Intelligent physical examination data integration modeling system

Abstract

The invention belongs to the field of physical examination data analysis, and particularly discloses an intelligent physical examination data integration modeling system which comprises a data acquisition module, a data storage module, a data analysis modeling module and a visual display module. A matrix key generation method is designed to encrypt the physical examination data, and the encrypted physical examination data are dispersedly stored in a block chain network, so that relatively strong encryption protection is provided for the physical examination data, and the data are dispersed in a plurality of nodes, so that the risks of single-point failure and data loss are reduced; according to the method, a regression model and a fuzzy reasoning model are constructed to predict the health condition, an integrated bagging method and an integrated boosting method are adopted for training, respective advantages are fully exerted, and the prediction stability and the prediction precision are improved.

Description

Technical Field

[0001] The present invention relates to the field of physical examination data analysis, and in particular to an intelligent physical examination data integration modeling system. Background Art

[0002] Physical examination data analysis is a comprehensive evaluation and interpretation of the results of the physical examination. It combines the individual's health status, family medical history, etc. to make a comprehensive judgment and interpretation to determine whether there are health problems and give corresponding suggestions and treatment plans. At present, physical examination data analysis is largely done manually. Doctors may have certain subjectivity or omissions in the analysis of certain data due to differences in experience, cognition, etc., or fatigue from long-term work, resulting in inconsistent results and slow speed; physical examination data are usually stored in centralized databases, which are prone to vulnerabilities, triggering network attacks and leaking patient privacy information; the current intelligent analysis of physical examination data is highly dependent on data, and lacks flexibility for some special and complex data, resulting in inaccurate prediction results. Summary of the invention

[0003] In view of the above situation, in order to overcome the defects of the prior art, the present invention provides an intelligent physical examination data integration modeling system. In view of the technical problem that the current physical examination data analysis is largely performed manually, doctors may have certain subjectivity or omissions in the analysis of certain data due to differences in experience, cognition, etc., or fatigue from long-term work, resulting in inconsistent results and slow speed, the present invention constructs an intelligent physical examination data integration modeling system, which encrypts and stores physical examination data through blockchain technology, and uses machine learning to analyze physical examination data. It can process a large amount of physical examination data in a short time, is not affected by human emotions and subjective factors, and can quickly obtain analysis results; physical examination data are usually stored in a centralized database, which is prone to vulnerabilities, triggering network attacks, and leaking patients' privacy information. In order to solve the technical problem, the present invention designs a matrix key generation method to encrypt the physical examination data, and stores the encrypted physical examination data in a dispersed manner in the blockchain network, which provides strong encryption protection for the physical examination data and ensures that the encrypted physical examination data is difficult to tamper with once stored, and disperses the data in multiple nodes, reducing the risk of single point failure and data loss; in view of the technical problem that the current intelligent analysis of physical examination data is highly dependent on data and lacks flexibility for some special and complex data, resulting in inaccurate prediction results, the present invention constructs a regression model and a fuzzy reasoning model to predict the health status, and adopts integrated bagging and integrated boosting methods for training, giving full play to their respective advantages, improving the stability and accuracy of the prediction, and avoiding the problem of overfitting of the prediction.

[0004] The technical solution adopted by the present invention is as follows: The present invention provides an intelligent physical examination data integration modeling system, which includes a data acquisition module, a data storage module, a data analysis modeling module and a visualization display module;

[0005] The data acquisition module collects physical examination data and transmits the physical examination data to the data storage module;

[0006] The data storage module includes a local server and a decentralized database, and uses a matrix key-based method to encrypt the physical examination data and then save the physical examination data using blockchain technology;

[0007] The data analysis modeling module uses a fuzzy reasoning method based on integrated machine learning to establish a data analysis model, builds a regression model and a fuzzy reasoning model to predict health status, and uses integrated bagging and integrated boosting methods for training to analyze physical examination data and judge physical health status;

[0008] The visualization display module displays the physical health status obtained from the analysis of the physical examination data in the form of charts and graphs.

[0009] Furthermore, the data storage module uses a blockchain method based on a matrix key to store the physical examination data. The blockchain method based on a matrix key specifically includes the following steps:

[0010] Step B1: Calculate the frequency of the value of each physical examination item in the physical examination data to obtain a frequency matrix, divide the frequency matrix by a prime number to reduce the frequency of the physical examination data to obtain a reduced frequency matrix, wherein the prime number has a value range of [29, 97], add the values ​​of the reduced frequency matrix to their respective positions in the matrix, and convert them into characters to obtain a character matrix, and generate a symbol matrix according to the character matrix, wherein the ASCII value of each symbol in the symbol matrix is ​​greater than the character matrix, and if there are repeated values ​​in the symbol matrix, remove the repeated values ​​from the symbol matrix to obtain a unique matrix, wherein the frequency matrix, reduced frequency matrix, character matrix, symbol matrix and unique matrix are all 6-dimensional matrices;

[0011] Step B2: The unique matrix is ​​stored as a key in a local server, and the physical examination data is encrypted using the unique matrix to obtain encrypted data, and the encrypted data is stored in a decentralized database, wherein the decentralized database includes an IPFS and an OrbitDB database, and the OrbitDB database stores the encrypted data. IPFS performs a hash operation on the encrypted data to obtain hash data, and the hash data is sent to the blockchain network;

[0012] Step B3: The blockchain network includes a registration unit, a certificate issuing unit, a key generating unit and a distributed storage unit. The doctor submits a registration request to the blockchain network through the registration unit. After the registration request is approved, the key generating unit generates a key pair for the doctor, the certificate issuing unit creates a user certificate, and generates a user token, and sends the user token and the key pair to the doctor together;

[0013] Step B4: When the doctor accesses the physical examination data, the blockchain network calls the chain code containing the access control policy to automatically trigger the smart contract for identity authentication. If the identity authentication passes, the physical examination data is read; otherwise, an error prompt is returned.

[0014] Furthermore, in the data analysis modeling module, the fuzzy reasoning method based on integrated machine learning specifically includes the following steps:

[0015] Step C1: Collect an initial physical examination data set, which includes physical examination data of historical physical examination subjects and corresponding physical health conditions. The physical examination data in the initial physical examination data set are converted into standardized data through data conversion, missing value check and data scaling. The data is used as a feature for machine learning, and the corresponding physical condition is used as a label. The standardized data is divided into a training set and a test set in a ratio of 7:3;

[0016] Step C2: Establish and initialize a regression model, input the training set into the regression model, and use integrated bagging and integrated boosting methods for training;

[0017] Step C3: Assume that there are m features in the training set, randomly select k features from them, record them as random features, build a decision tree, find the optimal split of the k selected features, use the optimal split to divide the node into child nodes until reaching the leaf node, repeat this process until you get a forest, and predict the health status. The formula used is as follows: ;

[0018] In the formula, is the predicted health status, is the number of decision trees, are the characteristics of the training set, It is Decision tree pair The prediction results;

[0019] Step C4: Dynamically adjust the weight of the training set, initialize the training set and assign the same initial weight to each training data. Set the initial weight to 1 / N, where N is the total number of data in the training set. Find the training data that the regression model predicts incorrectly, and calculate the weight of each decision tree in the final decision. The formula used is as follows: ;

[0020] In the formula, is the weight of each decision tree in the final decision, is the total number of training data that the regression model predicted incorrectly;

[0021] Step C5: Iteratively update the weights using the following formula: ;

[0022] In the formula, For the The updated weights after iterations are In the base of natural logarithms;

[0023] Step C5: Establish and initialize the fuzzy inference model, set the fuzzy inference rules, input the health status predicted by the regression model into the fuzzy inference model, and output the health status vector value as the final health status prediction result.

[0024] The beneficial effects achieved by the present invention using the above scheme are as follows:

[0025] (1) In view of the current physical examination data analysis is largely done manually, doctors may have certain subjectivity or omissions in the analysis of certain data due to differences in experience, cognition, etc., or fatigue from long-term work, resulting in inconsistent results and slow speed. The present invention constructs an intelligent physical examination data integration modeling system, which encrypts and stores physical examination data through blockchain technology and analyzes physical examination data using machine learning. It can process a large amount of physical examination data in a short time, is not affected by human emotions and subjective factors, and quickly obtains analysis results;

[0026] (2) In view of the technical problem that physical examination data is usually stored in a centralized database, which is prone to loopholes, triggering network attacks and leaking patient privacy information, the present invention designs a matrix key generation method to encrypt the physical examination data, and stores the encrypted physical examination data in a decentralized manner in the blockchain network, providing strong encryption protection for the physical examination data, ensuring that the encrypted physical examination data is difficult to tamper with once stored, and distributing the data across multiple nodes, reducing the risk of single point failure and data loss;

[0027] (3) In view of the technical problems that the current intelligent analysis of physical examination data is highly dependent on data and lacks flexibility for some special and complex data, resulting in inaccurate prediction results, the present invention constructs a regression model and a fuzzy reasoning model to predict health status, and uses integrated bagging and integrated boosting methods for training, giving full play to their respective advantages, improving the stability and accuracy of the prediction, and avoiding the problem of overfitting of the prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 A module connection diagram of an intelligent physical examination data integration modeling system provided by the present invention.

[0029] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention. DETAILED DESCRIPTION

[0030] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments; based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0031] Example 1: See Figure 1 , this embodiment provides an intelligent physical examination data integration modeling system, the intelligent physical examination data integration modeling system includes a data acquisition module, a data storage module, a data analysis modeling module and a visualization display module;

[0032] The data acquisition module collects physical examination data and transmits the physical examination data to the data storage module;

[0033] The data storage module includes a local server and a decentralized database, and uses a matrix key-based method to encrypt the physical examination data and then save the physical examination data using blockchain technology;

[0034] The data analysis modeling module uses a fuzzy reasoning method based on integrated machine learning to establish a data analysis model, builds a regression model and a fuzzy reasoning model to predict health status, and uses integrated bagging and integrated boosting methods for training to analyze physical examination data and judge physical health status;

[0035] The visualization display module displays the physical health status obtained from the analysis of the physical examination data in the form of charts and graphs, which is convenient for doctors and examinees to understand.

[0036] Through the above operations, in view of the technical problem that the current physical examination data analysis is largely performed manually, doctors may have certain subjectivity or omissions in the analysis of certain data due to differences in experience, cognition, etc., or fatigue from long-term work, resulting in inconsistent results and slow speed. The present invention constructs an intelligent physical examination data integration modeling system, which encrypts and stores physical examination data through blockchain technology, and uses machine learning to analyze physical examination data. It can process a large amount of physical examination data in a short time, is not affected by human emotions and subjective factors, and can quickly obtain analysis results.

[0037] Example 2: See Figure 1 This embodiment is based on the above embodiment. The data storage module uses a blockchain method based on a matrix key to store physical examination data. The blockchain method based on a matrix key specifically includes the following steps:

[0038] Step B1: Calculate the frequency of the value of each physical examination item in the physical examination data to obtain a frequency matrix, divide the frequency matrix by a prime number to reduce the frequency of the physical examination data to obtain a reduced frequency matrix, wherein the prime number has a value range of [29, 97], add the values ​​of the reduced frequency matrix to their respective positions in the matrix, and convert them into characters to obtain a character matrix, and generate a symbol matrix according to the character matrix, wherein the ASCII value of each symbol in the symbol matrix is ​​greater than the character matrix, and if there are repeated values ​​in the symbol matrix, remove the repeated values ​​from the symbol matrix to obtain a unique matrix, wherein the frequency matrix, reduced frequency matrix, character matrix, symbol matrix and unique matrix are all 6-dimensional matrices;

[0039] Step B2: The unique matrix is ​​stored as a key in a local server, and the physical examination data is encrypted using the unique matrix to obtain encrypted data, and the encrypted data is stored in a decentralized database, wherein the decentralized database includes an IPFS and an OrbitDB database, and the OrbitDB database stores the encrypted data. IPFS performs a hash operation on the encrypted data to obtain hash data, and the hash data is sent to the blockchain network;

[0040] Step B3: The blockchain network includes a registration unit, a certificate issuing unit, a key generating unit and a distributed storage unit. The doctor submits a registration request to the blockchain network through the registration unit. After the registration request is approved, the key generating unit generates a key pair for the doctor, the certificate issuing unit creates a user certificate, and generates a user token, and sends the user token and the key pair to the doctor together;

[0041] Step B4: When the doctor accesses the physical examination data, the blockchain network calls the chain code containing the access control policy to automatically trigger the smart contract for identity authentication. If the identity authentication passes, the physical examination data is read; otherwise, an error prompt is returned.

[0042] Through the above operations, in view of the technical problem that physical examination data are usually stored in a centralized database, which is prone to loopholes, triggering network attacks, and leaking patient privacy information, the present invention designs a matrix key generation method to encrypt the physical examination data, and stores the encrypted physical examination data in a dispersed manner in the blockchain network, providing strong encryption protection for the physical examination data, ensuring that the encrypted physical examination data is difficult to tamper with once stored, and dispersing the data across multiple nodes, reducing the risk of single point failure and data loss.

[0043] Example 3: See Figure 1 This embodiment is based on the above embodiment. In the data analysis modeling module, the fuzzy reasoning method based on integrated machine learning specifically includes the following steps:

[0044] Step C1: Collect an initial physical examination data set, which includes physical examination data of historical physical examination subjects and corresponding physical health conditions. The physical examination data in the initial physical examination data set are converted into standardized data through data conversion, missing value check and data scaling. The data is used as a feature for machine learning, and the corresponding physical condition is used as a label. The standardized data is divided into a training set and a test set in a ratio of 7:3;

[0045] Step C2: Establish and initialize a regression model, input the training set into the regression model, and use integrated bagging and integrated boosting methods for training;

[0046] Step C3: Assume that there are m features in the training set, randomly select k features from them, record them as random features, build a decision tree, find the optimal split of the k selected features, use the optimal split to divide the node into child nodes until reaching the leaf node, repeat this process until you get a forest, and predict the health status. The formula used is as follows: ;

[0047] In the formula, is the predicted health status, is the number of decision trees, are the characteristics of the training set, It is Decision tree pair The prediction results;

[0048] Step C4: Dynamically adjust the weight of the training set, initialize the training set and assign the same initial weight to each training data. Set the initial weight to 1 / N, where N is the total number of data in the training set. Find the training data that the regression model predicts incorrectly, and calculate the weight of each decision tree in the final decision. The formula used is as follows: ;

[0049] In the formula, is the weight of each decision tree in the final decision, is the total number of training data that the regression model predicted incorrectly;

[0050] Step C5: Iteratively update the weights using the following formula: ;

[0051] In the formula, For the The updated weights after iterations are In the base of natural logarithms;

[0052] Step C5: Establish and initialize the fuzzy inference model, set the fuzzy inference rules, input the health status predicted by the regression model into the fuzzy inference model, and output the health status vector value as the final health status prediction result.

[0053] Through the above operations, in order to solve the technical problem that the current intelligent analysis of physical examination data is highly dependent on data and lacks flexibility for some special and complex data, resulting in inaccurate prediction results, the present invention constructs a regression model and a fuzzy reasoning model to predict health status, and adopts integrated bagging and integrated boosting methods for training, giving full play to their respective advantages, improving the stability and accuracy of prediction, and avoiding the problem of overfitting of prediction.

[0054] Embodiment 4: This embodiment is based on the above embodiment. The physical examination data of historical physical examination subjects mentioned in this scheme is obtained by launching a large-scale health survey project. It is implemented in compliance with the provisions of the Personal Information Protection Law and other relevant laws, is obtained under legal circumstances, and is a case of legal use.

[0055] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0056] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

[0057] The present invention and its embodiments are described above, and such description is not restrictive. The drawings show only one embodiment of the present invention, and the actual structure is not limited thereto. In short, if ordinary technicians in the field are inspired by it, without departing from the purpose of the invention, they can design a structure and embodiment similar to the technical solution without creativity, which should belong to the protection scope of the present invention.

Claims

1. An intelligent physical examination data integration modeling system, characterized in that: The system includes data acquisition module, data storage module, data analysis and modeling module and visualization display module; The data acquisition module collects physical examination data and transmits the physical examination data to the data storage module; The data storage module includes a local server and a decentralized database, and uses a matrix key-based method to encrypt the physical examination data and then save the physical examination data using blockchain technology; The data analysis modeling module uses a fuzzy reasoning method based on integrated machine learning to establish a data analysis model, builds a regression model and a fuzzy reasoning model to predict health status, and uses integrated bagging and integrated boosting methods for training to analyze physical examination data and judge physical health status; The visualization display module displays the physical health status obtained from the analysis of the physical examination data in the form of charts and graphs.

2. According to the intelligent physical examination data integration modeling system of claim 1, the data storage module uses a blockchain method based on a matrix key to store physical examination data, and the blockchain method based on a matrix key specifically includes the following steps: Step B1: Calculate the frequency of the value of each physical examination item in the physical examination data to obtain a frequency matrix, divide the frequency matrix by a prime number to reduce the frequency of the physical examination data to obtain a reduced frequency matrix, wherein the prime number has a value range of [29, 97], add the values ​​of the reduced frequency matrix to their respective positions in the matrix, and convert them into characters to obtain a character matrix, and generate a symbol matrix according to the character matrix, wherein the ASCII value of each symbol in the symbol matrix is ​​greater than the character matrix, and if there are repeated values ​​in the symbol matrix, remove the repeated values ​​from the symbol matrix to obtain a unique matrix, wherein the frequency matrix, reduced frequency matrix, character matrix, symbol matrix and unique matrix are all 6-dimensional matrices; Step B2: The unique matrix is ​​stored as a key in a local server, and the physical examination data is encrypted using the unique matrix to obtain encrypted data, and the encrypted data is stored in a decentralized database, wherein the decentralized database includes an IPFS and an OrbitDB database, and the OrbitDB database stores the encrypted data. IPFS performs a hash operation on the encrypted data to obtain hash data, and the hash data is sent to the blockchain network; Step B3: The blockchain network includes a registration unit, a certificate issuing unit, a key generating unit and a distributed storage unit. The doctor submits a registration request to the blockchain network through the registration unit. After the registration request is approved, the key generating unit generates a key pair for the doctor, the certificate issuing unit creates a user certificate, and generates a user token, and sends the user token and the key pair to the doctor together; Step B4: When the doctor accesses the physical examination data, the blockchain network calls the chain code containing the access control policy to automatically trigger the smart contract for identity authentication. If the identity authentication passes, the physical examination data is read; otherwise, an error prompt is returned.

3. The intelligent physical examination data integration modeling system according to claim 1, characterized in that: In the data analysis modeling module, the fuzzy reasoning method based on integrated machine learning specifically includes the following steps: Step C1: Collect an initial physical examination data set, which includes physical examination data of historical physical examination subjects and corresponding physical health conditions. The physical examination data in the initial physical examination data set are converted into standardized data through data conversion, missing value check and data scaling. The data is used as a feature for machine learning, and the corresponding physical condition is used as a label. The standardized data is divided into a training set and a test set in a ratio of 7:3; Step C2: Establish and initialize a regression model, input the training set into the regression model, and use integrated bagging and integrated boosting methods for training; Step C3: Assume that there are m features in the training set, randomly select k features from them, record them as random features, build a decision tree, find the optimal split of the k selected features, use the optimal split to divide the node into child nodes until reaching the leaf node, repeat this process until you get a forest, and predict the health status. The formula used is as follows: ； In the formula, is the predicted health status, is the number of decision trees, are the characteristics of the training set, It is Decision tree pair The prediction results; Step C4: Dynamically adjust the weight of the training set, initialize the training set and assign the same initial weight to each training data. Set the initial weight to 1 / N, where N is the total number of data in the training set. Find the training data that the regression model predicts incorrectly, and calculate the weight of each decision tree in the final decision. The formula used is as follows: ； In the formula, is the weight of each decision tree in the final decision, is the total number of training data that the regression model predicted incorrectly; Step C5: Iteratively update the weights using the following formula: ； In the formula, For the The updated weights after iterations are In the base of natural logarithms; Step C5: Establish and initialize the fuzzy inference model, set the fuzzy inference rules, input the health status predicted by the regression model into the fuzzy inference model, and output the health status vector value as the final health status prediction result.

Images