A health examination report data analysis and management system based on the Internet

Abstract

The present invention relates to the technical field of data processing, and provides a health checkup report data analysis and management system based on the Internet, including: analyzing the frequent pathologies corresponding to the physical examination users; a data query and calculation module, used to query the detection indexes corresponding to the frequent pathologies, and calculate the index deviation corresponding to the detection indexes; an incidence rate calculation module, used to calculate the pathological incidence rates corresponding to the life behaviors; a correlation coefficient calculation module, used to extract the user sign factors corresponding to the physical examination users, respectively standardize the abnormal physical examination data and the user sign factors, obtain the standardized data and the standardized factors, extract the data features corresponding to the standardized data, and calculate the correlation coefficient between the data features and the standardized factors; a program formulation module, used to determine the associated signs in the user signs according to the correlation coefficient, and formulate the health management plan corresponding to the frequent pathologies. The present invention aims to improve the accuracy of health checkup report data analysis and management.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to an Internet-based health examination report data analysis and management system. Background Art

[0002] A health check-up refers to a comprehensive assessment and screening of an individual's physical condition through a series of physical examinations, laboratory tests, and imaging examinations. It can detect potential health problems or diseases in the human body and take preventive or therapeutic measures in a timely manner to achieve the purpose of maintaining health, preventing disease and early diagnosis.

[0003] The existing health examination report data analysis and management system adopts an objective analysis system. At present, in the research of physical examination data, most of them only use descriptive analysis methods. However, due to the different dimensions of various health standards, it is difficult to measure personal health levels, which in turn leads to a decrease in the accuracy of health management. Therefore, a system is needed to improve the accuracy of health examination report data analysis and management. Summary of the invention

[0004] The present invention provides a health examination report data analysis and management system based on the Internet, the main purpose of which is to improve the accuracy of health examination report data analysis and management.

[0005] To achieve the above purpose, the present invention provides an Internet-based health examination report data analysis and management system, comprising:

[0006] A pathology analysis module is used to collect the health examination data of the physical examination user, screen out abnormal physical examination data in the health examination data, and analyze the frequent pathology corresponding to the physical examination user according to the abnormal physical examination data;

[0007] A data query and calculation module, used to query the detection index corresponding to the frequent pathology, calculate the index deviation corresponding to the detection index according to the abnormal physical examination data, and conduct a health assessment on the physical examination user in combination with the detection index, the index deviation and the frequent pathology to obtain the user's health level;

[0008] An incidence rate calculation module is used to obtain the life behavior corresponding to the physical examination user, and calculate the pathological incidence rate corresponding to the life behavior by combining the abnormal physical examination data and the frequent pathology;

[0009] A correlation coefficient calculation module, used to extract the user physical sign factors corresponding to the physical examination user, input the abnormal physical examination data and the user physical sign factors as input data into a preset regression model, use the feedforward neural network in the regression model to standardize the abnormal physical examination data and the user physical sign factors respectively to obtain standardized data and standardized factors, use the convolutional neural network in the regression model to extract data features corresponding to the standardized data, and use the regression function in the regression model to calculate the correlation coefficient between the data features and the standardized factors;

[0010] A plan formulation module is used to determine the associated physical signs of the user according to the correlation coefficient, and formulate a health management plan corresponding to the frequent pathology in combination with the associated physical signs, the user's health level and the pathological incidence rate.

[0011] Further, analyzing the frequent pathology corresponding to the physical examination user according to the abnormal physical examination data includes:

[0012] Identifying abnormal items corresponding to the abnormal physical examination data, and calculating item similarities between the abnormal items;

[0013] Classifying the abnormal items according to the item similarities to obtain classified items;

[0014] Extracting the item detection value corresponding to the classified item from the abnormal physical examination data, and querying the historical medical data corresponding to the physical examination user;

[0015] Extracting historical medical advice from the historical medical data, and determining the historical pathology of the physical examination user according to the historical medical advice;

[0016] Determine the current pathology of the physical examination user by combining the historical pathology and the item detection value;

[0017] The pathology frequency corresponding to the current pathology is counted, and the frequent pathology corresponding to the physical examination user is determined according to the pathology frequency.

[0018] Further, the calculating the project similarity between the abnormal projects includes:

[0019] The project similarity between the abnormal projects is calculated by the following formula:

[0020]

[0021] Among them, A represents the item similarity between abnormal items, B a represents the item vector corresponding to the ath item in the abnormal item, E a+1represents the item vector corresponding to the a+1th item in the abnormal items, and a and a+1 represent the item sequence numbers corresponding to the abnormal items.

[0022] Furthermore, the health assessment of the physical examination user is performed in combination with the detection index, the index deviation and the frequent pathology to obtain the user's health level, including:

[0023] Calculating the contribution between the detection index and the frequent pathology, and assigning an index weight corresponding to the detection index according to the contribution;

[0024] Calculate the health score of the user undergoing physical examination by combining the indicator deviation and the indicator weight;

[0025] According to the health score, a user health level corresponding to the physical examination user is set.

[0026] Furthermore, the combining of the indicator deviation and the indicator weight to calculate the health score corresponding to the physical examination user includes:

[0027] The health score corresponding to the physical examination user is calculated by the following formula:

[0028]

[0029] Among them, D represents the health score of the physical examination user, b represents the serial number of the detection index corresponding to the index deviation and index weight, t represents the number of detection indicators, β b Indicates the index deviation corresponding to the bth index in the detection index, F b Indicates the indicator weight corresponding to the bth indicator in the detection indicator.

[0030] Furthermore, the combining of the abnormal physical examination data and the frequent pathology to calculate the pathological incidence rate corresponding to the living behavior includes:

[0031] Performing data cleaning processing on the abnormal physical examination data to obtain target physical examination data, and extracting data variables in the target physical examination data;

[0032] Combining the data variables, the target physical examination data and the life behavior, constructing a data scatter plot corresponding to the frequent pathology;

[0033] Calculating the slope coefficient corresponding to the data scatter plot, and determining the pathological correlation degree corresponding to the life behavior according to the slope coefficient;

[0034] Calculating the relative risk coefficient of each behavior in the life behavior according to the pathological correlation degree;

[0035] The pathological incidence rate corresponding to the living behavior is calculated according to the relative risk coefficient and the pathological correlation degree.

[0036] Furthermore, the calculating of the pathological incidence rate corresponding to the living behavior according to the relative risk coefficient and the pathological correlation degree includes:

[0037] The pathological incidence rate corresponding to the lifestyle behavior is calculated by the following formula:

[0038]

[0039] Among them, φ d represents the relative risk coefficient corresponding to the dth behavior in life behavior, H d represents the pathological correlation degree corresponding to the dth behavior in life behavior, d represents the serial number corresponding to the life behavior, 1-H d Represents the irrelevance coefficient between the dth behavior in life behavior and frequent pathology.

[0040] Furthermore, the feedforward neural network in the regression model is used to perform standardization processing on the abnormal physical examination data and the user's physical sign factors to obtain standardized data and standardized factors, including:

[0041] Using the input layer in the feedforward neural network to scale the abnormal physical examination data and the user's physical sign factors, respectively, to obtain scaled data and scaled physical sign factors;

[0042] Utilizing the linear function in the feedforward neural network to respectively calculate the linear values ​​corresponding to the scaled data and the scaled vital sign factors, to obtain the data linear value and the factor linear value;

[0043] Calculating the data standard deviation corresponding to the scaled data using a standard deviation function in the feedforward neural network according to the data linear value;

[0044] According to the factor linear value, using the standard deviation function to calculate the factor standard deviation corresponding to the factor linear value;

[0045] According to the data standard deviation and the factor standard deviation, the scaled data and the scaled physical sign factors are standardized respectively using the standardization function in the regression model to obtain standardized data and standardized factors.

[0046] Further, the calculating the correlation coefficient between the data feature and the standardization factor by using the regression function in the regression model includes:

[0047] The regression function includes:

[0048]

[0049] Among them, Q represents the correlation coefficient between the data feature and the standardized factor, e represents the serial number corresponding to the data feature, e+1 represents the serial number corresponding to the standardized factor, and P e Represents the vector corresponding to the e-th feature in the data features, T e+1 represents the vector corresponding to the e+1th factor in the standardized factor, minmin(ln(P e -T e+1 )) represents the minimum difference between the logarithm of the difference between the vector corresponding to the e-th feature and the vector corresponding to the e+1th factor, maxmax(ln(P e -T e+1 )) represents the maximum difference between the logarithm of the difference between the vector corresponding to the e-th feature and the vector corresponding to the e+1-th factor.

[0050] A method for analyzing and managing health examination report data based on the Internet, characterized in that the method comprises:

[0051] Collecting the health examination data of the user undergoing the physical examination, screening out abnormal physical examination data in the health examination data, and analyzing the frequent pathology corresponding to the user undergoing the physical examination based on the abnormal physical examination data;

[0052] Query the detection index corresponding to the frequent pathology, calculate the index deviation corresponding to the detection index according to the abnormal physical examination data, and conduct a health assessment on the physical examination user by combining the detection index, the index deviation and the frequent pathology to obtain the user's health level;

[0053] Acquire the life behavior corresponding to the physical examination user, and calculate the pathological incidence rate corresponding to the life behavior by combining the abnormal physical examination data and the frequent pathology;

[0054] Extracting user physical sign factors corresponding to the physical examination user, inputting the abnormal physical examination data and the user physical sign factors as input data into a preset regression model, using the feedforward neural network in the regression model to perform standardization processing on the abnormal physical examination data and the user physical sign factors respectively to obtain standardized data and standardized factors, using the convolutional neural network in the regression model to extract data features corresponding to the standardized data, and using the regression function in the regression model to calculate the correlation coefficient between the data features and the standardized factors;

[0055] According to the correlation coefficient, the associated physical signs among the user's physical signs are determined, and a health management plan corresponding to the frequent pathology is formulated in combination with the associated physical signs, the user's health level and the pathology incidence rate.

[0056] The present invention analyzes the frequent pathologies corresponding to the physical examination users based on the abnormal physical examination data, and can understand the multiple diseases in the physical examination users, so as to facilitate the relationship between subsequent diseases and related factors. The present invention conducts a health assessment on the physical examination users by combining the detection indicators, the indicator deviation and the frequent pathologies, so as to facilitate the classification of the health of the physical examination users, so as to understand the health level of the physical examination users and facilitate the calculation accuracy of the disease risk coefficient of subsequent life behaviors. The present invention calculates the pathological incidence rate corresponding to the life behaviors by combining the abnormal physical examination data and the frequent pathologies, so as to understand the life behaviors. The probability of occurrence of the corresponding pathology is obtained, thereby improving the accuracy of the analysis of the pathological influencing factors of the physical examination user, and providing a basis for the formulation of subsequent health management plans. The present invention uses the feedforward neural network in the regression model to standardize the abnormal physical examination data and the user's physical signs, respectively, so as to eliminate the difference between the abnormal physical examination data and the user's physical signs, so as to facilitate the subsequent calculation of correlation coefficients. The present invention combines the associated signs and the pathological incidence rate to formulate a health management plan corresponding to the frequent pathology, so as to improve the accuracy of the health management of the physical examination user through the health management plan. Therefore, an Internet-based health examination report data analysis and management system provided by an embodiment of the present invention can improve the accuracy of health examination report data analysis and management. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 A functional module diagram of an Internet-based health examination report data analysis and management system provided by one embodiment of the present invention;

[0058] Figure 2 A flowchart of a method for analyzing and managing health examination report data based on the Internet is provided in accordance with an embodiment of the present invention.

[0059] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0060] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution 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 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.

[0061] In fact, the server-side device deployed by the Internet-based health check report data analysis and management system may be composed of one or more devices. The above-mentioned Internet-based health check report data analysis and management system can be implemented as: business instance, virtual machine, hardware device. For example, the Internet-based health check report data analysis and management system can be implemented as a business instance deployed on one or more devices in the cloud node. In simple terms, the live broadcast service system can be understood as a software deployed on the cloud node, which is used to provide Internet-based health check report data analysis and management services for each user terminal. Alternatively, the Internet-based health check report data analysis and management system can also be implemented as a virtual machine deployed on one or more devices in the cloud node. Application software for managing each user terminal is installed in the virtual machine. Alternatively, the Internet-based health check report data analysis and management system can also be implemented as a server-side composed of many hardware devices of the same or different types, and one or more hardware devices are set to provide Internet-based health check report data analysis and management services for each user terminal.

[0062] In terms of implementation, the Internet-based health checkup report data analysis and management system and the user end are adapted to each other. That is, the Internet-based health checkup report data analysis and management system is an application installed on the cloud service platform, and the user end is a client that establishes a communication connection with the application; or the Internet-based health checkup report data analysis and management system is implemented as a website, and the user end is implemented as a web page; or the Internet-based health checkup report data analysis and management system is implemented as a cloud service platform, and the user end is implemented as a small program in an instant messaging application.

[0063] Reference Figure 1 , which is a functional module diagram of an Internet-based health examination report data analysis and management system provided by an embodiment of the present invention.

[0064] The Internet-based health checkup report data analysis and management system 100 of the present invention can be set in a cloud server. In terms of implementation, it can be used as one or more service devices, or it can be installed as an application on the cloud (such as a server or server cluster of a live broadcast service operator), or it can be developed as a website. According to the functions implemented, the Internet-based health checkup report data analysis and management system 100 includes a pathology analysis module 101, a data query and calculation module 102, an incidence calculation module 103, a correlation coefficient calculation module 104, and a program formulation module 105.

[0065] In the embodiment of the present invention, in the tracking of the health check report data analysis management based on the Internet, each of the above modules can be implemented independently and called with other modules. The call here can be understood as a module that can connect to multiple modules of another type and provide corresponding services to the multiple modules connected to it. In the health check report data analysis management system based on the Internet provided by the embodiment of the present invention, there is no need to modify the program code. The scope of application of the health check report data analysis management architecture based on the Internet can be adjusted by adding modules and directly calling them to achieve cluster-based horizontal expansion, so as to achieve the purpose of quickly and flexibly expanding the health check report data analysis management system based on the Internet. In actual applications, the above modules can be set in the same device or different devices, or they can be set in virtual devices, such as service instances in cloud servers.

[0066] In the following, in conjunction with specific embodiments, each component and specific workflow of the Internet-based health examination report data analysis and management system are described.

[0067] The pathology analysis module 101 is used to collect the health examination data of the health examination user, filter out abnormal health examination data in the health examination data, and analyze the frequent pathology corresponding to the health examination user according to the abnormal health examination data.

[0068] The present invention analyzes the frequent pathologies corresponding to the physical examination users based on the abnormal physical examination data, and can understand the multiple diseases among the physical examination users, thereby facilitating the relationship between subsequent diseases and related factors, wherein the healthy physical examination data are data obtained by performing a systematic examination on the physical examination users, and the abnormal physical examination data are data with abnormalities in the healthy physical examination data, such as high blood sugar or high blood lipids, etc. The frequent pathologies are multiple diseases among the physical examination users. Furthermore, screening out the abnormal physical examination data in the healthy physical examination data can be achieved by combining medical standards, for example, the blood pressure range is 80-100, and if it is not within this range, it indicates abnormality.

[0069] As an embodiment of the present invention, analyzing the frequent pathologies corresponding to the physical examination user based on the abnormal physical examination data includes: identifying the abnormal items corresponding to the abnormal physical examination data, calculating the item similarities between the abnormal items, classifying the abnormal items according to the item similarities to obtain classified items, extracting item detection values ​​corresponding to the classified items from the abnormal physical examination data, querying the historical medical data corresponding to the physical examination user, extracting historical medical suggestions from the historical medical data, determining the historical pathology of the physical examination user according to the historical medical suggestions, determining the current pathology of the physical examination user by combining the historical pathology and the item detection values, counting the pathology frequency corresponding to the current pathology, and determining the frequent pathology corresponding to the physical examination user according to the pathology frequency.

[0070] Among them, the abnormal item is the corresponding item in the abnormal physical examination data, such as a blood item, the item similarity represents the degree of similarity between the abnormal items, the classified item is the item with similarity in the abnormal items, the item detection value is the detection value corresponding to the classified item, and the historical medical advice is the doctor's diagnosis corresponding to the detection result of the physical examination user in the historical medical data.

[0071] Furthermore, identifying the abnormal items corresponding to the abnormal physical examination data can be achieved through OCR text recognition technology, classifying the abnormal items can be achieved through the K-means clustering method, extracting the item detection values ​​corresponding to the classified items from the abnormal physical examination data can be achieved through the l eft function, the historical medical data corresponding to the physical examination user can be obtained by querying from the medical database, extracting historical medical suggestions from the historical medical data can be achieved through the above-mentioned l eft function, the historical pathology of the physical examination user can be determined through the text information description in the historical medical suggestions, whether the historical pathology is cured can be analyzed according to the item detection value, and the corresponding other pathologies can be determined, so as to determine the current pathology of the physical examination user, and the frequent pathology corresponding to the physical examination user can be determined according to the pathology with the highest frequency of occurrence.

[0072] Further, as an optional embodiment of the present invention, the calculating the item similarity between the abnormal items includes:

[0073] The project similarity between the abnormal projects is calculated by the following formula:

[0074]

[0075] Among them, A represents the item similarity between abnormal items, B a represents the item vector corresponding to the ath item in the abnormal item, E a+1represents the item vector corresponding to the a+1th item in the abnormal items, and a and a+1 represent the item sequence numbers corresponding to the abnormal items.

[0076] The data query and calculation module 102 is used to query the detection index corresponding to the frequent pathology, calculate the index deviation corresponding to the detection index based on the abnormal physical examination data, and conduct a health assessment on the physical examination user in combination with the detection index, the index deviation and the frequent pathology to obtain the user's health level.

[0077] The present invention conducts a health assessment on the physical examination user by combining the detection index, the index deviation and the frequent pathology, so as to classify the health of the physical examination user, thereby understanding the health level of the physical examination user and facilitating the calculation accuracy of the disease risk coefficient of subsequent life behavior, wherein the detection index is the detection target corresponding to the frequent pathology, such as blood cell count, hemoglobin concentration, platelet count, white blood cell count and other indicators, the index deviation indicates the degree of deviation between the detection index and the standard value, and the user health level indicates the corresponding health level of the physical examination user. Furthermore, the detection data corresponding to the detection index can be extracted from the abnormal physical examination data, and the standard range corresponding to the detection index can be queried through a medical institution, the difference between the detection data and the standard range can be calculated, and the index deviation corresponding to the detection index can be determined according to the difference.

[0078] As an embodiment of the present invention, the health assessment of the physical examination user is performed in combination with the detection index, the index deviation and the frequent pathology to obtain the user health level, including: calculating the contribution between the detection index and the frequent pathology, allocating the index weight corresponding to the detection index according to the contribution, calculating the health score corresponding to the physical examination user in combination with the index deviation and the index weight, and setting the user health level corresponding to the physical examination user according to the health score.

[0079] Among them, the contribution degree represents the influence of the detection index on the occurrence of the frequent pathology, the index weight represents the importance corresponding to the detection index, and the health score represents the health score corresponding to the physical examination user, so as to understand the health level of the physical examination user.

[0080] Furthermore, the specific steps for calculating the contribution between the detection index and the frequent pathology are as follows: first, the correlation between the detection index and the frequent pathology is calculated by the Kendall correlation coefficient algorithm, and the correlation is used as the contribution corresponding to the detection index. The contribution can be converted into a score value, and the score value is used as the indicator weight corresponding to the detection index. For example, if the contribution is 80%, the indicator weight is 0.8. The mapping relationship between the health level and the health score is designed according to professional medical knowledge. For example, health is 80-100 points, sub-health is 60-80 points, mild abnormality is 40-60 points, and moderate abnormality is below 40 points. The user health level corresponding to the physical examination user is set in this way.

[0081] Further, as an optional embodiment of the present invention, the combining of the indicator deviation and the indicator weight to calculate the health score corresponding to the physical examination user includes:

[0082] The health score corresponding to the physical examination user is calculated by the following formula:

[0083]

[0084] Among them, D represents the health score of the physical examination user, b represents the serial number of the detection index corresponding to the index deviation and index weight, t represents the number of detection indicators, β b Indicates the index deviation corresponding to the bth index in the detection index, F b Indicates the indicator weight corresponding to the bth indicator in the detection indicator.

[0085] The morbidity rate calculation module 103 is used to obtain the life behavior corresponding to the physical examination user, and calculate the pathology morbidity rate corresponding to the life behavior in combination with the abnormal physical examination data and the frequent pathology.

[0086] The present invention calculates the pathological incidence rate corresponding to the living behavior by combining the abnormal physical examination data and the frequent pathology, so as to understand the probability of occurrence of the pathology corresponding to the living behavior, thereby improving the accuracy of analysis of the pathological influencing factors of the physical examination user and providing a basis for the formulation of subsequent health management plans, wherein the pathological incidence rate indicates the probability of the pathology caused by the living behavior. Furthermore, obtaining the living behavior corresponding to the physical examination user can be achieved through a questionnaire survey method.

[0087] As an embodiment of the present invention, the method of combining the abnormal physical examination data and the frequent pathology to calculate the pathological incidence rate corresponding to the living behavior includes: performing data cleaning processing on the abnormal physical examination data to obtain target physical examination data, extracting data variables in the target physical examination data, combining the data variables, the target physical examination data and the living behavior to construct a data scatter plot corresponding to the frequent pathology, calculating the slope coefficient corresponding to the data scatter plot, determining the pathological correlation degree corresponding to the living behavior according to the slope coefficient, calculating the relative risk coefficient of each behavior in the living behavior according to the pathological correlation degree, and calculating the pathological incidence rate corresponding to the living behavior according to the relative risk coefficient and the pathological correlation degree.

[0088] Among them, the target physical examination data is the data obtained after removing the duplicate data in the abnormal physical examination data, the data variable is the secondary indicator corresponding to the measurement indicator in the target physical examination data, such as blood pressure or heart rate, the slope coefficient represents the degree of inclination corresponding to the data scatter plot, the pathology correlation represents the relationship between the life behavior and the frequent pathology, and the relative risk coefficient represents the degree of risk difference between each behavior in the life behavior.

[0089] Furthermore, data cleaning processing of the abnormal physical examination data can be achieved through a cleaning tool, which is compiled by a scripting language. Extracting data variables from the target physical examination data can be achieved through an extraction tool, such as a Python tool, with the data variables and the life behavior as coordinate axes, and the target physical examination data as data points. A data scatter plot corresponding to the frequent pathology is constructed using a vi s io mapping tool. Calculating the slope coefficient corresponding to the data scatter plot can be achieved through a slope calculator, which is compiled by a programming language. The coefficient ratio between the pathology correlations can be calculated, and the relative risk coefficient of each behavior in the life behavior can be obtained according to the coefficient ratio.

[0090] Further, as an optional embodiment of the present invention, the calculating the pathological incidence rate corresponding to the living behavior according to the relative risk coefficient and the pathological correlation degree includes:

[0091] The pathological incidence rate corresponding to the lifestyle behavior is calculated by the following formula:

[0092]

[0093] Among them, φ d represents the relative risk coefficient corresponding to the dth behavior in life behavior, H d represents the pathological correlation degree corresponding to the dth behavior in life behavior, d represents the serial number corresponding to the life behavior, 1-Hd Represents the irrelevance coefficient between the dth behavior in life behavior and frequent pathology.

[0094] The correlation coefficient calculation module 104 is used to extract the user physical sign factors corresponding to the physical examination user, input the abnormal physical examination data and the user physical sign factors as input data into a preset regression model, use the feedforward neural network in the regression model to standardize the abnormal physical examination data and the user physical sign factors respectively to obtain standardized data and standardized factors, use the convolutional neural network in the regression model to extract data features corresponding to the standardized data, and use the regression function in the regression model to calculate the correlation coefficient between the data features and the standardized factors.

[0095] The present invention utilizes the feedforward neural network in the regression model to standardize the abnormal physical examination data and the user's physical sign factors respectively, thereby eliminating the difference between the abnormal physical examination data and the user's physical signs, so as to facilitate the subsequent calculation and processing of correlation coefficients, wherein the user's physical signs are the physical representations corresponding to the physical examination user, such as body temperature, weight or age, the regression model is a model for calculating the correlation coefficient between two variables, the feedforward neural network is a neural network for eliminating the difference between two variables, and the standardized data and the standardized factors are the data and factors obtained after eliminating the difference between the abnormal physical examination data and the user's physical signs.

[0096] As an embodiment of the present invention, the feedforward neural network in the regression model is used to standardize the abnormal physical examination data and the user physical sign factors respectively to obtain standardized data and standardized factors, including: using the input layer in the feedforward neural network to scale the abnormal physical examination data and the user physical sign factors respectively to obtain scaled data and scaled physical sign factors, using the linear function in the feedforward neural network to calculate the linear values ​​corresponding to the scaled data and the scaled physical sign factors respectively to obtain data linear values ​​and factor linear values, according to the data linear values, using the standard deviation function in the feedforward neural network to calculate the data standard deviation corresponding to the scaled data, according to the factor linear value, using the standard deviation function to calculate the factor standard deviation corresponding to the factor linear value, according to the data standard deviation and the factor standard deviation, using the standardization function in the regression model to standardize the scaled data and the scaled physical sign factors respectively to obtain standardized data and standardized factors.

[0097] Among them, the input layer is used to scale the input data in the feedforward neural network to reduce the interference between the abnormal physical examination data and the user's physical sign factors. The linear function is a function used to calculate the expression values ​​corresponding to the scaled data and the scaled physical sign factors, such as a linear linear function. The data linear value and the factor linear value are the expression values ​​corresponding to the scaled data and the scaled physical sign factors, respectively. The standard deviation function is compiled by JAVA language, and the standardization function includes a maximum-minimum standardization algorithm. Furthermore, the scaling function in the input layer is used to scale the abnormal physical examination data and the user's physical sign factors, respectively, and the scaling function includes a Min-Max scaling function.

[0098] The present invention calculates the correlation coefficient between the data feature and the standardized factor by utilizing the regression function in the regression model, thereby understanding the correlation between the data feature and the standardized factor, thereby facilitating improving the accuracy of subsequent health management corresponding to the frequent pathology, wherein the data feature corresponding to the standardized data can be extracted by the convolution kernel in the convolutional neural network.

[0099] As an embodiment of the present invention, the calculating the correlation coefficient between the data feature and the standardized factor by using the regression function in the regression model includes:

[0100] The regression function includes:

[0101]

[0102] Among them, Q represents the correlation coefficient between the data feature and the standardized factor, e represents the serial number corresponding to the data feature, e+1 represents the serial number corresponding to the standardized factor, and P e Represents the vector corresponding to the e-th feature in the data features, T e+1 represents the vector corresponding to the e+1th factor in the standardized factor, minmin(ln(P e -T e+1 )) represents the minimum difference between the logarithm of the difference between the vector corresponding to the e-th feature and the vector corresponding to the e+1th factor, maxmax(ln(P e -T e+1 )) represents the maximum difference between the logarithm of the difference between the vector corresponding to the e-th feature and the vector corresponding to the e+1-th factor.

[0103] The program formulation module 105 is used to determine the associated physical signs of the user according to the correlation coefficient, and formulate a health management program corresponding to the frequent pathology in combination with the associated physical signs, the user health level and the pathology incidence rate.

[0104] The present invention formulates a health management plan corresponding to the frequent pathology by combining the associated signs, the user health level and the pathological incidence rate, so as to improve the accuracy of the health management of the physical examination user through the health management plan, wherein the associated signs are signs in the user's physical signs that have an influencing relationship with the frequent pathology. Furthermore, the associated signs in the user's physical signs can be determined by comparing the correlation coefficient with a preset threshold value. If the correlation coefficient is greater than the preset threshold value, it indicates that the corresponding user sign is an associated sign. By analyzing the relationship between the associated signs and the pathological incidence rate, a health management plan corresponding to the frequent pathology can be formulated. For example, smoking can lead to an increased incidence of the frequent pathology, and the amount of smoking can be controlled according to the user's health level. For example, running can reduce the incidence of the frequent pathology, and running exercises can be performed regularly. The amount of exercise is formulated according to the user's health level.

[0105] The present invention analyzes the frequent pathologies corresponding to the physical examination users based on the abnormal physical examination data, and can understand the multiple diseases in the physical examination users, so as to facilitate the relationship between subsequent diseases and related factors. The present invention conducts a health assessment on the physical examination users by combining the detection indicators, the indicator deviation and the frequent pathologies, so as to facilitate the classification of the health of the physical examination users, so as to understand the health level of the physical examination users and facilitate the calculation accuracy of the disease risk coefficient of subsequent life behaviors. The present invention calculates the pathological incidence rate corresponding to the life behaviors by combining the abnormal physical examination data and the frequent pathologies, so as to understand the life behaviors. The probability of occurrence of the corresponding pathology is obtained, thereby improving the accuracy of the analysis of the pathological influencing factors of the physical examination user, and providing a basis for the formulation of subsequent health management plans. The present invention uses the feedforward neural network in the regression model to standardize the abnormal physical examination data and the user's physical signs, respectively, so as to eliminate the difference between the abnormal physical examination data and the user's physical signs, so as to facilitate the subsequent calculation of correlation coefficients. The present invention combines the associated signs and the pathological incidence rate to formulate a health management plan corresponding to the frequent pathology, so as to improve the accuracy of the health management of the physical examination user through the health management plan. Therefore, an Internet-based health examination report data analysis and management system provided by an embodiment of the present invention can improve the accuracy of health examination report data analysis and management.

[0106] Reference Figure 2 FIG. 1 is a flow chart of a method for analyzing and managing health examination report data based on the Internet provided by an embodiment of the present invention. In this embodiment, the method for analyzing and managing health examination report data based on the Internet includes:

[0107] Collecting the health examination data of the user undergoing the physical examination, screening out abnormal physical examination data in the health examination data, and analyzing the frequent pathology corresponding to the user undergoing the physical examination based on the abnormal physical examination data;

[0108] Query the detection index corresponding to the frequent pathology, calculate the index deviation corresponding to the detection index according to the abnormal physical examination data, and conduct a health assessment on the physical examination user by combining the detection index, the index deviation and the frequent pathology to obtain the user's health level;

[0109] Acquire the life behavior corresponding to the physical examination user, and calculate the pathological incidence rate corresponding to the life behavior by combining the abnormal physical examination data and the frequent pathology;

[0110] Extracting user physical sign factors corresponding to the physical examination user, inputting the abnormal physical examination data and the user physical sign factors as input data into a preset regression model, using the feedforward neural network in the regression model to perform standardization processing on the abnormal physical examination data and the user physical sign factors respectively to obtain standardized data and standardized factors, using the convolutional neural network in the regression model to extract data features corresponding to the standardized data, and using the regression function in the regression model to calculate the correlation coefficient between the data features and the standardized factors;

[0111] According to the correlation coefficient, the associated physical signs among the user's physical signs are determined, and a health management plan corresponding to the frequent pathology is formulated in combination with the associated physical signs, the user's health level and the pathology incidence rate.

[0112] In the several embodiments provided by the present invention, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are only illustrative, for example, the division of the modules is only a logical function division, and there may be other division methods in actual implementation.

[0113] In addition, each functional module in each embodiment of the present invention may be integrated into one processing unit, each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional modules.

[0114] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. A health examination report data analysis and management system based on the Internet, characterized in that: The system comprises: A pathology analysis module is used to collect the health examination data of the physical examination user, screen out abnormal physical examination data in the health examination data, and analyze the frequent pathology corresponding to the physical examination user according to the abnormal physical examination data; A data query and calculation module, used to query the detection index corresponding to the frequent pathology, calculate the index deviation corresponding to the detection index according to the abnormal physical examination data, and conduct a health assessment on the physical examination user in combination with the detection index, the index deviation and the frequent pathology to obtain the user's health level; An incidence rate calculation module is used to obtain the life behavior corresponding to the physical examination user, and calculate the pathological incidence rate corresponding to the life behavior by combining the abnormal physical examination data and the frequent pathology; A correlation coefficient calculation module, used to extract the user physical sign factors corresponding to the physical examination user, input the abnormal physical examination data and the user physical sign factors as input data into a preset regression model, use the feedforward neural network in the regression model to standardize the abnormal physical examination data and the user physical sign factors respectively to obtain standardized data and standardized factors, use the convolutional neural network in the regression model to extract data features corresponding to the standardized data, and use the regression function in the regression model to calculate the correlation coefficient between the data features and the standardized factors; A plan formulation module, for determining the associated physical signs of the user according to the correlation coefficient, and formulating a health management plan corresponding to the frequent pathology in combination with the associated physical signs, the user's health level and the pathology incidence rate; Analyzing the frequent pathology corresponding to the user undergoing physical examination according to the abnormal physical examination data includes: Identifying abnormal items corresponding to the abnormal physical examination data, and calculating item similarities between the abnormal items; Classifying the abnormal items according to the item similarities to obtain classified items; Extracting the item detection value corresponding to the classified item from the abnormal physical examination data, and querying the historical medical data corresponding to the physical examination user; Extracting historical medical advice from the historical medical data, and determining the historical pathology of the physical examination user according to the historical medical advice; Determine the current pathology of the user undergoing physical examination by combining the historical pathology and the item detection value; Counting the pathology frequency corresponding to the current pathology, and determining the frequent pathology corresponding to the physical examination user according to the pathology frequency; The calculating the item similarity between the abnormal items includes: The project similarity between the abnormal projects is calculated by the following formula: Among them, A represents the item similarity between abnormal items, B a represents the item vector corresponding to the ath item in the abnormal item, E a+1 represents the item vector corresponding to the a+1th item in the abnormal item, and a and a+1 represent the item sequence numbers corresponding to the abnormal item; The step of combining the detection index, the index deviation and the frequent pathology to perform a health assessment on the physical examination user to obtain the user health level includes: Calculating the contribution between the detection index and the frequent pathology, and assigning an index weight corresponding to the detection index according to the contribution; Calculate the health score of the user undergoing physical examination by combining the indicator deviation and the indicator weight; According to the health score, setting the user health level corresponding to the physical examination user; The step of combining the indicator deviation and the indicator weight to calculate the health score corresponding to the physical examination user includes: The health score corresponding to the physical examination user is calculated by the following formula: Among them, D represents the health score of the physical examination user, b represents the serial number of the detection index corresponding to the index deviation and index weight, t represents the number of detection indicators, β b Indicates the index deviation corresponding to the bth index in the detection index, F b Indicates the indicator weight corresponding to the bth indicator in the detection indicator; The combining of the abnormal physical examination data and the frequent pathology to calculate the pathological incidence rate corresponding to the living behavior includes: Performing data cleaning processing on the abnormal physical examination data to obtain target physical examination data, and extracting data variables in the target physical examination data; Combining the data variables, the target physical examination data and the life behavior, constructing a data scatter plot corresponding to the frequent pathology; Calculating the slope coefficient corresponding to the data scatter plot, and determining the pathological correlation degree corresponding to the life behavior according to the slope coefficient; Calculating the relative risk coefficient of each of the life behaviors according to the pathological correlation degree; Calculating the pathological incidence rate corresponding to the living behavior according to the relative risk coefficient and the pathological correlation degree; Calculating the pathological incidence rate corresponding to the living behavior according to the relative risk coefficient and the pathological correlation degree includes: The pathological incidence rate corresponding to the lifestyle behavior is calculated by the following formula: Among them, φ d represents the relative risk coefficient corresponding to the dth behavior in life behavior, H d represents the pathological correlation degree corresponding to the dth behavior in life behavior, d represents the serial number corresponding to the life behavior, 1-H d represents the irrelevance coefficient between the dth behavior in life behavior and frequent pathology; The calculating the correlation coefficient between the data feature and the standardized factor by using the regression function in the regression model includes: The regression function includes: Among them, Q represents the correlation coefficient between the data feature and the standardized factor, e represents the serial number corresponding to the data feature, e+1 represents the serial number corresponding to the standardized factor, and P e Represents the vector corresponding to the e-th feature in the data features, T e+1 represents the vector corresponding to the e+1th factor in the standardized factor, minmin(ln(P e -T e+1 )) represents the minimum difference between the logarithm of the difference between the vector corresponding to the e-th feature and the vector corresponding to the e+1th factor, maxmax(ln(P e -T e+1 )) represents the maximum difference between the logarithm of the difference between the vector corresponding to the e-th feature and the vector corresponding to the e+1-th factor.

2. The Internet-based health examination report data analysis and management system as claimed in claim 1, characterized in that: The method of using the feedforward neural network in the regression model to perform standardization processing on the abnormal physical examination data and the user's physical sign factors to obtain standardized data and standardized factors includes: Using the input layer in the feedforward neural network to scale the abnormal physical examination data and the user's physical sign factors, respectively, to obtain scaled data and scaled physical sign factors; Utilizing the linear function in the feedforward neural network to respectively calculate the linear values ​​corresponding to the scaled data and the scaled vital sign factors, to obtain the data linear value and the factor linear value; Calculating the data standard deviation corresponding to the scaled data using a standard deviation function in the feedforward neural network according to the data linear value; According to the factor linear value, using the standard deviation function to calculate the factor standard deviation corresponding to the factor linear value; According to the data standard deviation and the factor standard deviation, the scaled data and the scaled physical sign factors are standardized respectively using a standardization function in the regression model to obtain standardized data and standardized factors.

3. A method for analyzing and managing health examination report data based on the Internet, the method realizing the system as claimed in claim 1, characterized in that: The method comprises: Collecting the health examination data of the user undergoing the physical examination, screening out abnormal physical examination data in the health examination data, and analyzing the frequent pathology corresponding to the user undergoing the physical examination based on the abnormal physical examination data; Query the detection index corresponding to the frequent pathology, calculate the index deviation corresponding to the detection index according to the abnormal physical examination data, and conduct a health assessment on the physical examination user by combining the detection index, the index deviation and the frequent pathology to obtain the user's health level; Acquire the life behavior corresponding to the physical examination user, and calculate the pathological incidence rate corresponding to the life behavior by combining the abnormal physical examination data and the frequent pathology; Extracting user physical sign factors corresponding to the physical examination user, inputting the abnormal physical examination data and the user physical sign factors as input data into a preset regression model, using the feedforward neural network in the regression model to perform standardization processing on the abnormal physical examination data and the user physical sign factors respectively to obtain standardized data and standardized factors, using the convolutional neural network in the regression model to extract data features corresponding to the standardized data, and using the regression function in the regression model to calculate the correlation coefficient between the data features and the standardized factors; According to the correlation coefficient, the associated physical signs among the user's physical signs are determined, and a health management plan corresponding to the frequent pathology is formulated in combination with the associated physical signs, the user's health level and the pathology incidence rate.

Images