Optimization method, optimization system and information data processing terminal for health examination items

Abstract

The present invention discloses an optimization method, an optimization system and an information data processing terminal for health checkup items, which belong to the field of medical health technology, and is characterized in that the optimization method for health checkup items comprises the following steps: S1, obtaining basic data of the examinee, and dividing the basic data into different types of type data; S2, selecting a prediction model matching the type data for the content of each type of data; S3, obtaining the prediction result of each type of data through the prediction model; S4, performing a logical or operation on the prediction results of different types of data, and taking the prediction result with the greatest risk as a prediction indicator; S5, adjusting the physical examination items according to the prediction indicators and in combination with the physical examination guide. By adopting the above technical scheme, the present invention performs model pre-analysis on the examinee from three dimensions of questionnaire survey, clinical symptoms and physical examination report, and then formulates physical examination items suitable for the examinee.

Description

Technical Field

[0001] The present invention belongs to the field of medical health technology, and in particular relates to an optimization method, an optimization system and an information data processing terminal for health examination items. Background Art

[0002] As we all know, a health checkup is a physical examination using medical means and methods, including basic examinations in various clinical departments, such as ultrasound, electrocardiogram, radiation and other medical equipment inspections, and laboratory tests of blood, urine and stool around the human body. A health checkup is a physical examination centered on health. Generally, medical experts believe that a health checkup refers to a comprehensive examination of the body before obvious diseases appear in the body. It is convenient to understand the physical condition and screen for physical diseases. That is, a physical examination of healthy people using physical examination means is a "health checkup", or a "preventive health care checkup". A health checkup is used to understand the health status of the examinee, and according to the examination results, it is determined whether there are any abnormal signs, and further analyze the nature of these abnormal signs. Some abnormal signs are physiological variations in themselves, which can be reviewed regularly. Some abnormal signs may be risk factors for diseases, which need to be intervened and corrected through health promotion methods; and some signs are the basis for the diagnosis of diseases, which need further examination and confirmation.

[0003] In recent years, with the continuous improvement of human living standards, health check-ups have received more and more attention from people. In order to achieve comprehensive and accurate health check-up results, the items of health check-ups need to be continuously increased. Frequent health check-ups and repeated health check-ups of multiple items, on the one hand, increase the economic burden of the examinees, and on the other hand, cause a waste of health check-up resources. For this reason, different health check-up institutions have launched a wide variety of health check-up packages, but these health check-up packages are designed for different groups of examinees (such as male or female packages between 20-30 years old, children's packages between 3-10 years old, elderly packages between 50-70 years old, etc.), and do not have personalized functions, that is, the examinees often do a comprehensive shallow health check-up, but do not conduct a deep health check-up based on the individual differences of the examinees. For this reason, it is particularly important to design and develop an optimization method, an optimization system and an information data processing terminal for health check-up items that can optimize the health check-up items according to the individual differences of the examinees. Summary of the invention

[0004] Technical Purpose

[0005] The present invention provides a method for optimizing health checkup items, an optimization system and an information data processing terminal; before the physical examination, a model analysis is performed on the examinee from different dimensions (such as questionnaire survey, clinical symptoms, physical examination report), and then physical examination items suitable for the examinee are formulated.

[0006] Technical Solution

[0007] The first object of the present invention is to provide a method for optimizing health examination items, comprising the following steps:

[0008] S1. Obtain basic data of the examinee and divide the basic data into different types of data;

[0009] S2. For each type of data, select a prediction model that matches the data type;

[0010] S3. Obtain prediction results for each type of data through the prediction model;

[0011] S4. Perform a logical OR operation on the prediction results of different types of data, and use the prediction result with the greatest risk as the prediction indicator;

[0012] S5. Adjust the physical examination items based on the prediction indicators and in combination with the physical examination guidelines.

[0013] Preferably, on the basis of the above technical solution:

[0014] The basic data of this application mainly include: questionnaire survey report, clinical symptoms and historical physical examination report. For this purpose, the said types of data are divided into three parts: questionnaire survey report data, clinical symptom data and physical examination report data.

[0015] According to the specific characteristics of the above different types of data, the following three prediction models are mainly used:

[0016] In S2, when the type of data is questionnaire report data, the prediction model includes a logistic regression model and a Rothman-Keller model.

[0017] In S2, when the type data is clinical symptom data, the prediction model is: Where: Y is the risk level; n is the number of clinical symptoms; A i is the weight factor of the ith symptom.

[0018] In S2, when the type of data is physical examination report data, the prediction model includes:

[0019]

[0020]

[0021] Where: h(t,x) is the risk function; t is the prediction time; x is the prediction variable; h0(t) is the baseline risk; m is the number of items tested; B i is the weight of the i-th reported result; C i Report the result for the ith item.

[0022] In order to quickly and intuitively display the prediction results, the prediction results include different degrees of risk levels, such as low risk, medium risk, and high risk.

[0023] As an example, we continuously improve the accuracy of predictions, including:

[0024] S6. Perform a physical examination using the adjusted physical examination items and obtain the physical examination results;

[0025] S7. Compare the physical examination results with the prediction indicators; and adjust the prediction model and / or model parameters according to the comparison results; specifically:

[0026] When the physical examination result is inconsistent with the prediction index, the basic data of the examinee is extracted and stored separately, and counted. When the count result reaches the set value, the basic data of the examinee stored separately is found for statistical analysis, and the prediction model and / or model parameters are adjusted according to the analysis results;

[0027] When the physical examination results are consistent with the prediction indicators, the existing prediction model is saved.

[0028] The second object of the present invention is to provide a health check-up item optimization system, comprising:

[0029] The data acquisition module acquires the basic data of the examinee and divides the basic data into different types of data;

[0030] The model matching module selects a prediction model that matches the data of each type according to the content of the data;

[0031] The preliminary result prediction module obtains the prediction results of each type of data through the prediction model;

[0032] The secondary result operation module performs logical OR operation on the prediction results of different types of data and takes the prediction result with the greatest risk as the prediction indicator;

[0033] The optimization module adjusts the physical examination items based on the prediction indicators and combined with the physical examination guidelines.

[0034] Preferably, on the basis of the above technical solution:

[0035] The basic data of this application mainly include: questionnaire survey report, clinical symptoms and historical physical examination report. For this purpose, the data of this type is divided into three parts: questionnaire survey report data, clinical symptom data and physical examination report data. In order to obtain the three parts of data, the following three signal acquisition modules are used:

[0036] Human-computer interaction module for obtaining questionnaire report data, such as APP;

[0037] An inquiry module for obtaining clinical symptom data; such as a recording module and a text conversion module; or a computer input module directly based on the inquiry results;

[0038] Recognition module for obtaining physical examination report data.

[0039] According to the specific characteristics of the above different types of data, the data acquisition module mainly adopts the following three prediction models:

[0040] The first prediction model receives the questionnaire survey report data and uses the logistic regression model and Rothman-Keller model to analyze and process the questionnaire survey report data;

[0041] The second prediction model receives clinical symptom data and analyzes and processes the clinical symptom data, specifically:

[0042] Where: Y is the risk level; n is the number of clinical symptoms; A i For the ith symptom

[0043] Weight factor;

[0044] The third prediction model receives the physical examination report data and analyzes and processes the physical examination report data, specifically:

[0045]

[0046]

[0047] Where: h(t,x) is the risk function; t is the prediction time; x is the prediction variable; h0(t) is the baseline risk; m is the number of items tested; B i is the weight of the i-th reported result; C i Report the result for the ith item.

[0048] In order to continuously improve the prediction accuracy, the optimization module includes submodules:

[0049] The physical examination results of the adjusted physical examination items are compared with the prediction indicators; the prediction model and / or model parameters are adjusted according to the comparison results; specifically:

[0050] When the physical examination result is inconsistent with the prediction index, the basic data of the examinee is extracted and stored separately, and counted. When the count result reaches the set value, the basic data of the examinee stored separately is found for statistical analysis, and the prediction model and / or model parameters are adjusted according to the analysis results;

[0051] When the physical examination results are consistent with the prediction indicators, the existing prediction model is saved.

[0052] In order to achieve fast and convenient information entry: the identification module includes:

[0053] AI recognition module, used to obtain image information of paper reports, extract test results from the image information, and convert the test results into digital information;

[0054] The data interaction module extracts the results of the physical examination report from the Internet of Things or data storage devices.

[0055] The third invention objective of this patent is to provide an information data processing terminal that implements the optimization method of the above-mentioned health check-up items.

[0056] The fourth invention objective of this patent is to provide a computer-readable storage medium, including instructions, which, when run on a computer, enables the computer to execute the above-mentioned method for optimizing health examination items.

[0057] The advantages and positive effects of the present invention are:

[0058] The present invention firstly performs model analysis on the examinee from different dimensions (such as questionnaire survey, clinical symptoms, and physical examination report), and then performs logical OR operation on the analysis results of different models, so as to ensure the comprehensiveness of risk screening and avoid risk omission; finally, according to the risk prediction level and the requirements of the health guide, a physical examination package suitable for each examinee is formulated;

[0059] The present invention analyzes and statistically compares the physical examination results and the prediction results, and then continuously optimizes the prediction model, so that as the data increases, the accuracy of the prediction model is continuously improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 is a flow chart of a first preferred embodiment of the present invention;

[0061] Figure 2 is a structural diagram of type data in the first preferred embodiment of the present invention;

[0062] Figure 3 is a system block diagram of a first preferred embodiment of the present invention;

[0063] Figure 4 is a flow chart of a second preferred embodiment of the present invention;

[0064] Figure 5 It is a structural block diagram of the identification module in the preferred embodiment of the present invention. DETAILED DESCRIPTION

[0065] In order to further understand the content, features and effects of the present invention, the following embodiments are given as examples and described in detail with reference to the accompanying drawings.

[0066] First preferred embodiment: Please refer to Figure 1 , an optimization method for health examination items, including the following steps:

[0067] S1. Obtain the basic data of the examinee from different dimensions and divide the basic data into different types of data. In general, the basic data mainly include: questionnaire survey reports, clinical symptoms and historical physical examination reports, among which: the content of the questionnaire survey report mainly includes living habits, family disease history, personal disease history, etc.; the content of clinical symptoms mainly includes data on self-feeling; the content of historical physical examination reports mainly includes the most recent one or two test reports, etc.;

[0068] like Figure 2 As shown: This preferred embodiment divides the type of data into three parts: questionnaire survey report data, clinical symptom data and physical examination report data. In actual situations, the questionnaire survey report data and clinical symptom data are both non-quantitative forms, and the two can be combined into one or separated independently;

[0069] S2. According to the content of each type of data, a prediction model matching the type of data is selected; according to the specific characteristics of the above different types of data, this application mainly adopts the following three prediction models:

[0070] When the type of data is questionnaire survey report data, the prediction model includes logistic regression model and Rothman-Keller model. For example:

[0071] The logistic regression model for male lung cancer is:

[0072] logitP=-6.409+1.632*X1+0.773*X2+0.362*X3-0.842*X4

[0073] -1.301*X5+1.811*X6+0.141*X7+1.519*X8+1.215*X9

[0074] -1.198*X10+1.248*X11+1.668*X12+0.396*X13+1.182*X14

[0075] Among them, logitP is the prediction result, X1 to X14 are the specific items of the questionnaire survey report data, here there are 14 items, and in practice the items can be increased or decreased as needed;

[0076] Rothman-Keller Model of Lung Cancer in Men:

[0077] Individual risk of disease = [(3.685-1)*[smoking]+(0.206-1)*[passive smoking]+(2.191-1)*[drinking]+(2.731-1)*[family history of cancer]+(4.014-1)*[history of tuberculosis]+(2.669-1)*[history of bronchial asthma]+(2.211-1)*[thinness]+(2.751-1)*[50-60 years old]+(3.911-1)*[ >=60 years old]+(2.173-1)*[Has diabetes]+[I am blocked text 1.444]×[I am blocked text 1.795]×[I am blocked text 1.845]×[I am blocked text 1.787]]×male lung cancer incidence rate (97.09 / 100,000); where: "I am blocked text" is personal privacy content; if there are multiple places of personal privacy content, they are all represented by "I am blocked text", which does not mean that the content of each place is consistent;

[0078] The logistic regression model for female lung cancer is:

[0079] logitP=-6.615+1.588*X1+1.673*X2+1.262*X3-1.942*X4

[0080] -1.301*X5+1.811*X6+0.141*X7+1.519*X8+1.215*X9

[0081] -1.198*X10+1.248*X11+1.668*X12+0.723*X13+1.182*X14

[0082] The Rothman-Keller model for lung cancer in women is:

[0083] Individual risk of disease = [(2.783-1)*[smoking]+(2.206-1)*[passive smoking]+(2.191-1)*[drinking]+(2.731-1)*[family history of tumor]+(2.014-1)*[history of tuberculosis]+(0.669-1)*[history of bronchial asthma]+(0.211-1)*[thinness]+(5.735-1)*[50-60 years old]+(2.414-1)*[>=60 years old]+(2.735-1)*[diabetes]+[I am blocked text 1.444]×[I am blocked text 0.895]×[I am blocked text 0.645]×[I am blocked text 0.987]]× female lung cancer incidence rate (90.20 / 100,000) Among them: "I am blocked text" is personal privacy content;

[0084] Among them: X1 is smoking, X2 is passive smoking, X3 is drinking, X4 is vegetable intake, X5 is regular exercise, X6 is a family history of cancer, X7 is a history of tuberculosis, X8 is a history of bronchial asthma, X9 is thin, X10 is overweight, X11 is I am blocked text, X12 is I am blocked text, X13 is I am blocked text, X14 is I am blocked text;

[0085] When the type data is clinical symptom data, the prediction model is: Where: Y is the risk level; n is the number of clinical symptoms; A i is the weight factor of the ith symptom. For example, the prediction of lung cancer;

[0086] Y=9 blood in sputum + I am blocking text + history of occupational exposure (asbestos, beryllium, uranium, radon, etc.) + family history of lung cancer + chronic obstructive pulmonary disease + history of diffuse pulmonary fibrosis + lung nodules have been detected;

[0087] A Y value greater than or equal to 1 indicates medium-high risk. A Y value greater than or equal to 3 indicates high risk.

[0088] When the type of data is physical examination report data, the prediction model includes:

[0089]

[0090]

[0091] Where: h(t,x) is the risk function; t is the prediction time; x is the prediction variable, this item is the inspection item; h0(t) is the baseline risk; m is the number of items to be tested; B i is the weight of the i-th reported result; C i Report the result for item i; for example, the following prediction of lung cancer based on the physical examination report:

[0092]

[0093]

[0094] S3. Import different types of data into corresponding prediction models, and obtain prediction results for each type of data through the prediction model; in order to quickly and intuitively display the prediction results, the prediction results include different degrees of risk levels, such as low risk, medium risk, and high risk.

[0095] S4, perform a logical OR operation on the prediction results of different types of data, and use the prediction result with the greatest risk as the prediction indicator; for example, if the prediction results of S3 include low risk and medium risk, the final prediction result is medium risk;

[0096] S5. Adjust the physical examination items according to the prediction indicators and in combination with the physical examination guidelines. For example, if the person being examined is predicted to have a high risk of lung cancer, it is necessary to add lung cancer-related physical examination items in a targeted manner based on the physical examination requirements for lung cancer patients in the physical examination guidelines on the basis of the original basic physical examination items;

[0097] Second preferred embodiment: Please refer to Figure 3 , an optimization system for health examination items, comprising:

[0098] The data acquisition module obtains the basic data of the examinee and divides the basic data into different types of data. In general, the basic data mainly includes: questionnaire survey reports, clinical symptoms and historical physical examination reports, among which: the content of the questionnaire survey report mainly includes living habits, family disease history, personal disease history, etc.; the content of clinical symptoms mainly includes data on self-feeling; the content of the historical physical examination report mainly includes the most recent one or two test reports, etc.;

[0099] like Figure 2 As shown: This preferred embodiment divides the type of data into three parts: questionnaire survey report data, clinical symptom data and physical examination report data. In actual situations, the questionnaire survey report data and clinical symptom data are both non-quantitative forms, and the two can be combined into one or separated independently;

[0100] The model matching module selects a prediction model that matches the content of each type of data. According to the specific characteristics of the above different types of data, this application mainly adopts the following three prediction models:

[0101] When the type of data is questionnaire survey report data, the prediction model includes logistic regression model and Rothman-Keller model. For example:

[0102] The logistic regression model for male lung cancer is:

[0103] logitP=-6.409+1.632*X1+0.773*X2+0.362*X3-0.842*X4

[0104] -1.301*X5+1.811*X6+0.141*X7+1.519*X8+1.215*X9

[0105] -1.198*X10+1.248*X11+1.668*X12+0.396*X13+1.182*X14

[0106] Among them, logitP is the prediction result, X1 to X14 are the specific items of the questionnaire survey report data, here there are 14 items, and in practice the items can be increased or decreased as needed;

[0107] Rothman-Keller Model of Lung Cancer in Men:

[0108] Individual risk of disease = [(3.685-1)*[smoking]+(0.206-1)*[passive smoking]+(2.191-1)*[drinking]+(2.731-1)*[family history of cancer]+(4.014-1)*[history of tuberculosis]+(2.669-1)*[history of bronchial asthma]+(2.211-1)*[thinness]+(2.751-1)*[50-60 years old]+(3.911-1)*[ >=60 years old]+(2.173-1)*[Has diabetes]+[I am blocked text 1.444]×[I am blocked text 1.795]×[I am blocked text 1.845]×[I am blocked text 1.787]]×male lung cancer incidence rate (97.09 / 100,000); where: "I am blocked text" is personal privacy content; if there are multiple places of personal privacy content, they are all represented by "I am blocked text", which does not mean that the content of each place is consistent;

[0109] The logistic regression model for female lung cancer is:

[0110] logitP=-6.615+1.588*X1+1.673*X2+1.262*X3-1.942*X4

[0111] -1.301*X5+1.811*X6+0.141*X7+1.519*X8+1.215*X9

[0112] -1.198*X10+1.248*X11+1.668*X12+0.723*X13+1.182*X14

[0113] The Rothman-Keller model for lung cancer in women is:

[0114] Individual risk of disease = [(2.783-1)*[smoking]+(2.206-1)*[passive smoking]+(2.191-1)*[drinking]+(2.731-1)*[family history of tumor]+(2.014-1)*[history of tuberculosis]+(0.669-1)*[history of bronchial asthma]+(0.211-1)*[thinness]+(5.735-1)*[50-60 years old]+(2.414-1)*[>=60 years old]+(2.735-1)*[diabetes]+[I am blocked text 1.444]×[I am blocked text 0.895]×[I am blocked text 0.645]×[I am blocked text 0.987]]× female lung cancer incidence rate (90.20 / 100,000) Among them: "I am blocked text" is personal privacy content;

[0115] Among them: X1 is smoking, X2 is passive smoking, X3 is drinking, X4 is vegetable intake, X5 is regular exercise, X6 is a family history of cancer, X7 is a history of tuberculosis, X8 is a history of bronchial asthma, X9 is thin, X10 is overweight, X11 is I am blocked text, X12 is I am blocked text, X13 is I am blocked text, X14 is I am blocked text;

[0116] When the type data is clinical symptom data, the prediction model is: Where: Y is the risk level; n is the number of clinical symptoms; A i is the weight factor of the ith symptom. For example, the prediction of lung cancer;

[0117] Y=9 Blood in sputum + I am blocking text + History of occupational exposure (asbestos, beryllium, uranium, radon, etc.) + Family history of lung cancer + Chronic obstructive pulmonary disease + History of diffuse pulmonary fibrosis + Pulmonary nodules have been detected;

[0118] A Y value greater than or equal to 1 indicates medium-high risk. A Y value greater than or equal to 3 indicates high risk.

[0119] When the type of data is physical examination report data, the prediction model includes:

[0120]

[0121]

[0122] Where: h(t,x) is the risk function; t is the prediction time; x is the prediction variable, this item is the inspection item; h0(t) is the baseline risk; m is the number of items to be tested; B i is the weight of the i-th reported result; C i Report the result for item i; for example, the following prediction of lung cancer based on the physical examination report:

[0123]

[0124]

[0125] The preliminary result prediction module obtains the prediction results of each type of data through the prediction model;

[0126] The secondary result operation module performs logical OR operation on the prediction results of different types of data and takes the prediction result with the greatest risk as the prediction indicator;

[0127] The optimization module adjusts the physical examination items based on the prediction indicators and combined with the physical examination guidelines.

[0128] Among them: human-computer interaction module for obtaining questionnaire survey report data; such as APP;

[0129] An inquiry module for obtaining clinical symptom data; such as a recording module and a text conversion module; or a computer input module directly based on the inquiry results;

[0130] Recognition module for obtaining physical examination report data.

[0131] In order to continuously improve the accuracy of prediction, based on the second preferred embodiment, the optimization module includes submodules:

[0132] The physical examination results of the adjusted physical examination items are compared with the prediction indicators; the prediction model and / or model parameters are adjusted according to the comparison results; specifically:

[0133] When the physical examination result is inconsistent with the prediction index, the basic data of the examinee is extracted and stored separately, and counted. When the count result reaches the set value, the basic data of the examinee stored separately is found for statistical analysis, and the prediction model and / or model parameters are adjusted according to the analysis results;

[0134] When the physical examination results are consistent with the prediction indicators, the existing prediction model is saved.

[0135] For quick and easy information entry, please refer to Figure 5 , the identification module comprises:

[0136] AI recognition module, used to obtain image information of paper reports, extract test results from the image information, and convert the test results into digital information;

[0137] The data interaction module extracts the results of the physical examination report from the Internet of Things or data storage devices.

[0138] See also Figure 4 , based on the first preferred embodiment, the accuracy of the prediction model is continuously improved; further comprising:

[0139] S6. Perform a physical examination using the adjusted physical examination items and obtain the physical examination results;

[0140] S7. Compare the physical examination results with the prediction indicators; and adjust the prediction model and / or model parameters according to the comparison results; specifically:

[0141] When the physical examination results are inconsistent with the prediction indicators, the basic data of the examinee is extracted and saved separately, and counted. When the counting result reaches the set value, such as 500, the basic data of the examinee saved separately is found for statistical analysis, and the prediction model and / or model parameters are adjusted according to the analysis results; for example, if the similarity of eating habits is high or the regional overlap rate is high, the weight factor of the high similarity of eating habits or the high regional overlap rate is increased.

[0142] When the physical examination results are consistent with the prediction indicators, the existing prediction model is saved.

[0143] An information data processing terminal for realizing the above-mentioned optimization method for health examination items.

[0144] A computer-readable storage medium includes instructions, which, when executed on a computer, enable the computer to execute the above-mentioned method for optimizing health examination items.

[0145] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When the use is implemented in whole or in part in the form of a computer program product, the computer program product includes one or more computer instructions. When the computer program instructions are loaded or executed on a computer, the process or function described in the embodiment of the present invention is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from one website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL) or wireless (e.g., infrared, wireless, microwave, etc.) mode) to another website site, computer, server or data center. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrated. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state hard disk SolidState Disk (SSD)), etc.

[0146] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Any simple modification, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention are within the scope of the technical solution of the present invention.

Claims

1. A method for optimizing health examination items, characterized in that: Model analysis is performed on the examinee from three dimensions: questionnaire survey, clinical symptoms, and physical examination report, and then physical examination items suitable for the examinee are formulated; the optimization method includes the following steps: S1. Obtain the basic data of the examinee, and divide the basic data into different types of data; the types of data include questionnaire survey report data, clinical symptom data and physical examination report data. The questionnaire survey report data includes living habits, family disease history, and personal disease history. The clinical symptom data includes the data of the examinee's own feelings. The historical physical examination report includes the most recent or the most recent two test reports. S2. For each type of data, select a prediction model that matches the data type; specifically: When the type data is questionnaire report data, the prediction model includes a logistic regression model corresponding to the gender of the physical examinee and a Rothman-Keller model corresponding to the gender of the physical examinee; When the type data is clinical symptom data, the prediction model is: Where: Y is the risk level; n is the number of clinical symptoms; A i is the weight factor of the ith symptom; When the type of data is physical examination report data, the prediction model includes: Where: h(t,x) is the risk function; t is the prediction time; x is the prediction variable; h0(t) is the baseline risk; m is the number of items tested; B i is the weight of the i-th reported result; C i Report the result for item i; S3. Obtain prediction results for each type of data through the prediction model, wherein the multiple prediction results include risk levels of different degrees, the risk level is used to indicate the risk level of the examinee suffering from the target disease, and the multiple risk levels include low risk, medium risk, and high risk; S4, performing a "logical OR" operation on the prediction results of different types of data, and taking the prediction result with the highest risk level among the multiple prediction results as the prediction indicator corresponding to the target disease; S5. According to the risk level of the target disease indicated by the prediction indicator and in combination with the physical examination requirements for patients with the target disease in the physical examination guide, the physical examination items of the examinee are adjusted in terms of the physical examination items related to the target disease.

2. The method for optimizing health checkup items according to claim 1, characterized in that: Also includes: S6. Perform a physical examination using the adjusted physical examination items and obtain the physical examination results; S7. Compare the physical examination results with the prediction indicators; and adjust the prediction model and / or model parameters according to the comparison results; specifically: When the physical examination result is inconsistent with the prediction index, the basic data of the examinee is extracted and stored separately, and counted. When the count result reaches the set value, the basic data of the examinee stored separately is found for statistical analysis, and the prediction model and / or model parameters are adjusted according to the analysis results; When the physical examination results are consistent with the prediction indicators, the existing prediction model is saved.

3. A health check-up item optimization system, characterized in that: Model analysis is performed on the examinee from three dimensions: questionnaire survey, clinical symptoms, and physical examination report, and then physical examination items suitable for the examinee are formulated; the optimization system includes: A data acquisition module is used to acquire basic data of the examinee and divide the basic data into different types of data; the data acquisition module includes: a human-computer interaction module for acquiring questionnaire survey report data; an inquiry module for acquiring clinical symptom data; and an identification module for acquiring physical examination report data; the questionnaire survey report data includes living habits, family disease history, and personal disease history; the clinical symptom data includes data on the examinee's own feelings; and the historical physical examination report includes the most recent or the most recent two test reports; The model matching module selects a prediction model that matches the data of each type according to the content of the data; the prediction model includes: A first prediction model receives the questionnaire survey report data, and analyzes and processes the questionnaire survey report data to form a logistic regression model corresponding to the gender of the physical examinee, and a Rothman-Keller model corresponding to the gender of the physical examinee; The second prediction model receives clinical symptom data and analyzes and processes the clinical symptom data, specifically: Where: Y is the risk level; n is the number of clinical symptoms; A i is the weight factor of the ith symptom; The third prediction model receives the physical examination report data and analyzes and processes the physical examination report data, specifically: Where: h(t,x) is the risk function; t is the prediction time; x is the prediction variable; h0(t) is the baseline risk; m is the number of items tested; B i is the weight of the i-th reported result; C i Report the result for item i; A preliminary result prediction module, which obtains prediction results for each type of data through a prediction model, wherein the multiple prediction results include risk levels of different degrees, the risk level is used to indicate the risk level of the examinee suffering from the target disease, and the multiple risk levels include low risk, medium risk, and high risk; The secondary result calculation module performs a "logical OR" operation on the prediction results of different types of data, and uses the prediction result with the highest risk level among the multiple prediction results as the prediction indicator corresponding to the target disease; The optimization module adjusts the physical examination items of the examinee related to the target disease according to the risk level of the target disease indicated by the prediction index and in combination with the physical examination requirements for patients with the target disease in the physical examination guide.

4. The health check-up item optimization system according to claim 3, characterized in that: The optimization module includes submodules: The physical examination results of the adjusted physical examination items are compared with the prediction indicators; the prediction model and / or model parameters are adjusted according to the comparison results; specifically: When the physical examination result is inconsistent with the prediction index, the basic data of the examinee is extracted and stored separately, and counted. When the count result reaches the set value, the basic data of the examinee stored separately is found for statistical analysis, and the prediction model and / or model parameters are adjusted according to the analysis results; When the physical examination results are consistent with the prediction indicators, the existing prediction model is saved.

5. The health check-up item optimization system according to claim 3, characterized in that: The identification module comprises: AI recognition module, used to obtain image information of paper reports, extract test results from the image information, and convert the test results into digital information; The data interaction module extracts the results of the physical examination report from the Internet of Things or data storage devices.

6. An information data processing terminal for implementing the method for optimizing health checkup items as described in any one of claims 1 to 2.

7. A computer-readable storage medium comprising instructions, which, when executed on a computer, enables the computer to execute the method for optimizing health examination items as claimed in any one of claims 1 to 2.

Images