Multi-dimensional index health management method and system based on physical examination report

Abstract

The invention is suitable for the field of artificial intelligence and health management, and provides a multi-dimensional index health management method and system based on a physical examination report, and the method comprises the steps: obtaining the basic information and physical examination data of a subject based on the physical examination report, and carrying out the preprocessing of the obtained data; inputting the preprocessed data into a pre-trained health assessment model, assessing health risks of the seven systems and outputting risk levels, and performing grading and labeling according to the risk levels; identifying key indexes according to a grading result, and mining an association relationship among the key indexes to determine and early warn potential risk indexes; and according to the grading result and the early-warning potential risk index, outputting a doctor-seeing guidance scheme and a supplementary checking suggestion for the subject. By means of multi-index multi-weight risk interpretation, quantitative tracking, personalized guidance, effect tracking and the like, the problem that single-index interpretation in an existing physical examination report is insufficient in health management is solved, and one-stop and accurate service is provided for examinees.

Description

Technical Field

[0001] The present invention belongs to the field of artificial intelligence and health management, and in particular, relates to a multi-dimensional indicator health management method and system based on physical examination reports. Background Art

[0002] With the continuous development of medical technology and the improvement of people's health awareness, health examinations have become an important means of preventing diseases and maintaining health. At present, health examination technology is quite mature. Each health examination center will provide users with various abnormal indicator interpretation reports. These health examination reports usually include a general inspection report and the conclusions of each inspection item. They are usually very professional medical descriptions. It is difficult for patients to understand the severity of the problem and the next action to be taken. However, due to the differences in the health examination systems of different health examination institutions, different health examination structures have their own health examination systems and reports. To compare the results of previous health examinations, you can only rely on purchasing professional offline report interpretation services.

[0003] There are data silos between different physical examination institutions. Patients' reports from different physical examination institutions cannot be presented to users in a unified and intuitive manner. There is also a lack of clear action suggestions for the next steps after the physical examination and a lack of intuitive expression of the severity of the problem. Therefore, improvement is necessary. Summary of the invention

[0004] The purpose of the present invention is to provide a multi-dimensional indicator health management method based on physical examination reports, aiming to solve at least one of the problems mentioned in the background technology; the present invention uses a digitized and quantified approach, combined with individual differences of users, to achieve in-depth interpretation and continuous analysis of physical examination reports, and provide personalized health guidance and management suggestions.

[0005] The present invention is implemented as follows: a multi-dimensional indicator health management method based on physical examination reports, the method comprising: Obtain the examinee's basic information and physical examination data based on the physical examination report, and pre-process the acquired data; The pre-processed data is input into the pre-trained health assessment model to assess the health risks of the seven major systems and output the risk level, and then grade and label them according to the risk level; the seven major systems include: cardiovascular system, endocrine system, respiratory system, digestive system, urinary system, reproductive system, and skeletal system; Identify key indicators based on the classification results and explore the correlation between key indicators to determine and warn of potential risk indicators; Based on the grading results and the potential risk indicators warned, the patient's treatment guidance plan and supplementary examination suggestions are output.

[0006] Furthermore, the method further comprises: In the step of grading and marking according to risk levels, marking is performed by a three-color light method or a highlight method.

[0007] Furthermore, the method also includes: combining the acquired physical examination data of the examinee with historical physical examination data to generate a data comparison trend chart, and displaying it in a graphical manner.

[0008] Furthermore, the basic information of the examinee includes at least age, gender, and family medical history records; the physical examination data of the examinee includes at least biochemical indicators and imaging test data.

[0009] Furthermore, in the step of identifying key indicators based on the grading results and mining the correlation between the key indicators, the correlation between the key indicators is mined by calling a deep feature mining module; wherein the deep feature mining module includes at least two convolutional layers, two pooling layers and one fully connected layer.

[0010] The second object of the present invention is to provide a multi-dimensional indicator health management system based on physical examination reports, which is used in the above method, and the system includes: The data acquisition module is used to obtain the basic information and physical examination data of the examinee and pre-process the acquired data; The risk assessment and grading module is used to input the pre-processed data into the pre-trained health assessment model, assess the health risks of the seven major systems and output the risk level, and grade and label them according to the risk level; Potential risk warning module, which is used to identify key indicators based on the classification results and explore the correlation between key indicators to determine and warn potential risk indicators; The result output and suggestion module is used to output the treatment guidance plan and supplementary examination suggestions for the examinee based on the grading results and potential risk indicators warned.

[0011] Furthermore, the system also includes: a marking module, which is used to mark by a three-color light method or a highlight method in the step of grading and marking according to the risk level.

[0012] Furthermore, the system also includes: a graphical module, which is used to combine the acquired physical examination data of the examinee with historical physical examination data to generate a data comparison trend chart and display it in a graphical manner.

[0013] The present invention provides a multi-dimensional indicator health management method based on physical examination reports. This method displays health risks in an intuitive and easy-to-understand way, enhances the examinee's awareness of their own health status, and thus improves their enthusiasm and initiative in participating in health management; this enhanced health awareness helps the examinee pay more attention to their own health and take active preventive and treatment measures; the pre-trained health assessment model analyzes multiple indicators, can accurately identify the examinee's health risks in different systems, and tailor a personalized health management plan for them; it can discover potential risks under "normal" indicators and provide early warning and intervention suggestions for the examinee. This forward-looking health management strategy helps to intervene when the disease has not yet formed or is in the early stages, thereby effectively curbing the development of the disease and improving treatment effects and survival rates. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 A schematic diagram of the principle of a multi-dimensional indicator health management method based on a physical examination report provided by an embodiment of the present invention; Figure 2 A flowchart of a multi-dimensional indicator health management method based on a physical examination report provided by an embodiment of the present invention; Figure 3 This is a schematic diagram of a first sub-process of an embodiment of the present invention; Figure 4 This is a schematic diagram of a second sub-process of an embodiment of the present invention; Figure 5 An internal structural block diagram of a multi-dimensional indicator health management system based on physical examination reports provided by an embodiment of the present invention; Figure 6 Another internal structural block diagram of a multi-dimensional indicator health management system based on physical examination reports provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0015] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0016] The specific implementation of the present invention is described in detail below in conjunction with specific embodiments.

[0017] Figure 1 A schematic diagram of the principle of a multi-dimensional indicator health management method based on a physical examination report provided by an embodiment of the present invention; Figure 2 A flowchart of a multi-dimensional indicator health management method based on a physical examination report provided by an embodiment of the present invention. In one embodiment, a multi-dimensional indicator health management method based on a physical examination report includes the following steps: S101: Obtaining basic information and physical examination data of the examinee based on the physical examination report, and preprocessing the acquired data; In this step, the physical examination report may be one or more copies; the one or more physical examination reports may be digitized through OCR technology, thereby obtaining the basic information and physical examination data of the examinee.

[0018] Among them, preprocessing includes data cleaning, deduplication, and normalization, etc.

[0019] S102: Input the pre-processed data into a pre-trained health assessment model to assess the health risks of the seven systems and output the risk levels, and classify and label them according to the risk levels; the seven systems include: cardiovascular system, endocrine system, respiratory system, digestive system, urinary system, reproductive system, and skeletal system; Among them, the health assessment model can convolve and pool the data related to the seven major systems respectively, and then fuse and output the data; Exemplary, evaluation of the health assessment model: Taking the determination of cardiovascular and cerebrovascular system risk level as an example: the subject's disease risk level = age base * age weight + gender weight + obesity factor * obesity weight + previous medical history weight + key indicator quantitative weight, calculate together to get the final risk value; refer to Figure 1 As shown; The core function of this weighted algorithm analysis is to combine the subject's age, gender, underlying diseases and other information to perform weighted algorithm analysis on seven major systems, including the cardiovascular system, respiratory system, and digestive system, to achieve health risk grading.

[0020] S103: Identify key indicators based on the classification results, and explore the correlation between key indicators to determine and warn of potential risk indicators; For example, the key indicators of the cardiovascular system may include blood lipids, blood pressure, etc.; the key indicators are identified based on the classification results, specifically more than 200 key indicators; When exploring the correlation between key indicators, the complex relationships between more than 200 key indicators are integrated and analyzed to discover potential risks under "normal" indicators; avoiding hidden risks that were ignored in the analysis of single indicators and single diseases in previous physical examination reports.

[0021] For example, in the evaluation of the cardiovascular system, taking low-density lipoprotein as an example, the medical definition of its normal range value should be less than or equal to 3.4mmol / L, and this standard is based on the fact that all relevant health indicators are maintained at an ideal level. However, in actual situations, if a subject also suffers from hypertension, the "normal" threshold of his low-density lipoprotein is not necessarily 3.4mmol / L. It is necessary to comprehensively consider the overall condition of his blood pressure and blood lipids and set a more stringent range that is lower than the conventional standard. The isolated analysis of a single indicator often generally includes patients with hypertension and hyperlipidemia in the "safe" category of less than or equal to 3.4mmol / L, thereby ignoring the real risks hidden under the illusion of health. This method is to make up for this deficiency. By deeply analyzing the interactions and associations between multiple health indicators, an accurate health range is tailored for each individual; it keenly captures those health risks that seem "normal" under the traditional single indicator evaluation, but are actually hidden, and realizes the early detection, early screening, and early intervention of hidden risks.

[0022] S104: Output the medical guidance plan and supplementary examination suggestions for the examinee based on the classification results and the potential risk indicators warned; For example, the medical guidance plan for the examinee is customized according to the risk assessment results; Figure 1 In the process, the physical examination report of the examinee can be an electronic report or a paper report, both of which can be converted into a PDF report file. In the computer device, this type of PDF report file is uploaded, and then the indicator data in the PDF report file can be extracted through text scanning technology, such as OCR (Optical Character Recognition) technology; the project indicator template is matched according to the indicator data; and then the analysis and decision-making module is called to convert the indicator data into variables, and the disease risk level is calculated according to the variables.

[0023] Exemplarily, the calculation of the disease risk level can be considered from two aspects, namely, calculating the basic risk factor value and calculating the risk factor value (or key indicator factor value) that may be caused by the key indicator; then, calculating the evaluation result value according to the basic risk factor value and the key indicator factor value, thereby judging the risk factor conditions; and outputting the risk level suggestion and adding the indicator term explanation according to the judgment situation, and finally ending.

[0024] For example, the calculation of basic risk factor values ​​mainly considers age, gender, body mass index, family history or past medical history, etc.; the calculation of key indicator factor values ​​mainly considers key indicators such as four blood lipids, blood pressure, and blood sugar.

[0025] In an example of this embodiment, the medical consultation guidance plan includes detailed information such as the medical department, time of medical consultation, reason for medical consultation, etc. The medical consultation guidance plan can clearly indicate to the examinee which specific system is at high risk, which department to go to for medical consultation, which abnormal items need further examination, when to go for medical consultation, and approximately how long it will take to do a re-examination.

[0026] In an example of this embodiment, supplementary examination suggestions: based on the physical examination data and possible data missing, suggestions for supplementary examination items are made to help the examinee complete the examination items and ensure the comprehensiveness of the health assessment.

[0027] In one example of this embodiment, Figure 3 As shown, the method also includes: S201: In the step of grading and marking according to risk levels, marking is performed by a three-color light method or a highlight method.

[0028] The three-color light method, for example: "three-color light" marking: in the data generated by the physical examination report, the "three-color light" (green, yellow, and red) is used to intuitively mark the health risk level of each system, which is convenient for the examinee or user to quickly understand; the highlight method is basically the same as the three-color light method.

[0029] In one example of this embodiment, Figure 4 As shown, the method further includes: combining the acquired physical examination data of the examinee with historical physical examination data to generate a data comparison trend chart, and displaying it in a graphical manner; For example, for continuously uploaded physical examination reports, the changing trends of various system indicators can be automatically compared to generate an intuitive data comparison trend chart for the examinee to view and refer to.

[0030] Exemplarily, the physical examination data of the examinee includes at least biochemical indicators and imaging test data.

[0031] In an example of this embodiment, in the step of identifying key indicators based on grading results and mining the correlation between key indicators, the correlation between key indicators is mined by calling a deep feature mining module; wherein the deep feature mining module includes at least two convolutional layers, two pooling layers and one fully connected layer.

[0032] In this example, the deep feature mining module is a deep learning model that deeply mines the correlation between key indicators, identifies and warns of indicators that are within the normal range but have potential risks, and provides early intervention recommendations for the examinees.

[0033] As mentioned above, a multi-dimensional indicator health management method based on physical examination reports can improve health intention and participation: by displaying health risks in an intuitive and easy-to-understand way, the examinees' awareness of their own health status is enhanced, thereby increasing their enthusiasm and initiative in participating in health management; this enhanced health awareness helps the examinees pay more attention to their own health and take active preventive and treatment measures. Achieve precise health management: by conducting multi-indicator and multi-weight risk assessment, the examinees' health risks in different systems can be accurately identified, and personalized health management plans can be tailored for them; this precise management not only improves the pertinence and effectiveness of health advice, but also avoids the waste of resources and poor results caused by traditional "one-size-fits-all" management. Early warning and intervention can also be carried out: potential risks under "normal" indicators can be discovered, and early warning and intervention suggestions can be provided to the examinees; this forward-looking health management strategy helps to intervene when the disease has not yet formed or is in the early stages, thereby effectively curbing the development of the disease and improving treatment effects and survival rates. And optimize the use of medical resources: by providing personalized medical guidance plans, it helps the examinees to accurately connect to the departments that can provide further diagnosis and treatment, reducing the blindness of patients' medical treatment and unnecessary trips to and from the hospital; this not only saves patients' time and energy, but also optimizes the allocation and utilization of medical resources and reduces the burden on the medical system. Enhance patient satisfaction and trust: by providing comprehensive, accurate and personalized health management services, it can significantly improve patients' satisfaction and trust in medical services; patients feel the professionalism and care of medical services, and are more willing to follow the doctor's advice for health management, thus forming a good doctor-patient relationship.

[0034] In another embodiment, Figure 5 As shown, a multi-dimensional indicator health management system based on physical examination reports is used in the above method, and the system includes: The data acquisition module 100 is used to obtain the basic information and physical examination data of the examinee and pre-process the obtained data; The risk assessment and grading module 200 is used to input the pre-processed data into a pre-trained health assessment model, assess the health risks of the seven systems and output the risk level, and grade and label them according to the risk level; A potential risk warning module 300 is used to identify key indicators based on the classification results, and to mine the correlation between key indicators to determine and warn potential risk indicators; The result output and suggestion module 400 is used to output a treatment guidance plan and supplementary examination suggestions for the examinee based on the classification results and the potential risk indicators warned.

[0035] In one example of this embodiment, Figure 6As shown, the system further includes: a marking module 500, which is used to mark in a three-color light manner or a highlight manner in the step of grading and marking according to risk levels.

[0036] In one example of this embodiment, Figure 6 As shown, the system further includes: a graphical module 600, which is used to combine the acquired physical examination data of the examinee with historical physical examination data to generate a data comparison trend chart and display it in a graphical manner.

[0037] The present invention provides a multi-dimensional indicator health management method based on physical examination reports. This method displays health risks in an intuitive and easy-to-understand way, enhances the examinee's awareness of their own health status, and thus improves their enthusiasm and initiative in participating in health management; this enhanced health awareness helps the examinee pay more attention to their own health and take active preventive and treatment measures; the pre-trained health assessment model analyzes multiple indicators, can accurately identify the examinee's health risks in different systems, and tailor a personalized health management plan for them; it can discover potential risks under "normal" indicators and provide early warning and intervention suggestions for the examinee. This forward-looking health management strategy helps to intervene when the disease has not yet formed or is in the early stages, thereby effectively curbing the development of the disease and improving treatment effects and survival rates.

[0038] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A multi-dimensional indicator health management method based on physical examination reports, characterized in that: The method comprises the following steps: Obtain the examinee's basic information and physical examination data based on the physical examination report, and pre-process the acquired data; The pre-processed data is input into the pre-trained health assessment model to assess the health risks of the seven major systems and output the risk level, and then grade and label them according to the risk level; the seven major systems include: cardiovascular system, endocrine system, respiratory system, digestive system, urinary system, reproductive system, and skeletal system; Identify key indicators based on the classification results and explore the correlation between key indicators to determine and warn of potential risk indicators; Based on the grading results and the potential risk indicators warned, the patient's treatment guidance plan and supplementary examination suggestions are output.

2. The method according to claim 1, characterized in that The method further comprises: In the step of grading and marking according to risk levels, marking is performed by a three-color light method or a highlight method.

3. The method according to claim 1, characterized in that The method further includes: combining the acquired physical examination data of the examinee with historical physical examination data to generate a data comparison trend chart, and displaying it in a graphical manner.

4. The method according to claim 1, characterized in that The basic information of the examinee shall at least include age, gender, and family medical history records; the physical examination data of the examinee shall at least include biochemical indicators and imaging test data.

5. The method according to claim 1, characterized in that In the step of identifying key indicators based on the grading results and mining the correlation between the key indicators, the correlation between the key indicators is mined by calling a deep feature mining module; wherein the deep feature mining module includes at least two convolutional layers, two pooling layers and one fully connected layer.

6. A multi-dimensional indicator health management system based on physical examination reports, used in the method according to any one of claims 1 to 5, characterized in that: The system comprises: A data acquisition module is used to obtain the basic information and physical examination data of the examinee based on the physical examination report and pre-process the acquired data; The risk assessment and grading module is used to input the pre-processed data into the pre-trained health assessment model, assess the health risks of the seven major systems and output the risk level, and grade and label them according to the risk level; Potential risk warning module, which is used to identify key indicators based on the classification results and explore the correlation between key indicators to determine and warn potential risk indicators; The result output and suggestion module is used to output the treatment guidance plan and supplementary examination suggestions for the examinee based on the grading results and potential risk indicators warned.

7. The system according to claim 6, characterized in that The system further comprises: a marking module, which is used to mark in a three-color light mode or a highlight mode in the step of grading and marking according to the risk level.

8. The system according to claim 6, characterized in that The system further comprises: a graphical module, which is used to combine the acquired physical examination data of the examinee with the historical physical examination data to generate a data comparison trend chart, and display it in a graphical manner.

Images