Disease prediction and prevention system and method based on large medical model and medium

Abstract

According to the disease prediction and prevention system and method based on the large medical model and the medium, the multi-dimensional health data of the patient is obtained and preprocessed to obtain the target multi-dimensional health data, processing of the multi-dimensional health data is achieved, a basis is provided for model construction, and the disease prediction and prevention method based on the large medical model and the medium are obtained based on the health evaluation indexes. The method comprises the following steps: establishing a large medical model in combination with deep learning, performing large medical model optimization based on historical patient data to obtain a target large medical model, realizing establishment and optimization of the model, ensuring the accuracy and the generalization ability of the model, performing disease prediction on target multi-dimensional health data based on the target large medical model to obtain a disease prediction result, and performing disease prediction on the target multi-dimensional health data based on the target large medical model. And finally, the disease prediction result is matched with the disease prevention information, and a disease prevention strategy is generated according to the matching result, so that disease prevention based on the disease prediction result is realized, and comprehensive medical information is provided for a patient.

Description

Technical Field

[0001] The present invention relates to the field of medical technology, and in particular to a disease prediction and prevention system, method and medium based on a medical big model. Background Art

[0002] Medical big models refer to mathematical models with complexity and high accuracy obtained through deep learning training of medical data. These models combine deep learning and big data technologies, can handle complex data relationships, and provide more accurate predictions and diagnostic results. The basic principle of medical big models is based on deep learning neural networks, with a large number of parameters and hierarchical structures; medical big models can provide medical professionals with new tools for disease prediction and risk assessment. These tools are designed to improve the accuracy of disease diagnosis, promote personalized health management plans, and achieve precision in disease prevention strategies. With the explosive growth of medical data and the continuous advancement of computing technology, it has become possible to develop AI models that can efficiently process this data and provide accurate predictions.

[0003] However, although existing technologies have made certain progress, they still face several key challenges and technical difficulties. First, how to effectively integrate and process multi-dimensional health data from different sources to ensure the quality and consistency of the data is a prerequisite for building an efficient prediction model. Second, existing models need to be strengthened in improving the accuracy of disease prediction and the generalization ability of models, especially when facing complex and rare diseases. When a disease is predicted, there is a lack of operations to prevent the disease. Summary of the invention

[0004] The present invention provides a disease prediction and prevention system, method and medium based on a medical big model to solve the problems raised in the background technology.

[0005] A disease prediction and prevention system based on a medical big model, comprising:

[0006] A data acquisition module, used to acquire multi-dimensional health data of a patient, and pre-process the multi-dimensional health data to obtain target multi-dimensional health data;

[0007] The model building module is used to build a medical big model based on health evaluation indicators and combined with deep learning, and optimize the medical big model based on historical patient data to obtain the target medical big model;

[0008] The disease prediction module is used to predict diseases based on the target multi-dimensional health data based on the target medical big model to obtain disease prediction results;

[0009] The disease prevention module is used to match the disease prediction results with the disease prevention information and generate disease prevention strategies based on the matching results.

[0010] Preferably, the data acquisition module includes:

[0011] A scanning unit, used to scan various examination sheets of the patient and the doctor's diagnosis results of the patient to obtain health information;

[0012] A matching unit, used to extract health attributes and health values ​​from health information, match the health attributes with the multi-dimensional health template, and configure the health value to a position corresponding to the corresponding multi-dimensional health template according to the matching result to obtain multi-dimensional health data;

[0013] The processing unit is used to pre-process the multi-dimensional health data based on the model data requirements to obtain target multi-dimensional health data.

[0014] Preferably, the processing unit comprises:

[0015] A feature acquisition unit, used to acquire standard data features from model data requirements and determine initial data features for dimensional health data;

[0016] A method determination unit, used for obtaining a target data preprocessing method from the data preprocessing method table to convert the initial data features into the standard data features;

[0017] The data processing unit is used to preprocess the multi-dimensional health data according to the target data preprocessing method to obtain target multi-dimensional health data.

[0018] Preferably, the model building module includes:

[0019] An indicator analysis unit, used to obtain an evaluation indicator set corresponding to each disease type from the health evaluation indicators, and determine the disease indicator range and health indicator range of each evaluation indicator set;

[0020] A model building unit is used to select a machine learning algorithm based on the disease type and the numerical relationship between the disease indicator range and the health indicator range of the corresponding evaluation indicator set to build a large medical model;

[0021] A data labeling unit, configured to determine a first labeling rule based on the disease type, establish a second labeling rule based on a disease indicator range and a health indicator range of an evaluation indicator set corresponding to the disease type, and label the historical patient data based on the first labeling rule and the second labeling rule to obtain labeling data containing the first labeling and the second labeling;

[0022] A data processing unit, configured to extract the annotation features in the annotation data to obtain key data, and convert the key data into a feature form based on the model data requirements of the medical large model to obtain target key data;

[0023] A data division unit, used for randomly dividing the target key data into a plurality of training-validation data groups according to a preset ratio;

[0024] A training unit, used for performing multiple training and verification on the target medical macromodel based on the training-verification data set to obtain an intermediate medical macromodel;

[0025] The optimization unit is used to apply the intermediate medical large model to the actual medical environment for optimization to obtain the target medical large model.

[0026] Preferably, the optimization unit comprises:

[0027] An application unit, used for obtaining application results of the intermediate medical large model during the medical test in the actual medical environment, and obtaining prediction information and feedback information from the application results;

[0028] A result analysis unit is used to determine a prediction evaluation result of the prediction information based on the feedback information, and optimize the intermediate medical big model based on the prediction evaluation result to obtain a target medical big model.

[0029] Preferably, the disease prediction module includes:

[0030] A prediction unit, used for inputting the target multi-dimensional health data into the target medical macro model to obtain abnormal health data, health data trends and disease prediction results;

[0031] The display unit is used to display the abnormal health data, health data trends and disease prediction results in charts and texts.

[0032] Preferably, the disease prevention module includes:

[0033] A curve analysis unit, used for obtaining abnormal health index values ​​from the disease prediction results, and generating an abnormal health index value curve based on all abnormal health index values, and comparing the abnormal health index value curve with a standard abnormal health index value curve to obtain a difference curve;

[0034] A threshold judgment unit, used to compare the difference curve with a curve threshold, and judge whether there is a curve point that is not within the range of the curve threshold;

[0035] If so, based on the indicator type of the abnormal value of the health indicator, multiple disease prevention measures corresponding to the indicator type are matched from the disease prevention information, and the multiple disease prevention measures are used as target disease prevention measures;

[0036] Otherwise, based on the numerical difference between the value of the curve point and the curve threshold, the multiple disease prevention measures are adjusted to obtain multiple adjusted disease prevention measures, and the multiple adjusted disease prevention measures are used as target disease prevention measures;

[0037] The strategy determination unit is used to simplify multiple target disease prevention measures based on the prevention effects between the multiple target disease prevention measures to obtain multiple optimized disease prevention measures, and to determine the execution order and execution interval of the multiple optimized disease prevention measures based on the time association and operation association of the multiple optimized disease prevention measures to obtain the disease prevention strategy.

[0038] Preferably, the threshold judgment unit adjusts the multiple disease prevention measures based on the numerical difference between the value of the curve point and the curve threshold, and obtains the multiple adjusted disease prevention measures as follows:

[0039] Based on the fluctuating characteristics of each disease prevention measure, determining the adjustment weight for each disease prevention measure, and determining the adjustment value for each disease prevention measure based on the numerical difference;

[0040] Based on the product of the adjustment weight and the adjustment value, a target adjustment value for the disease prevention measure is determined, and the disease prevention measure is adjusted according to the target adjustment value to obtain an adjusted disease prevention measure.

[0041] A prevention method of a disease prediction and prevention system based on a medical big model, comprising:

[0042] S1: Acquire multi-dimensional health data of a patient, and pre-process the multi-dimensional health data to obtain target multi-dimensional health data;

[0043] S2: Based on health evaluation indicators and deep learning, a medical big model is established, and the medical big model is optimized based on historical patient data to obtain the target medical big model;

[0044] S3: Based on the target medical big model, disease prediction is performed on the target multi-dimensional health data to obtain disease prediction results;

[0045] S4: Match the disease prediction results with the disease prevention information and generate a disease prevention strategy based on the matching results.

[0046] A storage medium stores a computer program, which, when executed by a processor, implements the steps of a disease prediction and prevention method based on a medical big model.

[0047] Compared with the prior art, the present invention has achieved the following beneficial effects:

[0048] By acquiring the patient's multi-dimensional health data and pre-processing the multi-dimensional health data to obtain the target multi-dimensional health data, the processing of the multi-dimensional health data is realized, the quality and consistency of the data are guaranteed, and a basis is provided for model construction. Based on health evaluation indicators, a medical big model is established in combination with deep learning, and the medical big model is optimized based on historical patient data to obtain the target medical big model, and the establishment and optimization of the model is realized to ensure the accuracy of the model and the generalization ability of the model. Based on the target medical big model, the target multi-dimensional health data is predicted for disease to obtain the disease prediction results to ensure the accuracy of the model-based disease prediction. Finally, the disease prediction results are matched with the disease prevention information, and the disease prevention strategy is generated according to the matching results to realize disease prevention based on the disease prediction results, and provide patients with comprehensive medical information.

[0049] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures specifically pointed out in this application document.

[0050] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0052] Figure 1 is a structural diagram of a disease prediction and prevention system based on a medical big model in an embodiment of the present invention;

[0053] Figure 2 is a structural diagram of the data acquisition module described in an embodiment of the present invention;

[0054] Figure 3 It is a structural diagram of the disease prediction and prevention method based on the medical big model in an embodiment of the present invention. DETAILED DESCRIPTION

[0055] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0056] Embodiment 1:

[0057] The embodiment of the present invention provides a disease prediction and prevention system based on a medical big model, such as Figure 1 As shown, including:

[0058] A data acquisition module, used to acquire multi-dimensional health data of a patient, and pre-process the multi-dimensional health data to obtain target multi-dimensional health data;

[0059] The model building module is used to build a medical big model based on health evaluation indicators and combined with deep learning, and optimize the medical big model based on historical patient data to obtain the target medical big model;

[0060] The disease prediction module is used to predict diseases based on the target multi-dimensional health data based on the target medical big model to obtain disease prediction results;

[0061] The disease prevention module is used to match the disease prediction results with the disease prevention information and generate disease prevention strategies based on the matching results.

[0062] In this embodiment, the multi-dimensional health data includes, for example, blood pressure, heart rate, body temperature, respiratory rate, blood oxygen saturation, EEG activity, EMG activity, intraocular pressure, medical history data, imaging data, etc.

[0063] In this embodiment, the health evaluation indicators include, for example, body weight, body temperature, organ function, etc.

[0064] In this embodiment, historical patient data is obtained from a historical database.

[0065] In this embodiment, the disease prevention information specifies the prevention measures corresponding to different diseases.

[0066] The beneficial effects of the above design scheme are: by acquiring the patient's multi-dimensional health data and pre-processing the multi-dimensional health data to obtain the target multi-dimensional health data, the processing of the multi-dimensional health data is realized, the quality and consistency of the data are guaranteed, and a basis is provided for model construction. Based on health evaluation indicators, a medical big model is established in combination with deep learning, and the medical big model is optimized based on historical patient data to obtain the target medical big model, so as to realize the establishment and optimization of the model, ensure the accuracy of the model and the generalization ability of the model, and predict the disease of the target multi-dimensional health data based on the target medical big model to obtain the disease prediction results, and ensure the accuracy of the disease prediction based on the model. Finally, the disease prediction results are matched with the disease prevention information, and the disease prevention strategy is generated according to the matching results to realize disease prevention based on the disease prediction results, and provide patients with comprehensive medical information.

[0067] Embodiment 2:

[0068] Based on Example 1, the present invention provides a disease prediction and prevention system based on a medical big model, such as Figure 2 As shown, the data acquisition module includes:

[0069] A scanning unit, used to scan various examination sheets of the patient and the doctor's diagnosis results of the patient to obtain health information;

[0070] A matching unit, used to extract health attributes and health values ​​from health information, match the health attributes with the multi-dimensional health template, and configure the health value to a position corresponding to the corresponding multi-dimensional health template according to the matching result to obtain multi-dimensional health data;

[0071] The processing unit is used to pre-process the multi-dimensional health data based on the model data requirements to obtain target multi-dimensional health data.

[0072] In this embodiment, health attributes include blood pressure, heart rate, body temperature, respiratory rate, blood oxygen saturation, EEG activity, EMG activity, intraocular pressure, medical records, imaging data, etc.

[0073] In this embodiment, the multi-dimensional health template is pre-set according to the health condition of the human body.

[0074] The beneficial effects of the above design scheme are: by scanning the patient's various examination sheets and the doctor's diagnosis results of the patient, health information is obtained, health attributes and health values ​​are extracted from the health information, and matching is performed with the multidimensional health template based on the health attributes. According to the matching results, the health value is configured to the corresponding position of the corresponding multidimensional health template to obtain multidimensional health data, and based on the model data requirements, the multidimensional health data is pre-processed to obtain the target multidimensional health data, thereby realizing the acquisition and processing of multidimensional health data, ensuring the quality and consistency of the data, and providing a basis for model construction.

[0075] Embodiment 3:

[0076] Based on Example 2, the embodiment of the present invention provides a disease prediction and prevention system based on a medical big model, wherein the processing unit includes:

[0077] A feature acquisition unit, used to acquire standard data features from model data requirements and determine initial data features for dimensional health data;

[0078] A method determination unit, used for obtaining a target data preprocessing method from the data preprocessing method table to convert the initial data features into the standard data features;

[0079] The data processing unit is used to preprocess the multi-dimensional health data according to the target data preprocessing method to obtain target multi-dimensional health data.

[0080] In this embodiment, the target data preprocessing method is a continuous processing process, for example, data cleaning is first performed on the data, and then data conversion is performed. The specific processing parameters are set according to the differences between the features.

[0081] The beneficial effects of the above design scheme are: by obtaining standard data features from the model data requirements, determining the initial data features of the dimensional health data, obtaining the target data preprocessing method from the data preprocessing method table from the initial data features to the standard data features, preprocessing the dimensional health data according to the target data preprocessing method, obtaining the target multi-dimensional health data, realizing the processing of multi-dimensional health data, ensuring the quality and consistency of the data, and providing a basis for model construction.

[0082] Embodiment 4:

[0083] Based on Example 1, the present invention provides a disease prediction and prevention system based on a medical big model, wherein the model building module includes:

[0084] An indicator analysis unit, used to obtain an evaluation indicator set corresponding to each disease type from the health evaluation indicators, and determine the disease indicator range and health indicator range of each evaluation indicator set;

[0085] A model building unit is used to select a machine learning algorithm based on the disease type and the numerical relationship between the disease indicator range and the health indicator range of the corresponding evaluation indicator set to build a large medical model;

[0086] A data labeling unit, configured to determine a first labeling rule based on the disease type, establish a second labeling rule based on a disease indicator range and a health indicator range of an evaluation indicator set corresponding to the disease type, and label the historical patient data based on the first labeling rule and the second labeling rule to obtain labeling data containing the first labeling and the second labeling;

[0087] A data processing unit, configured to extract the annotation features in the annotation data to obtain key data, and convert the key data into a feature form based on the model data requirements of the medical large model to obtain target key data;

[0088] A data division unit, used for randomly dividing the target key data into a plurality of training-validation data groups according to a preset ratio;

[0089] A training unit, used for performing multiple training and verification on the target medical macromodel based on the training-verification data set to obtain an intermediate medical macromodel;

[0090] The optimization unit is used to apply the intermediate medical large model to the actual medical environment for optimization to obtain the target medical large model.

[0091] In this embodiment, the machine learning algorithm includes logistic regression, support vector machine, decision tree, etc.

[0092] In this embodiment, the training-validation data set divides the target key data into 10 groups, for example, and selects one group as the validation group in sequence, and the others as the training groups, to obtain 10 training-validation data sets.

[0093] The beneficial effects of the above design scheme are: by obtaining the evaluation index set corresponding to each disease type from the health evaluation index, and determining the disease index range and health index range of each evaluation index set, based on the disease type and the numerical relationship between the disease index range and health index range of the corresponding evaluation index set, a machine learning algorithm is selected to establish a large medical model, a first labeling rule is determined based on the disease type, a second labeling rule is established based on the disease index range and health index range of the evaluation index set corresponding to the disease type, and historical patient data is marked based on the first labeling rule and the second labeling rule to obtain labeled data containing the first label and the second labeling, a suitable machine learning algorithm is selected according to the specific problem, the model performance is guaranteed from the model itself, and the labeled data is labeled based on the first labeling rule and the second labeling rule. The labeled features are used to extract data to obtain key data, and based on the model data requirements of the medical big model, the key data are converted into feature form to obtain target key data, and the target key data are randomly divided into multiple training-validation data groups according to a preset ratio. The performance and generalization ability of the model are improved through data feature conversion, the consistency of the data is ensured, and a basis for cross-validation is provided. The target medical big model is trained and verified multiple times based on the training-validation data group to obtain an intermediate medical big model. The generalization ability of the model is ensured through methods such as cross-validation. The intermediate medical big model is applied to the actual medical environment for optimization to obtain the target medical big model, and the trained model is deployed to the actual environment for actual application and continuous monitoring, and necessary adjustments and maintenance are performed according to feedback from actual applications.

[0094] Embodiment 5:

[0095] Based on Example 4, the present invention provides a disease prediction and prevention system based on a medical big model, wherein the optimization unit includes:

[0096] An application unit, used for obtaining application results of the intermediate medical large model during the medical test in the actual medical environment, and obtaining prediction information and feedback information from the application results;

[0097] A result analysis unit is used to determine a prediction evaluation result of the prediction information based on the feedback information, and optimize the intermediate medical big model based on the prediction evaluation result to obtain a target medical big model.

[0098] The beneficial effects of the above design scheme are: by obtaining the application results of the intermediate medical big model during the medical test in the actual medical environment, obtaining prediction information and feedback information from the application results, determining the prediction evaluation results of the prediction information based on the feedback information, and optimizing the intermediate medical big model based on the prediction evaluation results to obtain the target medical big model, thereby ensuring the model performance of the target medical big model.

[0099] Embodiment 6:

[0100] Based on Example 1, the embodiment of the present invention provides a disease prediction and prevention system based on a medical big model, wherein the disease prediction module includes:

[0101] A prediction unit, used for inputting the target multi-dimensional health data into the target medical macro model to obtain abnormal health data, health data trends and disease prediction results;

[0102] The display unit is used to display the abnormal health data, health data trends and disease prediction results in charts and texts.

[0103] The beneficial effect of the above design scheme is: by inputting the target multi-dimensional health data into the target medical big model, abnormal health data, health data trends and disease prediction results are obtained, and the abnormal health data, health data trends and disease prediction results are displayed in charts and texts to ensure the accuracy of model-based disease prediction.

[0104] Embodiment 7:

[0105] Based on Example 1, the embodiment of the present invention provides a disease prediction and prevention system based on a medical big model, wherein the disease prevention module includes:

[0106] A curve analysis unit, used for obtaining abnormal health index values ​​from the disease prediction results, and generating an abnormal health index value curve based on all abnormal health index values, and comparing the abnormal health index value curve with a standard abnormal health index value curve to obtain a difference curve;

[0107] A threshold judgment unit, used to compare the difference curve with a curve threshold, and judge whether there is a curve point that is not within the range of the curve threshold;

[0108] If so, based on the indicator type of the abnormal value of the health indicator, multiple disease prevention measures corresponding to the indicator type are matched from the disease prevention information, and the multiple disease prevention measures are used as target disease prevention measures;

[0109] Otherwise, based on the numerical difference between the value of the curve point and the curve threshold, the multiple disease prevention measures are adjusted to obtain multiple adjusted disease prevention measures, and the multiple adjusted disease prevention measures are used as target disease prevention measures;

[0110] The strategy determination unit is used to simplify multiple target disease prevention measures based on the prevention effects between the multiple target disease prevention measures to obtain multiple optimized disease prevention measures, and to determine the execution order and execution interval of the multiple optimized disease prevention measures based on the time association and operation association of the multiple optimized disease prevention measures to obtain the disease prevention strategy.

[0111] In this embodiment, if multiple target disease prevention measures have the same prevention effects, the measures may be deleted.

[0112] In this embodiment, the standard abnormal index value curve is determined according to the disease prevention information.

[0113] The beneficial effects of the above design scheme are: by obtaining health indicator abnormal values ​​from disease prediction results, and generating a health indicator abnormal value curve based on all health indicator abnormal values, the health indicator abnormal value curve is compared with the standard abnormal indicator value curve to obtain a difference curve, and the difference curve is compared with the curve threshold to determine whether there is a curve point that is not within the range of the curve threshold. If so, based on the indicator type of the health indicator abnormal value, multiple disease prevention measures corresponding to the indicator type are matched from the disease prevention information, and the multiple disease prevention measures are used as target disease prevention measures. Otherwise, based on the numerical difference between the value of the curve point and the curve threshold, the multiple disease prevention measures are adjusted to obtain multiple adjusted disease prevention measures, and the multiple adjusted disease prevention measures are used as target disease prevention measures. Based on the prevention effects between the multiple target disease prevention measures, the multiple target disease prevention measures are simplified to obtain multiple optimized disease prevention measures, and based on the time association and operation association of the multiple optimized disease prevention measures, the execution order and execution interval of the multiple optimized disease prevention measures are determined to obtain a disease prevention strategy, so as to achieve disease prevention based on disease prediction results and provide patients with comprehensive medical information.

[0114] Embodiment 8:

[0115] Based on Example 7, the embodiment of the present invention provides a disease prediction and prevention system based on a medical big model. The threshold judgment unit adjusts the multiple disease prevention measures based on the numerical difference between the value of the curve point and the curve threshold, and obtains multiple adjusted disease prevention measures, which are specifically:

[0116] Based on the fluctuating characteristics of each disease prevention measure, determining the adjustment weight for each disease prevention measure, and determining the adjustment value for each disease prevention measure based on the numerical difference;

[0117] Based on the product of the adjustment weight and the adjustment value, a target adjustment value for the disease prevention measure is determined, and the disease prevention measure is adjusted according to the target adjustment value to obtain an adjusted disease prevention measure.

[0118] In this embodiment, the more fluctuating characteristics there are and the larger the range is, the greater the corresponding adjustment weight is.

[0119] The beneficial effects of the above design scheme are: by determining the adjustment weight for each disease prevention measure based on the fluctuating characteristics of each disease prevention measure, determining the adjustment value for each disease prevention measure based on the numerical difference, determining the target adjustment value for the disease prevention measure based on the product of the adjustment weight and the adjustment value, and adjusting the disease prevention measure according to the target adjustment value to obtain adjusted disease prevention measures, thereby providing accurate disease prevention measures for the generation of disease prevention strategies.

[0120] Embodiment 9:

[0121] Based on Example 1, the present invention provides a prevention method of a disease prediction and prevention system based on a medical big model, such as Figure 3 As shown, including:

[0122] S1: Acquire multi-dimensional health data of a patient, and pre-process the multi-dimensional health data to obtain target multi-dimensional health data;

[0123] S2: Based on health evaluation indicators and deep learning, a medical big model is established, and the medical big model is optimized based on historical patient data to obtain the target medical big model;

[0124] S3: Based on the target medical big model, disease prediction is performed on the target multi-dimensional health data to obtain disease prediction results;

[0125] S4: Match the disease prediction results with the disease prevention information and generate a disease prevention strategy based on the matching results.

[0126] In this embodiment, the multi-dimensional health data includes, for example, blood pressure, heart rate, body temperature, respiratory rate, blood oxygen saturation, EEG activity, EMG activity, intraocular pressure, medical history data, imaging data, etc.

[0127] In this embodiment, the health evaluation indicators include, for example, body weight, body temperature, organ function, etc.

[0128] In this embodiment, historical patient data is obtained from a historical database.

[0129] In this embodiment, the disease prevention information specifies the prevention measures corresponding to different diseases.

[0130] The beneficial effects of the above design scheme are: by acquiring the patient's multi-dimensional health data and pre-processing the multi-dimensional health data to obtain the target multi-dimensional health data, the processing of the multi-dimensional health data is realized, the quality and consistency of the data are guaranteed, and a basis is provided for model construction. Based on health evaluation indicators, a medical big model is established in combination with deep learning, and the medical big model is optimized based on historical patient data to obtain the target medical big model, so as to realize the establishment and optimization of the model, ensure the accuracy of the model and the generalization ability of the model, and predict the disease of the target multi-dimensional health data based on the target medical big model to obtain the disease prediction results, and ensure the accuracy of the disease prediction based on the model. Finally, the disease prediction results are matched with the disease prevention information, and the disease prevention strategy is generated according to the matching results to realize disease prevention based on the disease prediction results, and provide patients with comprehensive medical information.

[0131] Embodiment 10:

[0132] A storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the disease prediction and prevention method based on a medical big model as described in Example 9.

[0133] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention belong to the scope of this application document and its equivalent technology, the present invention is also intended to include these changes and variations.

Claims

1. A disease prediction and prevention system based on a medical big model, characterized in that: include: A data acquisition module, used to acquire multi-dimensional health data of a patient, and pre-process the multi-dimensional health data to obtain target multi-dimensional health data; The model building module is used to build a medical big model based on health evaluation indicators and combined with deep learning, and optimize the medical big model based on historical patient data to obtain the target medical big model; The disease prediction module is used to predict diseases based on the target multi-dimensional health data based on the target medical big model and obtain disease prediction results; The disease prevention module is used to match the disease prediction results with the disease prevention information and generate disease prevention strategies based on the matching results.

2. A disease prediction and prevention system based on a medical big model according to claim 1, characterized in that: The data acquisition module comprises: A scanning unit, used to scan various examination sheets of the patient and the doctor's diagnosis results of the patient to obtain health information; A matching unit, used to extract health attributes and health values ​​from health information, match the health attributes with the multi-dimensional health template, and configure the health value to a position corresponding to the corresponding multi-dimensional health template according to the matching result to obtain multi-dimensional health data; The processing unit is used to pre-process the multi-dimensional health data based on the model data requirements to obtain target multi-dimensional health data.

3. A disease prediction and prevention system based on a medical big model according to claim 2, characterized in that: The processing unit comprises: A feature acquisition unit, used to acquire standard data features from model data requirements and determine initial data features for dimensional health data; A method determination unit, used for obtaining a target data preprocessing method from the data preprocessing method table to convert the initial data features into the standard data features; The data processing unit is used to preprocess the multi-dimensional health data according to the target data preprocessing method to obtain target multi-dimensional health data.

4. The disease prediction and prevention system based on the medical big model according to claim 1 is characterized in that: The model building module comprises: An indicator analysis unit, used to obtain an evaluation indicator set corresponding to each disease type from the health evaluation indicators, and determine the disease indicator range and health indicator range of each evaluation indicator set; A model building unit is used to select a machine learning algorithm based on the disease type and the numerical relationship between the disease indicator range and the health indicator range of the corresponding evaluation indicator set to build a large medical model; A data labeling unit, configured to determine a first labeling rule based on the disease type, establish a second labeling rule based on a disease indicator range and a health indicator range of an evaluation indicator set corresponding to the disease type, and label the historical patient data based on the first labeling rule and the second labeling rule to obtain labeling data containing the first labeling and the second labeling; A data processing unit, configured to extract the annotation features in the annotation data to obtain key data, and convert the key data into a feature form based on the model data requirements of the medical large model to obtain target key data; A data division unit, used for randomly dividing the target key data into a plurality of training-validation data groups according to a preset ratio; A training unit, used for performing multiple training and verification on the target medical macromodel based on the training-verification data set to obtain an intermediate medical macromodel; The optimization unit is used to apply the intermediate medical large model to the actual medical environment for optimization to obtain the target medical large model.

5. A disease prediction and prevention system based on a medical big model according to claim 4, characterized in that: The optimization unit comprises: An application unit, used for obtaining application results of the intermediate medical large model during the medical test in the actual medical environment, and obtaining prediction information and feedback information from the application results; A result analysis unit is used to determine a prediction evaluation result of the prediction information based on the feedback information, and optimize the intermediate medical big model based on the prediction evaluation result to obtain a target medical big model.

6. The disease prediction and prevention system based on the medical big model according to claim 1 is characterized in that: The disease prediction module comprises: A prediction unit, used for inputting the target multi-dimensional health data into the target medical macro model to obtain abnormal health data, health data trends and disease prediction results; The display unit is used to display the abnormal health data, health data trends and disease prediction results in charts and texts.

7. The disease prediction and prevention system based on medical big model according to claim 1 is characterized in that: The disease prevention module comprises: A curve analysis unit, used for obtaining abnormal health index values ​​from the disease prediction results, and generating an abnormal health index value curve based on all abnormal health index values, and comparing the abnormal health index value curve with a standard abnormal health index value curve to obtain a difference curve; A threshold judgment unit, used to compare the difference curve with a curve threshold, and judge whether there is a curve point that is not within the range of the curve threshold; If so, based on the indicator type of the abnormal value of the health indicator, multiple disease prevention measures corresponding to the indicator type are matched from the disease prevention information, and the multiple disease prevention measures are used as target disease prevention measures; Otherwise, based on the numerical difference between the value of the curve point and the curve threshold, the multiple disease prevention measures are adjusted to obtain multiple adjusted disease prevention measures, and the multiple adjusted disease prevention measures are used as target disease prevention measures; The strategy determination unit is used to simplify multiple target disease prevention measures based on the prevention effects between the multiple target disease prevention measures to obtain multiple optimized disease prevention measures, and to determine the execution order and execution interval of the multiple optimized disease prevention measures based on the time association and operation association of the multiple optimized disease prevention measures to obtain the disease prevention strategy.

8. A disease prediction and prevention system based on a medical big model according to claim 7, characterized in that: The threshold judgment unit adjusts the multiple disease prevention measures based on the numerical difference between the value of the curve point and the curve threshold, and obtains multiple adjusted disease prevention measures, which are specifically: Based on the fluctuating characteristics of each disease prevention measure, determining the adjustment weight for each disease prevention measure, and determining the adjustment value for each disease prevention measure based on the numerical difference; Based on the product of the adjustment weight and the adjustment value, a target adjustment value for the disease prevention measure is determined, and the disease prevention measure is adjusted according to the target adjustment value to obtain an adjusted disease prevention measure.

9. A disease prediction and prevention method based on a medical big model, used in a disease prediction and prevention system as claimed in claim 1, characterized in that: include: S1: Acquire multi-dimensional health data of a patient, and pre-process the multi-dimensional health data to obtain target multi-dimensional health data; S2: Based on health evaluation indicators and deep learning, a medical big model is established, and the medical big model is optimized based on historical patient data to obtain the target medical big model; S3: Based on the target medical big model, disease prediction is performed on the target multi-dimensional health data to obtain disease prediction results; S4: Match the disease prediction results with the disease prevention information and generate a disease prevention strategy based on the matching results.

10. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the disease prediction and prevention method based on the medical big model as claimed in claim 9 are implemented.

Images