Diabetes health management method and system based on BP neural network
Through the diabetes health management method based on BP neural network, patient data is obtained for preprocessing and model training, and a personalized health management solution is generated, which solves the problem that personalized management cannot be provided in the existing technology, and achieves precise diabetes management and treatment effects.
Patent Information
- Application Number
- CN202510581177.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-05-07
- Publication Date
- 2025-08-15
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Existing diabetes health management technologies cannot achieve scientific and comprehensive provision of personalized predictions, interventions and treatment options for each patient.
A diabetes health management method based on BP neural network is adopted to generate personalized health management plans, including comprehensive measures such as diet, exercise, nutrients and psychological intervention by obtaining patient data, preprocessing, data division, model training and classification identification.
Accurate diabetes health management has been achieved, patients' prediction, intervention and treatment effects have been improved, and medical staff's diagnosis and treatment capabilities and work efficiency have been improved.
Smart Images
Figure CN120496741A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of diabetes health management, and in particular to a diabetes health management method and system based on BP neural network. Background Art
[0002] Diabetes is a chronic disease characterized by hyperglycemia, caused by an absolute or relative deficiency in insulin secretion and impaired utilization. Clinically, diabetes is divided into four types: insulin-dependent (type 1), non-insulin-dependent (type 2), gestational diabetes, and special types of diabetes. The etiology is primarily attributed to a combination of genetic and environmental factors, including pancreatic cell dysfunction leading to decreased insulin secretion, insensitivity to insulin, or both, which prevents the effective utilization and storage of glucose in the blood.
[0003] Metabolic disorders can lead to chronically high blood sugar levels. This disease is often accompanied by multiple complications, and chronically high blood sugar levels can severely harm the kidneys, cardiovascular system, and nervous system, significantly impacting both physical and mental health. The incidence of diabetes is increasing year by year and is occurring at younger ages, severely impacting public health and safety in my country and around the world. Therefore, research on diabetes prediction is necessary.
[0004] In actual diabetes clinical treatment, the causes and symptoms of diabetes vary from patient to patient, as do individual physical conditions and lifestyles. In this context, it can be extremely complex for healthcare professionals to develop tailored health management plans for each patient. Existing technologies cannot scientifically and comprehensively manage patients' diabetes health, let alone provide targeted prediction, intervention, and treatment plans.
[0005] Therefore, it is an urgent problem for those skilled in the art to propose a diabetes health management method and system based on BP neural network to solve the difficulties existing in the prior art. Summary of the Invention
[0006] In view of this, the present invention provides a diabetes health management method and system based on BP neural network to achieve precise health management, so that diabetic patients can obtain effective prediction, intervention and treatment.
[0007] In order to achieve the above object, the present invention adopts the following technical solutions:
[0008] A diabetes health management method based on BP neural network includes the following steps:
[0009] S1. Acquire data: Acquire information data of diabetic patients;
[0010] S2. Data preprocessing: Preprocessing the acquired diabetes patient information data to obtain preprocessed diabetes data;
[0011] S3. Data Partitioning: Divide the preprocessed diabetes data into training and test sets.
[0012] S4. Model training: Input the training set into the health management model to train the parameters in the health management model; update the weight parameters through the loss function. After several training cycles, a trained health management model is obtained;
[0013] S5. Classification and Recognition: Input the test set into the trained health management model to evaluate the diabetes health management results;
[0014] S6. Management plan: Develop a diabetes management plan for diabetic patients based on the results of diabetes health management.
[0015] In the above method, optionally, the specific content of obtaining the information data of the diabetic patient in S1 is: obtaining the health management data in the health file of the diabetic patient and the diabetes survey information data input by the questionnaire.
[0016] The above method may optionally perform data elimination, data cleaning, and data transformation preprocessing on the health management data obtained from the health records of the diabetic patients and the diabetes survey information data input by the questionnaire to obtain preprocessed diabetes data.
[0017] In the above method, optionally, in S3 , 80% of the pre-processed diabetes data is divided into a training set and 20% is divided into a test set.
[0018] In the above method, optionally, in S4, a health management model based on a BP neural network is constructed, and the specific content of training the parameters in the health management model is as follows:
[0019] S41 initializes the network and learning parameters, sets the network initial weight matrix and learning factors;
[0020] S42 inputs the training set into the BP neural network for training until the learning requirements are met;
[0021] S43 forward propagation process: for a given training pattern input, calculate the network output pattern and compare it with the expected pattern. If there is an error, execute S44; otherwise, return to S42;
[0022] S44 back propagation process: calculate the error of the units in the same layer, correct the weights and thresholds, and return to S42.
[0023] In the above method, the specific content of formulating a diabetes management plan for the diabetic patient based on the diabetes health management results in S6 is optional:
[0024] Form health records based on health management results, and form labels based on health records;
[0025] According to the label, one or more combinations of special dietary intervention plans, anti-glycation intervention plans, daily dietary intervention plans, exercise intervention plans, nutrient intervention plans, acupressure intervention plans, anti-aging intervention plans and psychological intervention plans are carried out.
[0026] A diabetes health management system based on BP neural network, applying any of the above-mentioned diabetes health management methods based on BP neural network, comprising: a data acquisition module, a data preprocessing module, a data partitioning module, a model training module, a classification and recognition module, and a management solution module;
[0027] A data acquisition module is connected to the input end of the data preprocessing module and is used to obtain information data of diabetic patients;
[0028] A data preprocessing module is connected to the input end of the data partitioning module and is used to preprocess the acquired diabetic patient information data to obtain preprocessed diabetic data;
[0029] A data partitioning module, connected to the input end of the model training module, is used to divide the preprocessed diabetes data into a training set and a test set;
[0030] The model training module is connected to the input end of the classification and recognition module, and is used to input the training set into the health management model and train the parameters in the health management model; the weight parameters are updated through the loss function, and after several training cycles, a trained health management model is obtained;
[0031] The classification and recognition module is connected to the input end of the management solution module and is used to input the test set into the trained health management model to evaluate the diabetes health management results;
[0032] The management plan module is used to formulate diabetes management plans for diabetic patients based on the results of diabetes health management.
[0033] It can be seen from the above technical solutions that, compared with the prior art, the present invention provides a diabetes health management method and system based on BP neural network, which has the following beneficial effects:
[0034] (1) The present invention collects patient data and information, conducts diabetes risk assessment through a health management model, and then generates a comprehensive health management plan based on a combination of special dietary intervention, nutrient intervention, and daily diet and exercise intervention by a health manager, thereby achieving precise health management and enabling diabetic patients to obtain effective prediction, intervention, and treatment;
[0035] (2) The present invention helps medical staff to judge the diabetic status of patients, understand the basic situation, analyze the health status, and improve diagnosis and treatment capabilities and work efficiency. BRIEF DESCRIPTION OF THE DRAWINGS
[0036] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are merely embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.
[0037] Figure 1 A flow chart of a diabetes health management method based on BP neural network provided by the present invention;
[0038] Figure 2 This is a structural block diagram of a diabetes health management system based on BP neural network provided by the present invention. DETAILED DESCRIPTION
[0039] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. 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 making creative efforts are within the scope of protection of the present invention.
[0040] Reference Figure 1 As shown, the present invention discloses a diabetes health management method based on BP neural network, comprising the following steps:
[0041] S1. Acquire data: Acquire information data of diabetic patients;
[0042] S2. Data preprocessing: Preprocessing the acquired diabetes patient information data to obtain preprocessed diabetes data;
[0043] S3. Data Partitioning: Divide the preprocessed diabetes data into training and test sets.
[0044] S4. Model training: Input the training set into the health management model to train the parameters in the health management model; update the weight parameters through the loss function. After several training cycles, a trained health management model is obtained;
[0045] S5. Classification and Recognition: Input the test set into the trained health management model to evaluate the diabetes health management results;
[0046] S6. Management plan: Develop a diabetes management plan for diabetic patients based on the results of diabetes health management.
[0047] Furthermore, the specific content of obtaining the information data of the diabetic patient in S1 is: obtaining the health management data in the health file of the diabetic patient and the diabetes survey information data input by the questionnaire.
[0048] Furthermore, the health management data in the health files of the diabetic patients and the diabetes survey information data input by the questionnaire are pre-processed by data elimination, data cleaning, and data transformation to obtain pre-processed diabetes data.
[0049] Furthermore, in S3, 80% of the preprocessed diabetes data is divided into a training set and 20% is divided into a test set.
[0050] Furthermore, in S4, a health management model based on BP neural network is constructed, and the specific content of training the parameters in the health management model is as follows:
[0051] S41 initializes the network and learning parameters, sets the network initial weight matrix and learning factors;
[0052] S42 inputs the training set into the BP neural network for training until the learning requirements are met;
[0053] S43 forward propagation process: for a given training pattern input, calculate the network output pattern and compare it with the expected pattern. If there is an error, execute S44; otherwise, return to S42;
[0054] S44 back propagation process: calculate the error of the units in the same layer, correct the weights and thresholds, and return to S42.
[0055] Specifically, the loss function in S4 adopts the cross entropy loss function or the L1Loss loss function.
[0056] Specifically, the BP network consists of an input layer, a hidden layer, and an output layer.
[0057] Input layer: the input end of information, used to read input data; hidden layer: the processing end of information, the number of layers of this hidden layer can be set, one hidden layer, q neurons; output layer: the output end of information, used to output results;
[0058] The process of BP neural network is mainly divided into two stages. The first stage is the forward propagation of the signal, from the input layer through the hidden layer, and finally to the output layer; the second stage is the back propagation of the error, from the output layer to the hidden layer, and finally to the input layer, adjusting the weights and bias from the hidden layer to the output layer, and the weights and bias from the input layer to the hidden layer in turn.
[0059] Furthermore, the specific content of formulating a diabetes management plan for diabetic patients based on the diabetes health management results in S6 is as follows:
[0060] Form health records based on health management results, and form labels based on health records;
[0061] According to the label, one or more combinations of special dietary intervention plans, anti-glycation intervention plans, daily dietary intervention plans, exercise intervention plans, nutrient intervention plans, acupressure intervention plans, anti-aging intervention plans and psychological intervention plans are carried out.
[0062] In a specific embodiment, the dataset used is from the UCI Machine Learning Repository. According to the description of the UCI Machine Learning Repository, the data in the dataset meets the following information: 1) experience of hospitalized patients; 2) encounter with "diabetes", that is, any type of diabetes is entered into the system as a diagnosis during this period; 3) the hospitalization time is at least 1 day and at most 14 days; 4) experimental tests are performed during hospitalization; 5) drug treatment is performed during hospitalization.
[0063] The health management data from the health records of diabetic patients and the diabetes survey information data entered in the questionnaire were obtained, and labels were set for each patient. The labels included patient number, race, gender, age, admission type, length of stay, number of examinations, glycated hemoglobin test results, insulin release test results, number of medications, diabetes medications, outpatient diagnosis records, number of emergency visits, and other attributes; data from 1,205 diabetic patients were used in advance.
[0064] The health management data in the health records of diabetic patients and the diabetes survey information data input by the questionnaire are preprocessed by data elimination, data cleaning and data transformation to obtain preprocessed diabetes data.
[0065] 80% of the preprocessed diabetes data was divided into a training set and 20% was divided into a test set.
[0066] The BP neural network is trained according to the training set, the loss function value is reduced and the network weight is updated. After continuous iterative learning, the trained BP neural network volume is obtained and a health management model is established.
[0067] The test set is input into the trained health management model to evaluate the diabetes health management results.
[0068] Develop diabetes management plans for diabetic patients based on diabetes health management results.
[0069] A health record is formed based on the results of health management, and a label is formed based on the health record; based on the label, one or more combinations of special dietary intervention plans, anti-glycation intervention plans, daily diet intervention plans, exercise intervention plans, nutrient intervention plans, acupoint massage intervention plans, anti-aging intervention plans and psychological intervention plans are carried out.
[0070] The present invention collects the patient's physical data information, conducts diabetes risk assessment through a health management model, and then generates a comprehensive health management plan based on a combination of special dietary therapy interventions, nutrient interventions, and daily diet and exercise interventions by a health manager, thereby achieving precise health management and enabling diabetic patients to receive effective prediction, intervention and treatment.
[0071] and Figure 1 Corresponding to the method described above, the embodiment of the present invention also provides a diabetes health management system based on BP neural network, the structural diagram of which is shown in FIG. Figure 2 As shown, it includes: data acquisition module, data preprocessing module, data partitioning module, model training module, classification and recognition module and management solution module;
[0072] A data acquisition module is connected to the input end of the data preprocessing module and is used to obtain information data of diabetic patients;
[0073] A data preprocessing module is connected to the input end of the data partitioning module and is used to preprocess the acquired diabetic patient information data to obtain preprocessed diabetic data;
[0074] A data partitioning module, connected to the input end of the model training module, is used to divide the preprocessed diabetes data into a training set and a test set;
[0075] The model training module is connected to the input end of the classification and recognition module, and is used to input the training set into the health management model and train the parameters in the health management model; the weight parameters are updated through the loss function, and after several training cycles, a trained health management model is obtained;
[0076] The classification and recognition module is connected to the input end of the management solution module and is used to input the test set into the trained health management model to evaluate the diabetes health management results;
[0077] The management plan module is used to formulate diabetes management plans for diabetic patients based on the results of diabetes health management.
[0078] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. Reference can be made to the common and similar parts between the various embodiments. For the systems disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple, and the relevant parts can be referred to the method description.
[0079] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not limited to the embodiments shown herein but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.
Claims
1. A diabetes health management method based on BP neural network, characterized in that: The following steps are involved: S1. Acquire data: Acquire information data of diabetic patients; S2. Data preprocessing: Preprocessing the acquired diabetes patient information data to obtain preprocessed diabetes data; S3. Data Partitioning: Divide the preprocessed diabetes data into training and test sets. S4. Model training: Input the training set into the health management model to train the parameters in the health management model; update the weight parameters through the loss function. After several training cycles, a trained health management model is obtained; S5. Classification and Recognition: Input the test set into the trained health management model to evaluate the diabetes health management results; S6. Management plan: Develop a diabetes management plan for diabetic patients based on the results of diabetes health management.
2. A diabetes health management method based on BP neural network according to claim 1, characterized in that: The specific content of obtaining the information data of the diabetic patient in S1 is: obtaining the health management data in the health file of the diabetic patient and the diabetes survey information data input by the questionnaire.
3. A diabetes health management method based on BP neural network according to claim 2, characterized in that: The health management data in the health records of diabetic patients and the diabetes survey information data input by the questionnaire are preprocessed by data elimination, data cleaning and data transformation to obtain preprocessed diabetes data.
4. The diabetes health management method based on BP neural network according to claim 1, characterized in that: In S3, 80% of the preprocessed diabetes data is divided into a training set and 20% is divided into a test set.
5. The diabetes health management method based on BP neural network according to claim 1, characterized in that: In S4, a health management model based on BP neural network is constructed. The specific content of training the parameters in the health management model is as follows: S41 initializes the network and learning parameters, sets the network initial weight matrix and learning factors; S42 inputs the training set into the BP neural network for training until the learning requirements are met; S43 forward propagation process: for a given training pattern input, calculate the network output pattern and compare it with the expected pattern. If there is an error, execute S44; Otherwise, return to S42; S44 back propagation process: calculate the error of the units in the same layer, correct the weights and thresholds, and return to S42.
6. The diabetes health management method based on BP neural network according to claim 1, characterized in that: The specific contents of formulating a diabetes management plan for diabetic patients based on the diabetes health management results in S6 are as follows: Form health records based on health management results, and form labels based on health records; According to the label, one or more combinations of special dietary intervention plans, anti-glycation intervention plans, daily dietary intervention plans, exercise intervention plans, nutrient intervention plans, acupressure intervention plans, anti-aging intervention plans and psychological intervention plans are carried out.
7. A diabetes health management system based on BP neural network, characterized in that: A diabetes health management method based on a BP neural network according to any one of claims 1 to 6 is applied, comprising: a data acquisition module, a data preprocessing module, a data partitioning module, a model training module, a classification and recognition module, and a management solution module; A data acquisition module is connected to the input end of the data preprocessing module and is used to obtain information data of diabetic patients; A data preprocessing module is connected to the input end of the data partitioning module and is used to preprocess the acquired diabetic patient information data to obtain preprocessed diabetic data; A data partitioning module, connected to the input end of the model training module, is used to divide the preprocessed diabetes data into a training set and a test set; The model training module is connected to the input end of the classification and recognition module, and is used to input the training set into the health management model and train the parameters in the health management model; the weight parameters are updated through the loss function, and after several training cycles, a trained health management model is obtained; The classification and recognition module is connected to the input end of the management solution module and is used to input the test set into the trained health management model to evaluate the diabetes health management results; The management plan module is used to formulate diabetes management plans for diabetic patients based on the results of diabetes health management.