is [0,1]; Indicates the adjustment amount corresponding to the network weight during the tth training; represents the network weight corresponding to the jth neuron in the hidden layer during the t-1th training; when t = 1, represents the optimal network weight; j = 1, 2, ..., J, J represents the total number of neurons in the neural network model; represents the adjusted network center, that is, the network center corresponding to the jth neuron in the hidden layer during the t+1th training; represents the network center before adjustment, that is, the network center corresponding to the jth neuron in the hidden layer during the tth training; η c represents the second learning rate; β c represents the second operation parameter value corresponding to the network center, and β c The value range of is [0,1]; Indicates the adjustment amount corresponding to the network center during the tth training, Indicates the network center corresponding to the jth neuron in the hidden layer during the t-1th training; when t = 1, represents the optimal network center; represents the adjusted network width, that is, the network width corresponding to the jth neuron in the hidden layer during the t+1th training; represents the network width before adjustment, that is, the network width corresponding to the jth neuron in the hidden layer during the tth training; η σ represents the third learning rate, β c represents the third operation parameter value corresponding to the network width, and β c The value range of is [0,1]; Indicates the adjustment amount corresponding to the network width during the tth training; represents the network width corresponding to the jth neuron in the hidden layer during the t-1th training; when t=1, represents the optimal network width.

9. The rehabilitation guidance system for chronic diseases according to claim 8, characterized in that: The first learning rate, the second learning rate, and the third learning rate all use dynamic learning rates: Where η represents the first learning rate, the second learning rate or the third learning rate, g represents the gradient, and ξ represents the learning rate calculation coefficient. represents the symbol for partial derivative, express or 10. A rehabilitation guidance method for chronic diseases, characterized in that: include: Collect the current basic data and current chronic disease indicators corresponding to the patients to be guided; Obtaining pre-stored historical data corresponding to chronic diseases and rehabilitation plans corresponding to the historical data, and obtaining a plurality of associated candidate rehabilitation plans and rehabilitation plan categories recommended by pre-stored doctors; wherein the historical data includes historical basic data and historical chronic disease indicators, and the rehabilitation plan includes rehabilitation medication and rehabilitation training; Match the current basic data and the current chronic disease indicators in the database, determine the closest target historical data, and determine the rehabilitation plan corresponding to the target historical data as the first target rehabilitation plan; Using a pre-deployed artificial intelligence model to identify the current basic data and the current chronic disease indicators, obtain the target rehabilitation program category, and determine the corresponding candidate rehabilitation program according to the target rehabilitation program category to obtain a second target rehabilitation program; Rehabilitation guidance suggestions are generated according to the first target rehabilitation plan and the second target rehabilitation plan, and the rehabilitation guidance suggestions are pushed to the user to achieve rehabilitation guidance for chronic diseases.

Images