Comprehensive health evaluation method and system based on in-vivo component metabolism multiple indexes

By establishing a medical knowledge graph and integrated model, and adaptively adding indicators and updating labels, the false alarm and false negative problems of the in vivo component detection model were solved, and the model was automatically optimized and its accuracy was improved.

CN118969284AActive Publication Date: 2024-11-15北京中生金域诊断技术股份有限公司

Patent Information

Application Number
CN202411410111.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-10-10
Publication Date
2024-11-15
Estimated Expiration
2044-10-10

AI Technical Summary

Technical Problem

Existing in vivo component detection models suffer from frequent false alarms or missed alarms due to a limited number of indicators and the lack of clinical data covering rare diseases. Furthermore, model optimization requires extensive business knowledge and professional experience, resulting in a large workload and high personnel qualifications.

Method used

By establishing a medical knowledge graph and integrated model, indicators are adaptively added and tags are updated. The model is optimized using similarity algorithms to achieve dynamic updates and automatic iteration, reducing reliance on business knowledge.

Benefits of technology

It improved the accuracy of the model, reduced false positives and false negatives, simplified the model optimization process, and lowered the requirements for professional knowledge.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118969284A_ABST
    Figure CN118969284A_ABST
Patent Text Reader

Abstract

The invention discloses a comprehensive health evaluation method based on multiple indexes of in-vivo component metabolism. The method comprises the following steps: establishing an integrated model of a basic classification algorithm and a medical knowledge graph database; for the condition that in-vivo component metabolism indexes are sufficient and the disease classification labels of the medical knowledge map database can fully cover clinical diseases, predicting disease classification and health condition evaluation of the user based on an integrated model of a basic classification algorithm; otherwise, after variable self-adaptive addition of the fixed indexes and the user basic information indexes and automatic updating of the disease classification labels are carried out, the disease classification and health condition evaluation of the user are predicted; and collecting corresponding newly-added clinical user information by aiming at a fixed index adaptively added by a variable, a corresponding user basic information index and an automatically updated disease classification label, and adding non-existing entity nodes and relationships so as to form updating of the medical knowledge graph database. The invention further discloses a system, electronic equipment and a computer readable storage medium.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of body component detection and evaluation, and in particular to a comprehensive health evaluation method and system based on multiple indicators of body component metabolism. Background Art

[0002] In the prior art, the characterization data of body components is completed based on the classification task of 6 body fluid indicators and basic characteristics. However, due to the small number of indicators and the fact that there are rare diseases that are not covered by clinical data, the existing classification models may have false positives or false negatives. In addition, the existing classification models are based on integrated algorithms or SVM and other technologies. These classification models are all for fixed features and specific labels. In actual work, the addition of features and label types requires a lot of business knowledge and relevant experience accumulation to determine more suitable features and label types, and then collect data according to the needs of professionals to gradually iterate and optimize the model so that the model is constantly improved. It consumes a lot of work and requires extremely high professional quality of personnel.

[0003] Therefore, there is a need to develop new evaluation methods and systems for multi-index detection of in vivo component metabolism. Summary of the invention

[0004] The purpose of the present invention is to provide a comprehensive health evaluation method and system based on multiple indicators of body component metabolism, which is used for comprehensive evaluation of diseases based on multiple indicators of body fluids tested by reagents and basic information characteristics of users. In view of the situation that model personnel are unable to obtain effective indicators in a short period of time due to lack of business knowledge or experience, it is possible to adaptively add indicators in the prediction of fixed indicator models, and automatically update the types of iterative labels, so as to achieve the technical effect of continuous optimization and continuous improvement of the accuracy of the model; at the same time, it solves the problem that the model personnel are misreporting or missing diseases due to lack of business knowledge or experience. Specifically, through the similarity and threshold setting operations of the medical knowledge graph library, relevant indicators can be obtained in a relatively short time and label optimization and completion operations can be performed. At the same time, the clinical data collected by the indicators can continuously optimize the medical knowledge graph library, forming a dynamic update comprehensive evaluation algorithm and database automatic update function.

[0005] The first aspect of the present invention is to provide a comprehensive health evaluation method based on multiple indicators of body component metabolism, comprising: S1, establishing an integrated model of a basic classification algorithm and a medical knowledge graph library; the medical knowledge graph library includes multiple disease classification labels and corresponding network structures, and the network structure includes indicator nodes and disease nodes; S2, if the in vivo component metabolism index is sufficient and the disease classification labels of the medical knowledge atlas library can fully cover clinical diseases, predict the user's disease classification and health status evaluation based on the integrated model of the basic classification algorithm; otherwise, execute S3; S3, in the case where the body component metabolism index is insufficient and / or the disease classification labels of the medical knowledge atlas library cannot fully cover clinical diseases, after the variables of the fixed index and the user basic information index are adaptively added and the disease classification labels are automatically updated, the disease classification and health status evaluation of the user are predicted based on all the fixed indexes and the user basic information indexes after the variables are adaptively added and the automatically updated disease classification labels; S4, for the fixed indicators adaptively added to the variables and the corresponding user basic information indicators and the automatically updated disease classification labels, collect the corresponding newly added clinical user information, and after analyzing the corresponding newly added clinical user information, determine the entity nodes and relationships that do not exist in the current medical knowledge graph library, and add the non-existent entity nodes and relationships to form an update to the medical knowledge graph library.

[0006] Preferably, the entity items of the medical knowledge graph library include: types of diseases and complications, clinical symptoms, metabolic indicators of body components, microbial communities, body parts, populations, treatment methods and regions; wherein, metabolic indicators of body components belong to fixed indicators, and microbial communities, body parts, populations, treatment methods and regions belong to user basic information indicators; Preferably, the relational items of the medical knowledge graph library include: symptoms corresponding to diseases and complications, examination items corresponding to metabolic indicators of body components, and the microbial communities involved.

[0007] Preferably, the process of establishing the integrated model of the classification algorithm based on S1 includes: S11, determining a training data set, wherein the training data set includes a fixed number of training features and a fixed number of labels; S12, inputting the data in the training data set into a classification integration algorithm for training, thereby establishing an initial integration model of the classification algorithm; S13, verifying the accuracy and recall of the initial integrated model of the classification algorithm. When the accuracy and recall meet the predetermined requirements, an integrated model of the classification algorithm based on the obtained integrated model is obtained; wherein the recall rate is related to the false alarm rate and the missed alarm rate.

[0008] Preferably, the indicator nodes and disease nodes of the medical knowledge graph library are adaptively added to the training features and labels of the training data set.

[0009] Preferably, S2 includes: S21, obtaining fixed indicators and user basic information indicators; wherein the fixed indicators include multiple indicators of body fluid detection; S22, predicting the user's disease classification and health status evaluation based on the fixed indicators and the user's basic information indicators.

[0010] Preferably, S3 includes: S31, obtain initial disease classification based on the integrated model of basic classification algorithms and current body component metabolic indicators; S32, respectively calculating the similarity between a plurality of disease classification labels in the medical knowledge graph library and the initial disease classification, and determining all disease classification labels whose similarity meets a first preset condition and the corresponding network structure; S33, parallelly calculating the first-degree and second-degree relationships of the network structures corresponding to all disease classification labels whose similarities satisfy the first preset condition; S34, selecting indicators and disease categories with a correlation degree higher than a second threshold from the first-degree and second-degree relationships to adaptively add variables of the fixed indicators and user basic information indicators and automatically update the disease classification labels; S35, predicting the user's disease classification and health status evaluation based on all fixed indicators and user basic information indicators after the variables are adaptively added and the automatically updated disease classification labels.

[0011] The second aspect of the present invention is to provide a comprehensive health evaluation system based on multiple indicators of body composition metabolism, which is used to implement the method of the first aspect, comprising: A model and knowledge graph library establishment module (101) is used to establish an integrated model of a basic classification algorithm and a medical knowledge graph library; the medical knowledge graph library includes a plurality of disease classification labels and corresponding network structures, and the network structure includes indicator nodes and disease nodes; The first prediction and evaluation module (102) is used to predict the disease classification and health status evaluation of the user based on the integrated model of the basic classification algorithm when the body component metabolism index is sufficient and the disease classification labels of the medical knowledge atlas library can fully cover clinical diseases; The second prediction and evaluation module (103) is used to predict the disease classification and health status evaluation of the user based on all the fixed indicators and user basic information indicators after the variables are adaptively added and the disease classification labels after the variables are automatically updated, in case the body component metabolism indicators are insufficient and / or the disease classification labels of the medical knowledge atlas library cannot fully cover the clinical diseases; The knowledge graph update module (104) collects the corresponding newly added clinical user information for the fixed indicators adaptively added to the variables and the corresponding user basic information indicators and the automatically updated disease classification labels, analyzes the corresponding newly added clinical user information, determines the entity nodes and relationships that do not exist in the current medical knowledge graph library, and adds the non-existent entity nodes and relationships to form an update to the medical knowledge graph library.

[0012] A third aspect of the present invention is to provide an electronic device, comprising a processor and a memory, wherein the memory stores a plurality of instructions, and the processor is used to read the instructions and execute the method described in the first aspect.

[0013] A fourth aspect of the present invention is to provide a computer-readable storage medium, wherein the computer-readable storage medium stores a plurality of instructions, and the plurality of instructions can be read by a processor to execute the method described in the first aspect.

[0014] Beneficial effects of the method and system of the present invention: The false positives and negatives in the prediction results of the integrated model are analyzed, and structures that are extremely similar to the feature and label network structures are found in the medical knowledge graph library. The disease and indicator nodes in this network result are added to the model features, which reduces the time for model personnel to find important features to optimize the model when they lack business knowledge and experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the related technologies, the drawings required for use in the specific embodiments or the related technical descriptions will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0016] Figure 1 A flow chart of a comprehensive health evaluation method based on multiple indicators of body component metabolism provided according to an embodiment of the present invention; Figure 2 A schematic diagram of a comprehensive health evaluation method based on multiple indicators of body component metabolism provided according to an embodiment of the present invention; Figure 3 A flowchart for establishing an integrated model of a basic classification algorithm provided according to an embodiment of the present invention; Figure 4 A flow chart of a method for predicting a user's disease classification and health status evaluation based on an integrated model of the basic classification algorithm in accordance with an embodiment of the present invention in which the metabolic indicators of body components are sufficient and the disease classification labels of the medical knowledge graph library can fully cover clinical diseases; Figure 5 A flowchart of predicting the user's disease classification and health status evaluation based on all the fixed indicators and user basic information indicators after the variables are adaptively added and the disease classification labels after the variables are automatically updated, in case that the body component metabolism indicators are insufficient and / or the disease classification labels of the medical knowledge graph library cannot fully cover the clinical diseases, according to an embodiment of the present invention; Figure 6 This is a diagram showing the architecture of a comprehensive health evaluation system based on multiple indicators of body composition metabolism according to an embodiment of the present invention; Figure 7 A schematic diagram of an electronic device provided according to an embodiment of the present invention. DETAILED DESCRIPTION

[0017] The technical solution of the present invention will be described clearly and completely below 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 creative work are within the scope of protection of the present invention.

[0018] In the description of the present invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc., indicating the orientation or positional relationship, are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", and "third" are used for descriptive purposes only, and cannot be understood as indicating or implying relative importance.

[0019] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances. Embodiment 1

[0020] like Figure 1-2 As shown, this embodiment provides a comprehensive health evaluation method based on multiple indicators of body component metabolism, including: S1, establishing an integrated model of a basic classification algorithm and a medical knowledge graph library; the medical knowledge graph library includes multiple disease classification labels and corresponding network structures, and the network structure includes indicator nodes and disease nodes; As a preferred embodiment, the entity items of the medical knowledge graph library include: types of diseases and complications, clinical symptoms, metabolic indicators of body components, microbial communities, body parts, populations, treatment methods and regions; among them, metabolic indicators of body components belong to fixed indicators, and microbial communities, body parts, populations, treatment methods and regions belong to user basic information indicators; As a preferred embodiment, the relational items of the medical knowledge graph library include: symptoms corresponding to diseases and complications, examination items corresponding to metabolic indicators of body components, and the microbial communities involved.

[0021] like Figure 3 As shown, as a preferred implementation, the process of establishing the integrated model of the classification algorithm based on S1 includes: S11, determining a training data set, wherein the training data set includes a fixed number of training features and a fixed number of labels; In this embodiment, the training features are set to 6 fixed indicators (x1, x2, x3, x4, x5 and x6 respectively); and 6 user basic information indicators (baseinfo1, baseinfo2, baseinfo3, baseinfo4, baseinfo5 and baseinfo6 respectively); and the labels are set to 8, namely disease1, disease2, disease3, disease4, disease5, disease6, disease7 and disease8.

[0022] As a preferred implementation, the indicator nodes and disease nodes of the medical knowledge graph library are adaptively added to the training features and labels of the training data set.

[0023] S12, inputting the data in the training data set into the classification integration algorithm for training, so as to establish an initial integration model of the classification algorithm; the classification integration algorithm adopted is, for example, xgboost. Of course, those skilled in the art should know that other appropriate classification integration algorithms can also be adopted, all of which are within the protection scope of the present invention.

[0024] S13, verifying the accuracy and recall of the initial integrated model of the classification algorithm. When the accuracy and recall meet the predetermined requirements, an integrated model of the classification algorithm based on the obtained integrated model is obtained; wherein the recall rate is related to the false alarm rate and the missed alarm rate.

[0025] S2, if the in vivo component metabolism index is sufficient and the disease classification labels of the medical knowledge atlas library can fully cover clinical diseases, predict the user's disease classification and health status evaluation based on the integrated model of the basic classification algorithm; otherwise, execute S3; like Figure 4 As shown, as a preferred embodiment, the S2 includes: S21, obtaining fixed indicators and user basic information indicators; wherein the fixed indicators include multiple indicators of body fluid detection; in this embodiment, there are 6 indicators, for example, six urine detection indicators in a medical weight loss project. Of course, those skilled in the art can also add or replace other detection indicators, which are all within the scope of protection of the present invention; S22, predicting the user's disease classification and health status evaluation based on the fixed indicators and the user's basic information indicators.

[0026] S3, in the case where the body component metabolism index is insufficient and / or the disease classification labels of the medical knowledge atlas library cannot fully cover clinical diseases, after the variables of the fixed index and the user basic information index are adaptively added and the disease classification labels are automatically updated, the disease classification and health status evaluation of the user are predicted based on all the fixed indexes and the user basic information indexes after the variables are adaptively added and the automatically updated disease classification labels; like Figure 5 As shown, as a preferred implementation, S3 includes: S31, obtain initial disease classification based on the integrated model of basic classification algorithms and current body component metabolic indicators; S32, respectively calculating the similarity between a plurality of disease classification labels in the medical knowledge graph library and the initial disease classification, and determining all disease classification labels whose similarity meets a first preset condition and the corresponding network structure; As a preferred implementation, the first preset condition is that the similarity is greater than a first threshold; in this embodiment, the first threshold is set to 0.7. Of course, those skilled in the art will set different thresholds according to the number of indicators added; for a large number of added indicators, the first threshold can be set less than 0.7, otherwise it can be greater than 0.7.

[0027] S33, parallelly calculating the first-degree and second-degree relationships of the network structure corresponding to all disease classification labels whose similarities meet the first preset condition.

[0028] The second-degree relationship refers to the relationship between the branches of the network structure discovered between one end of the network structure and the other end of the network structure through the intermediate structure as a bridge.

[0029] GraphX ​​uses Spark to represent graphs as RDD, a distributed data set that can be loaded into memory. Compared with MapReduce, which processes data sequentially, most Spark operations are completed in memory, given the natural random access characteristics of memory, so it is more suitable for processing graph problems; GraphX ​​also processes end-to-end graph iteration problems faster than Giraph in terms of running time, so GraphX ​​is used to mine and recommend first-degree and second-degree relationships.

[0030] S34, selecting indicators and disease categories with a correlation degree higher than the second threshold from the first-degree and second-degree relationships to adaptively add variables of the fixed indicators and user basic information indicators and automatically update the disease classification labels. This achieves automatic addition and optimization of features and labels, so that the false positive and false negative rates of the integrated model of the classification algorithm are continuously reduced, and the model is continuously optimized.

[0031] S35, predicting the user's disease classification and health status evaluation based on all fixed indicators and user basic information indicators after the variables are adaptively added and the automatically updated disease classification labels.

[0032] S4, for the fixed indicators added adaptively by the variables and the corresponding basic user information indicators and the automatically updated disease classification labels, the corresponding newly added clinical user information is collected, and the corresponding newly added clinical user information is analyzed to determine the entity nodes and relationships that do not exist in the current medical knowledge graph library, and the non-existent entity nodes and relationships are added to form an update to the medical knowledge graph library. This step enables the integrated model of the classification algorithm and the medical knowledge graph library to interact with each other to form a closed loop, so that the corresponding models and knowledge graphs of the two are continuously improved, further reducing the false positive and false negative rates. Embodiment 2

[0033] like Figure 6 As shown, this embodiment provides a comprehensive health evaluation system based on multiple indicators of body component metabolism, which is used to implement the method of embodiment 1, including: The model and knowledge graph library establishment module 101 is used to establish an integrated model of a basic classification algorithm and a medical knowledge graph library; the medical knowledge graph library includes a plurality of disease classification labels and corresponding network structures, and the network structure includes indicator nodes and disease nodes; The first prediction and evaluation module 102 is used to predict the disease classification and health status evaluation of the user based on the integrated model of the basic classification algorithm when the body component metabolism index is sufficient and the disease classification labels of the medical knowledge atlas library can fully cover clinical diseases; The second prediction and evaluation module 103 is used for predicting the disease classification and health status evaluation of the user based on all the fixed indicators and user basic information indicators after the variables are adaptively added and the disease classification labels after the variables are automatically updated, when the body component metabolism indicators are insufficient and / or the disease classification labels of the medical knowledge atlas library cannot fully cover the clinical diseases; The knowledge graph update module 104 collects the corresponding newly added clinical user information for the fixed indicators adaptively added to the variables and the corresponding user basic information indicators and the automatically updated disease classification labels, and analyzes the corresponding newly added clinical user information to determine the entity nodes and relationships that do not exist in the current medical knowledge graph library, and adds the non-existent entity nodes and relationships to form an update to the medical knowledge graph library.

[0034] Application examples: This embodiment can be used for multi-indicator diagnosis, and a comprehensive evaluation can be given in combination with the basic information of the user. For example, the six metabolic indicators (L-lactic acid, creatine, creatinine, urea, β-hydroxybutyric acid, L-glutamic acid) in the risk assessment of sarcopenia can be detected. By combining these six indicators and the user's age, gender, height, weight, liver and kidney function, skeletal muscle mass, strength and other basic information, the XGBOOST model can be used to determine the user's disease type.

[0035] For user features that are incorrectly predicted by the model, the network similarity algorithm is used to find the network structure with the highest similarity from the JYR_KG graph, and then the disease nodes and indicator nodes in the network structure are filled into the XGBOOST model. The XGBOOST model is optimized by collecting data.

[0036] The present invention also provides a memory storing a plurality of instructions, wherein the instructions are used to implement the method as in the first embodiment.

[0037] like Figure 7 As shown, the present invention also provides an electronic device, including a processor 301 and a memory 302 connected to the processor 301, the memory 302 stores a plurality of instructions, and the instructions can be loaded and executed by the processor so that the processor can execute the method as in the first embodiment.

[0038] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above embodiments can be implemented by software, or by means of software plus necessary general hardware platforms. Based on such understanding, the technical solutions of the above embodiments can be embodied in the form of software products, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.), including several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0039] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A comprehensive health evaluation method based on multiple indicators of body component metabolism, characterized in that: include: S1, establishing an integrated model of a basic classification algorithm and a medical knowledge graph library; the medical knowledge graph library includes multiple disease classification labels and corresponding network structures, and the network structure includes indicator nodes and disease nodes; S2, if the in vivo component metabolism index is sufficient and the disease classification labels of the medical knowledge atlas library can fully cover clinical diseases, predict the user's disease classification and health status evaluation based on the integrated model of the basic classification algorithm; otherwise, execute S3; S3, in the case where the body component metabolism index is insufficient and / or the disease classification labels of the medical knowledge atlas library cannot fully cover clinical diseases, after the variables of the fixed index and the user basic information index are adaptively added and the disease classification labels are automatically updated, the disease classification and health status evaluation of the user are predicted based on all the fixed indexes and the user basic information indexes after the variables are adaptively added and the automatically updated disease classification labels; S4, for the fixed indicators adaptively added to the variables and the corresponding user basic information indicators and the automatically updated disease classification labels, collect the corresponding newly added clinical user information, and after analyzing the corresponding newly added clinical user information, determine the entity nodes and relationships that do not exist in the current medical knowledge graph library, and add the non-existent entity nodes and relationships to form an update to the medical knowledge graph library.

2. A comprehensive health evaluation method based on multiple indicators of body composition metabolism according to claim 1, characterized in that: The entity items of the medical knowledge graph library include: types of diseases and complications, clinical symptoms, metabolic indicators of body components, microbiome, body parts, population, treatment methods and regions; among them, metabolic indicators of body components are fixed indicators, and microbiome, body parts, population, treatment methods and regions are user basic information indicators.

3. A comprehensive health evaluation method based on multiple indicators of body composition metabolism according to claim 2, characterized in that: The relational items of the medical knowledge graph library include: symptoms corresponding to diseases and complications, examination items corresponding to metabolic indicators of body components, and the microbial communities involved.

4. A comprehensive health evaluation method based on multiple indicators of body composition metabolism according to claim 3, characterized in that: The process of establishing the integrated model of the classification algorithm based on S1 includes: S11, determining a training data set, wherein the training data set includes a fixed number of training features and a fixed number of labels; S12, inputting the data in the training data set into a classification integration algorithm for training, thereby establishing an initial integration model of the classification algorithm; S13, verifying the accuracy and recall of the initial integrated model of the classification algorithm. When the accuracy and recall meet the predetermined requirements, an integrated model of the classification algorithm based on the obtained integrated model is obtained; wherein the recall rate is related to the false alarm rate and the missed alarm rate.

5. A comprehensive health evaluation method based on multiple indicators of body composition metabolism according to claim 4, characterized in that: The indicator nodes and disease nodes of the medical knowledge graph library are adaptively added to the training features and labels of the training data set.

6. A comprehensive health evaluation method based on multiple indicators of body composition metabolism according to claim 5, characterized in that: The S2 includes: S21, obtaining fixed indicators and user basic information indicators; wherein the fixed indicators include multiple indicators of body fluid detection; S22, predicting the user's disease classification and health status evaluation based on the fixed indicators and the user's basic information indicators.

7. A comprehensive health evaluation method based on multiple indicators of body composition metabolism according to claim 6, characterized in that: The S3 includes: S31, obtain initial disease classification based on the integrated model of basic classification algorithms and current body component metabolic indicators; S32, respectively calculating the similarity between a plurality of disease classification labels in the medical knowledge graph library and the initial disease classification, and determining all disease classification labels whose similarity meets a first preset condition and the corresponding network structure; S33, parallelly calculating the first-degree and second-degree relationships of the network structures corresponding to all disease classification labels whose similarities satisfy the first preset condition; S34, selecting indicators and disease categories with a correlation degree higher than a second threshold from the first-degree and second-degree relationships to adaptively add variables of the fixed indicators and user basic information indicators and automatically update the disease classification labels; S35, predicting the user's disease classification and health status evaluation based on all fixed indicators and user basic information indicators after the variables are adaptively added and the automatically updated disease classification labels.

8. A comprehensive health evaluation system based on multiple indicators of body composition metabolism, used to implement any of the methods described in claims 1-7, characterized in that: include: A model and knowledge graph library establishment module (101) is used to establish an integrated model of a basic classification algorithm and a medical knowledge graph library; the medical knowledge graph library includes a plurality of disease classification labels and corresponding network structures, and the network structure includes indicator nodes and disease nodes; The first prediction and evaluation module (102) is used to predict the disease classification and health status evaluation of the user based on the integrated model of the basic classification algorithm when the body component metabolism index is sufficient and the disease classification labels of the medical knowledge atlas library can fully cover clinical diseases; The second prediction and evaluation module (103) is used to predict the disease classification and health status evaluation of the user based on all the fixed indicators and user basic information indicators after the variables are adaptively added and the disease classification labels after the variables are automatically updated, in case the body component metabolism indicators are insufficient and / or the disease classification labels of the medical knowledge atlas library cannot fully cover the clinical diseases; The knowledge graph update module (104) collects the corresponding newly added clinical user information for the fixed indicators adaptively added to the variables and the corresponding user basic information indicators and the automatically updated disease classification labels, analyzes the corresponding newly added clinical user information, determines the entity nodes and relationships that do not exist in the current medical knowledge graph library, and adds the non-existent entity nodes and relationships to form an update to the medical knowledge graph library.

9. An electronic device, characterized in that: The method comprises a processor and a memory, wherein the memory stores a plurality of instructions, and the processor is used to read the instructions and execute the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a plurality of instructions, and the plurality of instructions can be read by a processor to execute the method according to any one of claims 1 to 7.

Citation Information

Patent Citations

  • Disease category evaluation device and method based on medical knowledge graph

    CN113035346A

  • Method for automatically complementing and updating knowledge graph

    CN115033706A

  • Method and system for constructing affair knowledge graph based on metabolic diseases

    CN115062162A

  • Health assessment method and device based on metabolism model, equipment and storage medium

    CN118039071A

  • Clustering device, method and program

    JP2016066269A

Cited By

  • Intelligent in-vivo component metabolism monitoring method and system

    CN119178873A