Diabetes early warning method and system based on machine learning

Abstract

The invention provides a diabetes early warning method and system based on machine learning. The method comprises the steps of collecting multi-dimensional data of a target user; extracting specified features related to diabetes from the multi-dimensional data of the target user; and inputting the specified features into a preset diabetes early-warning model for prediction to obtain a diabetes early-warning result, the diabetes early-warning model being obtained based on a sample data training machine learning model. According to the scheme, the specified features related to diabetes are extracted from the multi-dimensional data of the target user, and the specified features are input into the diabetes early warning model for prediction to obtain the diabetes early warning result, namely early warning of diabetes is performed through the machine learning model, and the user is helped to perform prevention and intervention in advance.

Description

Technical Field

[0001] The present invention relates to the field of medical artificial intelligence technology, and in particular to a diabetes early warning method and system based on machine learning. Background Art

[0002] With the improvement of living standards, diabetes has become one of the major health problems worldwide. Diabetes usually has a long incubation period and complex pathogenesis. Early detection and intervention are crucial to improving patient prognosis.

[0003] However, most diabetic patients will only seek medical examination after their bodies have obvious reactions. At this time, diabetes has usually developed to the middle and late stages. There is an urgent need for a method that can provide early warning of diabetes. Summary of the Invention

[0004] In view of this, an embodiment of the present invention provides a diabetes warning method and system based on machine learning to provide early warning of diabetes.

[0005] To achieve the above objectives, the embodiments of the present invention provide the following technical solutions:

[0006] A first aspect of an embodiment of the present invention discloses a diabetes early warning method based on machine learning, the method comprising:

[0007] Collect multi-dimensional data of target users;

[0008] extracting designated features related to diabetes from the multi-dimensional data of the target user;

[0009] The designated features are input into a preset diabetes early warning model for prediction to obtain a diabetes early warning result. The diabetes early warning model is obtained by training a machine learning model based on sample data. The diabetes early warning result includes: a warning level for at least one type of diabetes.

[0010] Preferably, the process of training a machine learning model based on sample data to obtain a diabetes early warning model includes:

[0011] Collecting sample data, the sample data at least including: multi-dimensional data of diabetic patients and non-diabetic patients;

[0012] extracting sample-specified features related to diabetes from the sample data;

[0013] Using the sample specified features corresponding to the sample data, a machine learning model is trained until the machine learning model converges to obtain a diabetes early warning model, where the machine learning model is a logistic regression model, a decision tree model, a support vector machine model, or a neural network model.

[0014] Preferably, the sample-specified features corresponding to the sample data are used to train a machine learning model until the machine learning model converges to obtain a diabetes early warning model, including:

[0015] Dividing the sample data into a training set and a test set;

[0016] Using the sample specified features corresponding to the training set, training a machine learning model until the machine learning model converges to obtain a diabetes early warning model;

[0017] The diabetes early warning model is verified using the sample-specified features corresponding to the test set.

[0018] Preferably, before extracting the sample-specified features related to diabetes from the sample data, the method further includes:

[0019] The sample data is preprocessed, and the preprocessing includes at least data cleaning, missing value processing, and feature normalization.

[0020] Preferably, after obtaining the diabetes warning result, the method further includes:

[0021] Obtaining corresponding recommended plans from a preset diabetes knowledge base based on the diabetes early warning result, wherein the recommended plans include at least a diet plan, a prevention plan, and an exercise plan;

[0022] Output the proposed solution.

[0023] A second aspect of an embodiment of the present invention discloses a diabetes early warning system based on machine learning, the system comprising:

[0024] A collection unit, used to collect multi-dimensional data of target users;

[0025] an extraction unit, configured to extract designated features related to diabetes from the multi-dimensional data of the target user;

[0026] A prediction unit is used to input the specified features into a preset diabetes warning model for prediction to obtain a diabetes warning result, wherein the diabetes warning model is obtained by training a machine learning model based on sample data, and the diabetes warning result includes: a warning level of at least one type of diabetes.

[0027] Preferably, the prediction unit includes:

[0028] An acquisition module, configured to acquire sample data, wherein the sample data includes at least multi-dimensional data of diabetic patients and non-diabetic patients;

[0029] An extraction module, configured to extract sample-specified features related to diabetes from the sample data;

[0030] A training module is used to use the sample specified features corresponding to the sample data to train a machine learning model until the machine learning model converges to obtain a diabetes early warning model, where the machine learning model is a logistic regression model, a decision tree model, a support vector machine model or a neural network model.

[0031] Preferably, the training module is specifically used for:

[0032] Dividing the sample data into a training set and a test set;

[0033] Using the sample specified features corresponding to the training set, training a machine learning model until the machine learning model converges to obtain a diabetes early warning model;

[0034] The diabetes early warning model is verified using the sample-specified features corresponding to the test set.

[0035] Preferably, the prediction unit further includes:

[0036] The preprocessing module is used to preprocess the sample data, and the preprocessing includes at least data cleaning, missing value processing, and feature normalization.

[0037] Preferably, it also includes:

[0038] The suggestion unit is used to obtain corresponding suggestion plans from a preset diabetes knowledge base based on the diabetes warning results, and the suggestion plans include at least a diet plan, a prevention plan and an exercise plan; and output the suggestion plans.

[0039] Based on the above-mentioned embodiments of the present invention, a machine learning-based diabetes early warning method and system is provided. The method comprises: collecting multi-dimensional data of a target user; extracting specific features related to diabetes from the target user's multi-dimensional data; and inputting the specific features into a preset diabetes early warning model for prediction to obtain a diabetes early warning result. The diabetes early warning model is obtained by training a machine learning model based on sample data. This solution extracts specific features related to diabetes from the target user's multi-dimensional data, inputs the specified features into the diabetes early warning model for prediction to obtain a diabetes early warning result, and thus provides early warning of diabetes through the machine learning model, helping users to prevent and intervene in advance. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] 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.

[0041] Figure 1 A flowchart of a diabetes early warning method based on machine learning provided by an embodiment of the present invention;

[0042] Figure 2 A flowchart of a diabetes early warning model obtained through training provided by an embodiment of the present invention;

[0043] Figure 3 An overall flow chart of a diabetes early warning method based on machine learning provided by an embodiment of the present invention;

[0044] Figure 4 A structural block diagram of a diabetes early warning system based on machine learning provided by an embodiment of the present invention;

[0045] Figure 5 Another structural block diagram of a diabetes early warning system based on machine learning provided by an embodiment of the present invention;

[0046] Figure 6 Another structural block diagram of a diabetes early warning system based on machine learning provided by an embodiment of the present invention;

[0047] Figure 7 Another structural block diagram of a diabetes early warning system based on machine learning provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0048] 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.

[0049] In this application, the terms "comprises," "comprising," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not preclude the presence of additional identical elements in the process, method, article, or apparatus that includes the element.

[0050] Diabetes has become a major global health problem. Diabetes typically has a long incubation period and complex pathogenesis, making early detection and intervention crucial for improving patient outcomes. Currently, early warning of diabetes relies primarily on physician experience and clinical examination results. However, most diabetic patients only seek medical attention after experiencing noticeable symptoms. By then, the disease has typically progressed to the mid-to-late stage, making early warning impossible.

[0051] Research has found that with the rapid development of machine learning technology, its application in healthcare is gaining increasing attention. Machine learning algorithms can process large amounts of multidimensional data and uncover hidden patterns within them, providing new insights into early warning of diabetes. By combining the analysis of multidimensional data with advanced machine learning algorithms, accurate warnings of diabetes risk can be achieved, demonstrating broad application prospects.

[0052] Therefore, an embodiment of the present invention provides a diabetes warning method and system based on machine learning, which extracts specified features related to diabetes from the multi-dimensional data of the target user, inputs the specified features into the diabetes warning model for prediction to obtain diabetes warning results, that is, early warning of diabetes is performed through the machine learning model to help users prevent and intervene in advance.

[0053] It should be noted that before this plan collects user data, authorization from relevant users has been obtained in advance, that is, this plan collects user data on the premise of legality and compliance.

[0054] See also Figure 1 , shows a flow chart of a diabetes early warning method based on machine learning provided by an embodiment of the present invention, the diabetes early warning method comprising:

[0055] Step S101: Collect multi-dimensional data of target users.

[0056] In the specific implementation of step S101 , multi-dimensional data of the target user (the user to be predicted) is collected. The multi-dimensional data includes but is not limited to: age, gender, medical history, living habits and other data.

[0057] Among them, medical history data may include previous disease diagnosis, treatment process, medication status, etc.; lifestyle habit data includes but is not limited to eating habits, exercise frequency, smoking and drinking status, etc.

[0058] Step S102: extracting designated features related to diabetes from the multi-dimensional data of the target user.

[0059] In the specific implementation of step S102, designated features related to diabetes are extracted from the multi-dimensional data of the target user. The designated features include but are not limited to: age, gender, family history of diabetes, BMI (Body Mass Index), fasting blood sugar, postprandial blood sugar, smoking history, drinking history, etc.

[0060] Step S103: inputting the designated features into a preset diabetes early warning model for prediction to obtain a diabetes early warning result.

[0061] It should be noted that the diabetes early warning model is obtained by training the machine learning model based on sample data. Figure 2 The content shown is used to explain in detail how to obtain the diabetes early warning model.

[0062] In the specific implementation of step S103, the designated features of the target user are input into the diabetes warning model for prediction, and the diabetes warning result output by the diabetes warning model for the target user is obtained. The diabetes warning result includes: a warning level of at least one type of diabetes.

[0063] For example: The diabetes warning results include warning levels for type 1 diabetes and type 2 diabetes.

[0064] It should be noted that the warning level can be divided into three levels: high, medium and low. The warning level can also be further subdivided according to actual needs. For example, the warning level can be represented by a percentage. The division method of the warning level is not limited here.

[0065] It is worth noting that in actual applications, the diabetes warning results correspond to the risk of diabetes, that is, there is a corresponding relationship between the diabetes warning results and the risk of diabetes, and they can be converted into each other.

[0066] That is to say, in actual applications, the diabetes warning results output by the diabetes warning model can be replaced by "diabetes onset risk", that is, the diabetes warning model outputs the predicted results of diabetes onset risk.

[0067] In some specific embodiments, after obtaining the diabetes warning result, the diabetes warning result is output, for example, the diabetes warning result is sent to and displayed to the target user.

[0068] In some specific embodiments, a diabetes knowledge base (medical knowledge base) is pre-built, which includes recommendations on prevention, diet, medication, exercise, psychology, and other aspects.

[0069] After obtaining the diabetes warning results of the target user, the corresponding recommended plan is obtained from the preset diabetes knowledge base based on the diabetes warning results. The recommended plan includes at least a diet plan, a prevention plan and an exercise plan; and the recommended plan is output (i.e., targeted recommendations are output).

[0070] For example: sending the acquired suggestion plan to the target user, intervening with the target user in advance through the suggestion plan, assisting the target user to adjust his lifestyle, carry out early treatment, etc., thereby reducing the risk of diabetes.

[0071] In an embodiment of the present invention, specified features related to diabetes are extracted from the multi-dimensional data of the target user, and the specified features are input into the diabetes early warning model for prediction to obtain diabetes early warning results, that is, early warning of diabetes is performed through a machine learning model to help users prevent and intervene in advance.

[0072] For the above embodiments of the present invention Figure 1 The diabetes early warning model involved in step S103 is shown in Figure 2 , shows a flow chart of obtaining a diabetes early warning model through training according to an embodiment of the present invention, including the following steps:

[0073] Step S201: Collect sample data.

[0074] In the specific implementation of step S201 , sample data (carrying real labels) is collected, and the sample data at least includes: multi-dimensional data of diabetic patients and non-diabetic patients.

[0075] That is to say, multi-dimensional data of diabetic patients and non-diabetic patients are collected, and the multi-dimensional data include but are not limited to: age, gender, medical history, living habits and other data.

[0076] Among them, medical history data may include previous disease diagnosis, treatment process, medication status, etc.; lifestyle habit data includes but is not limited to eating habits, exercise frequency, smoking and drinking status, etc.

[0077] In some specific embodiments, after the sample data is collected, the sample data is preprocessed, and the preprocessing includes at least data cleaning, missing value processing, and feature normalization. The sample data used subsequently is the sample data after preprocessing.

[0078] Among them, data cleaning is used to remove duplicate data and abnormal data, missing value processing can be done by filling and deleting, and feature normalization is used to convert feature values ​​of different dimensions to the same dimension for subsequent modeling.

[0079] Step S202: extracting sample-specific features related to diabetes from the sample data.

[0080] In the specific implementation of step S202 , sample-specific features related to diabetes are extracted from the sample data. This process is divided into two parts: feature extraction and feature selection.

[0081] Feature extraction: Extract sample-specific features related to the risk of diabetes from the collected sample data, including but not limited to age, gender, family history of diabetes, laboratory test results, etc.

[0082] Feature selection: Based on the pathogenesis of diabetes and related research, sample-specified features related to the risk of diabetes were selected. Specifically, statistical analysis methods were used for feature selection to screen out features that have a significant impact on diabetes early warning as sample-specified features. Through the analysis of a large amount of clinical data and reference to relevant medical research, it was determined that age, gender, family history of diabetes, BMI, fasting blood sugar, postprandial blood sugar, smoking history, drinking history and other characteristics are significantly correlated with the risk of diabetes. Therefore, "age, gender, family history of diabetes, BMI, fasting blood sugar, postprandial blood sugar, smoking history, drinking history and other characteristics" were used as sample-specified features.

[0083] Step S203: Using the sample-specified features corresponding to the sample data, the machine learning model is trained until the machine learning model converges to obtain a diabetes early warning model.

[0084] It should be noted that the machine learning model is a logistic regression model, a decision tree model, a support vector machine model or a neural network model.

[0085] Specifically, considering the efficiency and easy interpretability of the logistic regression model in dealing with binary classification problems, the machine learning model can give priority to the logistic regression model. Similarly, the machine learning model can also adopt other types of models such as decision tree models, support vector machine models or neural network models.

[0086] In the specific implementation of step S203 , the sample data is divided into a training set and a test set.

[0087] The machine learning model is trained using the specified features of the samples corresponding to the training set until the machine learning model converges to obtain a diabetes early warning model.

[0088] The diabetes early warning model is verified using the sample-specified features corresponding to the test set.

[0089] Specifically, the sample data is divided into training set and test set according to a certain ratio (such as 9:1), and the established machine learning model is trained using the sample specified features corresponding to the training set. The model performance is optimized by adjusting the model parameters (hyperparameter optimization), thereby training a diabetes warning model.

[0090] The trained diabetes early warning model is verified using the sample-specified features corresponding to the test set, thereby evaluating the prediction accuracy, recall rate, F1 value, etc. of the diabetes early warning model to ensure that the diabetes early warning model has good predictive and generalization capabilities.

[0091] The above embodiments of the present invention Figure 2 It is about the instructions for training to obtain a diabetes early warning model.

[0092] To better understand the content of this plan, the following Figure 3 The figure shows a rectification flow chart of a diabetes early warning method based on machine learning, which explains the scheme from the overall level. Figure 3 The steps include:

[0093] Step S301: Data collection and processing.

[0094] In the specific implementation of step S301, sample data is collected, and the sample data at least includes multi-dimensional data of diabetic patients and non-diabetic patients. The sample data is pre-processed.

[0095] Step S302: Feature extraction and selection.

[0096] In the specific implementation of step S302 , sample-specified features are extracted from the preprocessed sample data.

[0097] Step S303: Model establishment and training.

[0098] In the specific implementation of step S303, a machine learning model is selected, and the machine learning model is trained using sample data to obtain a diabetes early warning model, and the diabetes early warning model is verified.

[0099] Step S304: Model prediction and suggestions.

[0100] In the specific implementation of step S304 , the designated features of the target user are input into the diabetes early warning model for prediction to obtain a diabetes early warning result.

[0101] According to the diabetes warning results, corresponding suggestions are obtained from the preset diabetes knowledge base and output to the target users.

[0102] It should be noted that Figure 3 The execution principle of each step in the embodiment of the present invention can be found in the above embodiment of the present invention. Figure 1 and Figure 2 The content will not be repeated here.

[0103] This solution conducts a comprehensive analysis of the target users' multi-dimensional data, uses machine learning algorithms to build a model, and predicts diabetes warning results in advance, helping doctors and target users to carry out early intervention and prevention, while improving the model's generalization ability and prediction accuracy.

[0104] In general, this solution has the following beneficial effects:

[0105] High accuracy: Machine learning algorithms can process large amounts of data and extract hidden patterns from them. Compared with traditional prediction methods, they can more accurately predict the warning level of diabetes.

[0106] Early intervention: By predicting the warning level of diabetes in advance, early intervention can be carried out on high-risk patients to reduce the incidence of diabetes.

[0107] Strong scalability: Through flexible feature selection and model parameter adjustment mechanisms, targeted optimization can be performed based on different diabetes types (such as type 1 diabetes, type 2 diabetes, etc.) and different data characteristics (such as data differences in different regions and different populations), thereby improving the accuracy and adaptability of predictions.

[0108] Corresponding to the diabetes early warning method based on machine learning provided in the above embodiment of the present invention, see Figure 4 The embodiment of the present invention also provides a structural block diagram of a diabetes early warning system based on machine learning. The diabetes early warning system includes: an acquisition unit 100, an extraction unit 200 and a prediction unit 300.

[0109] The collection unit 100 is used to collect multi-dimensional data of the target user.

[0110] The extraction unit 200 is configured to extract designated features related to diabetes from the multi-dimensional data of the target user.

[0111] The prediction unit 300 is used to input the specified features into a preset diabetes warning model for prediction to obtain a diabetes warning result. The diabetes warning model is obtained by training a machine learning model based on sample data. The diabetes warning result includes: a warning level of at least one type of diabetes.

[0112] In an embodiment of the present invention, specified features related to diabetes are extracted from the multi-dimensional data of the target user, and the specified features are input into the diabetes early warning model for prediction to obtain diabetes early warning results, that is, early warning of diabetes is performed through a machine learning model to help users prevent and intervene in advance.

[0113] Preferably, see Figure 5 , shows another structural block diagram of a diabetes early warning system based on machine learning provided by an embodiment of the present invention, wherein the prediction unit 300 includes:

[0114] The acquisition module 301 is used to acquire sample data, where the sample data at least includes multi-dimensional data of diabetic patients and non-diabetic patients.

[0115] The extraction module 302 is configured to extract sample-specific features related to diabetes from the sample data.

[0116] The training module 303 is used to train the machine learning model using the sample specified features corresponding to the sample data until the machine learning model converges to obtain a diabetes early warning model. The machine learning model is a logistic regression model, a decision tree model, a support vector machine model or a neural network model.

[0117] In the specific implementation, the training module 303 is specifically used to: divide the sample data into a training set and a test set; use the sample specified features corresponding to the training set to train the machine learning model until the machine learning model converges to obtain a diabetes warning model; use the sample specified features corresponding to the test set to verify the diabetes warning model.

[0118] Preferably, combined Figure 5 , see Figure 6 , shows another structural block diagram of a diabetes early warning system based on machine learning provided by an embodiment of the present invention, wherein the prediction unit 300 further includes:

[0119] The preprocessing module 304 is used to preprocess the sample data. The preprocessing includes at least data cleaning, missing value processing, and feature normalization.

[0120] Preferably, see Figure 7 , shows another structural block diagram of a diabetes early warning system based on machine learning provided by an embodiment of the present invention, the diabetes early warning system also includes:

[0121] The suggestion unit 400 is used to obtain corresponding suggestion plans from a preset diabetes knowledge base based on the diabetes warning results. The suggestion plans include at least a diet plan, a prevention plan and an exercise plan; and output the suggestion plans.

[0122] In summary, the embodiments of the present invention provide a diabetes warning method and system based on machine learning, which extracts specified features related to diabetes from the multi-dimensional data of the target user, inputs the specified features into the diabetes warning model for prediction to obtain diabetes warning results, that is, early warning of diabetes is performed through the machine learning model to help users prevent and intervene in advance.

[0123] Each embodiment in this specification is described in a progressive manner. The same or similar parts between the embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments. In particular, for system or system embodiments, since they are basically similar to method embodiments, the description is relatively simple. For relevant parts, refer to the partial description of the method embodiment. The system and system embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. A person of ordinary skill in the art can understand and implement it without expending creative work.

[0124] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the above description has generally described the components and steps of each example according to their functions. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present invention.

[0125] 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 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 early warning method based on machine learning, characterized in that: The method comprises: Collect multi-dimensional data of target users; extracting designated features related to diabetes from the multi-dimensional data of the target user; The designated features are input into a preset diabetes early warning model for prediction to obtain a diabetes early warning result. The diabetes early warning model is obtained by training a machine learning model based on sample data. The diabetes early warning result includes: a warning level for at least one type of diabetes.

2. The method according to claim 1, characterized in that The process of training a machine learning model based on sample data to obtain a diabetes early warning model includes: Collecting sample data, the sample data at least including: multi-dimensional data of diabetic patients and non-diabetic patients; extracting sample-specified features related to diabetes from the sample data; Using the sample specified features corresponding to the sample data, a machine learning model is trained until the machine learning model converges to obtain a diabetes early warning model, where the machine learning model is a logistic regression model, a decision tree model, a support vector machine model, or a neural network model.

3. The method according to claim 2, characterized in that Using the sample-specified features corresponding to the sample data, training a machine learning model until the machine learning model converges to obtain a diabetes early warning model, including: Dividing the sample data into a training set and a test set; Using the sample specified features corresponding to the training set, training a machine learning model until the machine learning model converges to obtain a diabetes early warning model; The diabetes early warning model is verified using the sample-specified features corresponding to the test set.

4. The method according to claim 2, characterized in that Before extracting sample-specified features related to diabetes from the sample data, the method further includes: The sample data is preprocessed, and the preprocessing includes at least data cleaning, missing value processing, and feature normalization.

5. The method according to any one of claims 1 to 4, characterized in that: After getting the diabetes warning results, it also includes: Obtaining corresponding recommended plans from a preset diabetes knowledge base based on the diabetes early warning result, wherein the recommended plans include at least a diet plan, a prevention plan, and an exercise plan; Output the proposed solution.

6. A diabetes early warning system based on machine learning, characterized in that: The system comprises: A collection unit, used to collect multi-dimensional data of target users; an extraction unit, configured to extract designated features related to diabetes from the multi-dimensional data of the target user; A prediction unit is used to input the specified features into a preset diabetes warning model for prediction to obtain a diabetes warning result, wherein the diabetes warning model is obtained by training a machine learning model based on sample data, and the diabetes warning result includes: a warning level of at least one type of diabetes.

7. The system according to claim 6, characterized in that The prediction unit includes: An acquisition module, configured to acquire sample data, wherein the sample data includes at least multi-dimensional data of diabetic patients and non-diabetic patients; An extraction module, configured to extract sample-specified features related to diabetes from the sample data; A training module is used to use the sample specified features corresponding to the sample data to train a machine learning model until the machine learning model converges to obtain a diabetes early warning model, where the machine learning model is a logistic regression model, a decision tree model, a support vector machine model or a neural network model.

8. The system according to claim 7, characterized in that The training module is specifically used for: Dividing the sample data into a training set and a test set; Using the sample specified features corresponding to the training set, training a machine learning model until the machine learning model converges to obtain a diabetes early warning model; The diabetes early warning model is verified using the sample-specified features corresponding to the test set.

9. The system according to claim 7, wherein: The prediction unit further includes: The preprocessing module is used to preprocess the sample data, and the preprocessing includes at least data cleaning, missing value processing, and feature normalization.

10. The method according to any one of claims 6 to 9, characterized in that: Also includes: The suggestion unit is used to obtain corresponding suggestion plans from a preset diabetes knowledge base based on the diabetes warning results, and the suggestion plans include at least a diet plan, a prevention plan and an exercise plan; and output the suggestion plans.

Images