KR102028048B1 - System and Method for Providing and Evaluating Preventive Medical Information based on Data Base - Google Patents

Abstract

The present invention relates to an apparatus and method for optimizing disease prediction models through data collection and providing and evaluating data-based preventive medical information with improved reliability by monitoring and monitoring user behavior changes according to recommendations. A disease risk calculation and recommendation unit that calculates disease risk using disease prediction model and recommends drugs and behaviors for disease prevention and health promotion; Behavior change monitoring unit for verifying and monitoring changes in user behavior of recommended action items; Disease risk and occurrence monitoring unit for checking whether disease risk scores are reduced due to changes in user behavior and monitoring the actual occurrence of disease; Predictive models and effects to evaluate disease prevention, health promotion drugs, and behavioral effects Or wrong; intended to include; predictive model update section for updating the disease prediction model according to the prediction model, and the effect evaluation unit effects the result.

Description

System and Method for Providing and Evaluating Preventive Medical Information based on Data Base}

The present invention relates to the provision and management of preventive medical information, and specifically, to optimize disease prediction models through data collection, and to provide and evaluate data-based preventive medical information with improved reliability by monitoring and monitoring user behavioral changes according to recommendations. An apparatus and method are provided.

In recent years, medical institutions have established a medical information system and have digitalized all processes and records including patient reception, examination, prescription, treatment, hospitalization and discharge, and test results. Specifically, the medical image processing system PACS (Picture Archiving and Communications System), which transmits images taken by X-ray, CT, and MRI, and the Hospital Information System (HIS), which integrates all of them, It belongs to the information system.

In addition, information exchange system is being established to promote medical information exchange and standardization is also progressing for compatibility.

However, although the construction of the medical information system has advantages for improving the public health, there are many risks in terms of personal information protection and therefore, there are limitations in effectively utilizing it.

This is because medical information having patient information data is not only professional and public information related to the body of a patient, but also information that needs special protection among personal information. Such patient information digital data is easy to copy, and if personal information is leaked or inaccurate information is produced, there is a possibility that personal privacy is invaded. Therefore, there is a need for an effective technology for protecting personal information such as patient information data and preventing privacy invasion. In addition, the health information other than the medical information is scattered in various ways, and the system for integrating and utilizing reliability is still at the stage of opinion diversification and research.

In light of this trend, there are many limitations in that the accumulated biomedical information data should be used for the public interest.

In the clinical-based evidence-based medical paradigm, various types of bioinformation data have been accumulated, and the possibility of using the data-based medical paradigm is increasing.

Current methods for providing prevention and treatment recommendations in health care settings are often provided by health practitioners who are well trained or knowledgeable about what the latest guidelines for various diseases provide.

However, the results of higher-level and more reliable studies, such as multicenter clinical trials, require a lot of money for research, and it takes a long time for the results to be established and reflected in the guidelines.

In order to solve this problem, data-based methods can be supplemented in the areas where guidelines or results are not reported. Data-driven research can save time and money by complementing the evidence-based paradigm that is expensive over time.

In order to provide data-based preventive medical information, data collection and disease prediction model need to be established.

Current data collection typically receives information from wearable devices that users enter or wear directly through the app or the web.

If necessary, the information in the clinic can be obtained by entering a copy of the medical record or uploading a video copy.

Recently, a pilot project was established to establish a medical information exchange system to store meta information in the form of a central server and to effectively transmit medical information between hospitals with user approval using meta information.

In addition, several domestic and international studies are building user-oriented personalized health record (PHR) systems.

In addition, various disease prediction models are being researched and developed for a single disease.

However, in case of reporting the prediction model for the same disease, the research teams have different medical information data, and accordingly, the prediction model is also different.

The method of recruiting subjects differs slightly among research teams, depending on the design of the study, and may also be racial or regional.

Therefore, even in the aspect of conservatism of medical services, even if disease prediction models are developed, they are not effectively utilized, and it is difficult to know various types of predictive models from the user side, and they are not often used in practice.

Therefore, there is a demand for the development of new technologies for utilizing the data-based medical paradigm by using various types of bioinformation data in the medical field for the public interest.

Republic of Korea Patent Publication No. 10-2015-0121940 Republic of Korea Patent No. 10-20170268

The present invention is to solve the problem of the provision of preventive medical information of the prior art, to establish a disease prediction model through data collection, and to improve the reliability of data-based preventive medical information by monitoring the behavior change verification according to the recommendation Its purpose is to provide an apparatus and method for providing and evaluating.

The present invention calculates disease risk using a disease prediction model, identifies the risk factors that can be adjusted by the individual, and then recommends drugs and behaviors for disease prevention and health promotion, thereby enabling effective disease and health management. An object of the present invention is to provide an apparatus and method for providing and evaluating medical information.

An object of the present invention is to provide an apparatus and method for providing and evaluating data-based preventive medical information to accumulate various types of bioinformation data in the medical field and to realize a data-based medical paradigm using the public interest. There is this.

The present invention provides data-based preventive medical information that enables effective disease and health management by verifying and monitoring user behavioral changes, such as disease prevention, taking health-promoting drugs, and recommending behavior using disease prediction models established through data collection. And an apparatus and method for evaluation.

The present invention checks whether the disease risk score decreases due to a change in user behavior, monitors the actual occurrence of disease, evaluates the effects of disease prevention, drug administration, and behavioral execution for health promotion, and iteratively advances the disease prediction model. The purpose of the present invention is to provide an apparatus and method for providing and evaluating prophylactic medical information based on data that enables the provision of highly prophylactic medical information.

The objects of the present invention are not limited to the above-mentioned objects, and other objects that are not mentioned will be clearly understood by those skilled in the art from the following description.

Apparatus for providing and evaluating data-based preventive medical information according to the present invention for achieving the above object is a data collection unit for collecting data; calculating disease risk by using a disease prediction model, disease prevention and health promotion Disease risk calculation and recommendation unit for recommending drugs and behaviors; behavior change monitoring unit for verifying and monitoring changes in user behavior of disease prevention, health promotion drugs, and recommendation action items using the established disease prediction model; Disease risk and incidence monitoring unit to check whether the disease risk score is reduced and monitor the actual disease occurrence; predictive model and effect evaluation unit for evaluating the effects of disease prediction model and disease prevention, health promotion drugs, and behavioral effects; Predictive model to update disease prediction model according to the results of model and effect evaluation Update unit; characterized in that it comprises a.

Here, the data collection unit, the user's personal information, activity information and medical information that can determine the exposure to air pollutant or environmental pollutant exposure information by entering address information or activity information, health information entered from the app or the web, health Save screening information, exercise information in the cloud for users, store information about participation in health-related research and consent to use of information, whether or not they have health care-type insurance, and whether they are under management. According to the approval of the information characterized in that the user or group can analyze the health-related propensity.

In the data collection process in the data collection unit, the user authentication is performed by the user, a data generation expert or inspection company, and the personal information of the questionnaire response or activity information is made through the user authentication, and the health examination data cloud The certification in the process of storing is made through the Health Insurance Corporation certification, the inspection specialist certification, and the certification in the process of being stored in the medical institution medical information data cloud is the certification in the process of being stored in the medical institution certification, insurance information data cloud The certification of private insurance organizations, the participation in health research and the scope of consent are authenticated in the process of being stored in the data cloud, the certification of health research institutes, the direct input of data through app / web / wearable devices, or the reception of device information through device manufacturer certification. Characterized by It shall be.

In addition, the information of the user cloud by data collection in the data collection unit records only meta information, and in the data collection step, applying block chain technology to track forgery and tampering with health-related information. It features.

In addition, the disease risk calculation and recommendation unit sets a publicly available range so that the user can use the type of health-related information, disease information of interest, prediction model utilization information, and prediction model utilization results for group analysis. do.

The disease risk calculation and recommendation unit compares and analyzes the results of applying the various predictive models, provides the basis, calculates the disease risk score by applying the necessary variables to various published predictive models by using the approved health information. Analyze the values and behavioral dispositions of health habits, and the percentage of people actually performing in all or similar groups on the health promotion information of drugs or behaviors, success rate of achievement and reduction of risk score, and prevention of monitoring results. It provides a result.

In addition, the behavior change monitoring unit collects data on whether the user has selected and executed the recommended health promotion and prevention-related information and whether the target is achieved, and provides a change pattern and reliability evaluation over time to perform the corresponding action. It features.

The behavior change monitoring unit provides the average goal achievement success rate of the group whose behavior and pattern results show a similar pattern to the user, provides the methods selected by the subgroup with higher success rate, and the average success rate and selection of the group having different patterns. Methods are provided together.

The disease risk and occurrence monitoring unit may recalculate the predictive model risk score by using updated user cloud information and pattern information of behavior performance while performing specific health promotion and preventive actions.

The disease risk and occurrence monitoring unit may update the result indicators of the disease occurrence and death information, and repeat the risk score recalculation at the time when the cloud information of the users is updated by a predetermined portion or more.

The predictive model and the effect evaluation unit compare and analyze the accuracy of the predictive model by comparing the risk score change of various disease predictive models with the actual disease occurrence and death data, and analyze the user's interests by health promotion and disease prevention behavior. Predictive models with high target achievement rates are listed, and information on health promotion and prevention effects is assessed by the change in the risk score, the achievement rate of health promotion and disease prevention behavior, the actual health promotion index, and the reduction of disease and death.

In addition, the prediction model updating unit evaluates the similarity of users to users who are evaluated that the prediction model is not accurate while evaluating the effect of the prediction model, and improves the accuracy of the prediction model by reanalyzing the cloud data. do.

A method for providing and evaluating data-based preventive medical information according to the present invention for achieving another object includes a data collection step of constructing disease prediction models by collecting data; calculating disease risk by using disease prediction models, Identifying and controlling disease risk factors at an adjustable individual level, and then estimating and recommending disease risks for drug and behavioral recommendations for disease prevention and health promotion; using disease prediction models to take disease prevention and health promotion drugs and to recommend behaviors Behavior change monitoring step of verifying and monitoring changes in user behavior of implementation items; Disease risk and occurrence monitoring step to check whether disease risk score decreases due to user behavior change and monitor the actual occurrence of disease; Evaluate the effectiveness of disease prediction model and prevent disease Predictive models for evaluating drug and behavioral effects for health promotion, Efficacy evaluation step; Predictive model update step of updating the disease prediction model according to the effect evaluation results of the prediction model and the effect evaluation step; and characterized in that repeating these steps.

Such an apparatus and method for providing and evaluating data-based preventive medical information according to the present invention have the following effects.

First, data collection can be used to build disease prediction models, use existing predictive models, identify risk factors that can be controlled by individuals, recommend necessary measures, and verify user behavioral changes to increase the reliability of preventive care.

Second, the disease prediction model is used to calculate disease risk and effective disease and health management is possible by recommending drugs and behaviors for disease prevention and health promotion related to adjustable risk factors.

Third, it accumulates various types of bioinformation data in the medical field and uses it for the public interest to increase the possibility of utilizing the data-based medical paradigm.

Fourth, verify and monitor changes in user behavior such as disease prevention, taking health-promoting drugs, and conducting recommended behaviors, and reflect the degree of reliability of data so that users can effectively carry out disease and health care themselves.

Fifth, by continuously improving the disease prediction model by evaluating whether the disease risk score decreases due to changes in user behavior and monitoring the actual occurrence of disease, evaluating the effects of disease prediction models and the effects of drugs and behaviors for disease prevention and health promotion. It is possible to provide accurate and reliable preventive medical information.

1 is a block diagram showing a process for providing and evaluating data-based preventive medical information according to the present invention
2 is a block diagram of an apparatus for providing and evaluating preventive medical information based on data according to the present invention.
3 is a block diagram showing an authentication procedure in a data collection process
4 is a flow chart showing a data-based preventive medical information providing process according to the present invention
5 is a flowchart illustrating a disease prediction model updating process by providing and evaluating preventive medical information based on data according to the present invention.

Hereinafter, a preferred embodiment of an apparatus and method for providing and evaluating data-based preventive medical information according to the present invention will be described in detail.

Features and advantages of the apparatus and method for providing and evaluating data-based preventive medical information according to the present invention will become apparent from the detailed description of each embodiment below.

1 is a block diagram showing a process for providing and evaluating data-based preventive medical information according to the present invention, Figure 2 is a block diagram of a device for providing and evaluating data-based preventive medical information according to the present invention.

An apparatus and method for providing and evaluating data-based preventive medical information according to the present invention are to enable a data-based medical paradigm by accumulating various types of bioinformation data in the medical field and using it for the public interest.

To this end, the present invention constructs a disease prediction model through data collection, calculates disease risk using a configuration and disease prediction model for monitoring and monitoring user behavior changes according to recommendation, drugs for disease prevention and health promotion, and It may include a component that makes an action recommendation.

The present invention includes a configuration of updating disease prediction models by evaluating the effects of disease prediction models and evaluating disease prevention, health promotion drugs, and behavioral effects by checking whether disease risk scores are reduced due to changes in user behavior and monitoring actual disease occurrence. can do.

Specifically, as shown in Figure 1, the step of collecting data to build a disease prediction model (S101), calculating the disease risk using the disease prediction model, and the step of recommending drugs and behaviors for disease prevention and health promotion (S102) and verifying and monitoring user behavioral changes such as disease prevention, taking health promoting drugs, and conducting recommended behaviors using the established disease prediction model (S103), and checking whether disease risk scores are reduced due to user behavioral changes and Monitoring the actual occurrence of the disease (S104), and evaluating the effect of disease prediction model and disease prevention, health promotion drugs, behavioral effects (S105) and updating the disease prediction model (S106).

Apparatus for providing and evaluating the data-based preventive medical information according to the present invention, as shown in Figure 2, the data collection unit 10 to build a disease prediction model by collecting data, calculating the disease risk using a disease prediction model Changes in user behaviors, including disease risk calculation and recommendation unit 20 for recommending drugs and behaviors for disease prevention and health promotion, and disease prevention and health promotion drugs taking and recommending behavior using the established disease prediction model. Behavior change monitoring unit 30 to verify and monitor the disease risk score due to changes in user behavior, and whether the disease risk and occurrence monitoring unit 40 to monitor the actual occurrence of disease, and evaluate the effect of disease prediction model and Predictive model and effect evaluator 50 for evaluating disease prevention, health promoting agents, and behavioral effects, and according to the effect evaluation results of the predictive model and effect evaluator 50 Disease prediction prediction model to a model updating includes updating section 60.

Referring to step (S101) of building a disease prediction model by collecting data as follows.

Whether they are treated at a medical institution such as a hospital / health center, whether they have health information entered through the app or the web, have a medical examination such as genetic testing, blood tests, or imaging tests, or other health-related information (fitness, health, yoga). , Swimming, etc.) when the health-related information is recorded in the cloud (cloud) for the user.

This may involve taking part in a health-related study and what kind of testing was done in that study.

At the moment, such information is not properly managed, and from the user's point of view, it is not possible to identify and manage which researches he or she agrees with, and data is managed only at the individual researcher level.

In addition, the method may include recording information on whether the user is under health care type insurance coverage and management.

Currently, health-care insurance is being developed and self-testing is conducted. This information is recorded so that it can be helpful in identifying the user's health-related preferences or efforts.

In addition, the records generated as described above may include a step of separately recording the types of records, the reliability and the objectivity of the information, and the reliability and objectivity part may be guaranteed by the experts involved in generating the related information. Can be.

The present invention is to introduce the online activity information of the individual in the cloud to analyze the health-related propensity of the user or group according to whether the user's information approval or not, it is possible to set up similar groups reflecting the propensity and to recommend the behavior and contribute to more effective health promotion do.

3 is a block diagram showing an authentication procedure in a data collection process.

In the process of collecting data, the certification is done by a professional or inspection company, and the certification in the process of being stored in the health examination data cloud (survey, blood, gene, image, etc.) is made through the Health Insurance Corporation certification and the inspection specialist certification. .

In addition, authentication in the process of being stored in the healthcare institution data cloud is achieved through certification of the healthcare institution.

And the certification in the process of being stored in the insurance information data cloud is achieved through the certification of private insurance institutions.

In addition, certification in the process of being stored in the health research participation data cloud is achieved through the certification of the health research institute.

Direct input of data through the app / web / wearable device or reception of device information is achieved through device manufacturer certification.

In addition, there is data stored in SNS data cloud (KakaoTalk, Facebook, LinkedIn, etc.).

Information in the user cloud may be direct medical information, but only the meta information is recorded because of the security risks when centrally managing personal medical information.

In the data collection step, it is also possible to apply block chain technology to track the forgery of health-related information and to improve reliability.

When information is generated, you can create a block and, if a change occurs, create a block to track changes. This process is to store the specific information that the actual change has occurred. The reason for separating it is that it is difficult to make the health information that can be sensitive information into a block considering that it is open to many people due to the characteristics of blockchain technology. The contents are intended to be stored in a form other than blocks.

Table 1 shows an example of metadata change record management using blockchain technology.

And the disease risk calculation using the disease prediction model, and the steps for the drug and behavior recommendation for disease prevention and health promotion (S102) will be described in detail as follows.

This step is anonymized to reduce the risk of personal information exposure, so that users can use their own health-related information, disease information they are interested in, predictive model utilization information, and predictive model utilization results for group analysis. It is a process to set the public scope.

Using the meta-information of individual cloud, identify which health institutions, health check-up centers, insurance companies, research institutes, etc. to request data, and block access and changes (information on request) to cloud information while requesting according to approval To record.

In addition, the results of applying various predictive models can be analyzed and compared, and the evidence can be easily identified, and the risk risk score can be calculated by applying the necessary variables to various published predictive models using approved health information. have.

By using several prediction models, the variables used in each prediction model and the difference in scores, accuracy, and evidence are provided together. Such a process can provide a feed in a form easily explained by an expert.

And analyzing the values and behaviors of the user's health habits.

Through various contact media such as apps, web, and phone, users can be provided with health-related and preventive information and collect responses to diseases they choose to be interested in and diseases they are not interested in.

If the user approves the use of online-related information, the user's propensity can be analyzed using the user's online media response information, and the result can be analyzed along with the results of applying various prediction models.

This is to raise the reliability of the analysis result by asking the user's interest beyond the level of using only the public information, and providing various health information and using the response information.

When providing predictive model results, a group of people with the same disease of interest, the response to various health-related information, and the propensity analysis using online information can be set up.

In addition to the existing general demographic variables, the method of setting up groups can be used to reflect behavioral changes due to the inclination that is difficult to collect with data, which can be used for future behavioral recommendation.

In addition, a similar group may have a high degree of interest, but the user may have a disease that the user is not interested in.

In this case, if there is data on the user's cloud of health information, the predictive model score can be calculated.If there is no data found on the user's cloud, the data retention frequency in the similar group is provided and the data can be used. Predictive models and the process of providing evidence and accuracy.

This is to provide diseases and predictive model results that the user does not care about but need to be interested in, and to provide more accurate predictive model through testing, thus contributing to effective health care. .

And providing health promotion and prevention information for disease (s) with a high disease risk score.

This process provides for each piece of information the fraction of people actually performing in similar groups, their performance success rate and risk score reduction rate, and the results of monitoring outcomes.

Related information Information that can increase performance success rates (e.g. diet scheduling nutritionists, exercise coaches, non-smoking counselors, clubs that work together, medicine cabinets that alert you to taking medication) and methods chosen by people with high or low success rates. Frequency can also be provided.

In the process of such a process, it is possible to record in the block how far the user has seen in each step and which step is the progress.

In addition, a detailed description of the step S103 of monitoring and monitoring user behavioral changes, such as disease prevention, health promotion drug administration, and recommendation behavior, using the established disease prediction model is as follows.

Even if you recommend a variety of information, it is not easy to apply and maintain in your life. It is necessary to save and analyze data about the process of moving into actual behavior and making it into a habit.

The present invention allows the user to select a method that can be implemented among the recommended health promotion and prevention-related information.

It's possible to add a method that you've been running, even though it wasn't recommended, and you'll have the option to make it public for other users.

You can also choose to monitor actual numbers, take medications or exercise to collect data about whether you are running.

This can be done by using methods such as surveys, health institution and health-related cloud information update status, inspection device, wearable device information collection, fellowship certification to run together, and certification of experts to assist the implementation (coach, health care workers, etc.). .

And it may include the process of providing a change pattern and reliability evaluation over time whether or not to perform the action.

The behavior and pattern results may include providing an average success rate for a group showing a pattern similar to a user, and also providing methods selected by a subgroup having a higher success rate, and an average success rate and a selection method for groups having different patterns. It is also possible to provide them together.

In addition, the steps for checking whether the disease risk score decreases due to the change in user behavior and monitoring the actual disease occurrence are described in detail as follows.

This recalculates the predictive model scores by reflecting the update status of personal cloud information and individual behavior changes through the provision of information, to check whether the scores are reduced and to monitor the occurrence of diseases.

In the case of disease occurrence and death information, when the results of the initial prediction model are calculated and health promotion and prevention information is provided, success or failure rate reduction and prevention effect evaluation results are provided for each information.

Sensitive information is often reluctant to be disclosed, but we have seen the necessity in evaluating the prevention effect and informed that it is protected through anonymization of personal information.

And while performing specific health promotion and preventive actions, the updated user cloud information and pattern information of behavior performance are utilized to recalculate predictive model risk scores and update outcome indicators such as disease occurrence and death information. It can also be implemented when is updated over a certain fraction.

And the evaluation of the effect of disease prediction model and disease prevention, health promoting agent, behavioral effect (S105) to explain the step (S106) for updating the disease prediction model in detail as follows.

Comparing and evaluating the accuracy of predictive models by comparing the risk scores of different disease prediction models with actual disease occurrence and death data. By estimating the target, the reliability of the evaluation can be improved.

If the number of users is large, it is possible to provide the user with optimized prediction accuracy by evaluating the accuracy of the prediction model for each subgroup.

Lists predictive models with high user interest by health promotion and disease prevention behavior (also taking medications or supplements), assesses information on health promotion and prevention effects by varying risk scores, and predicts models Information such as accuracy, evidence, and evaluation history can be considered.

In addition, it is possible to distinguish between well-maintained people and those who do not by using the information on the success rate of the actions of health promotion and disease prevention behavior as well as the entire user group.

Evaluate the effects of health promotion and disease prevention behaviors on specific users and similar groups, and evaluate the effects of health promotion and disease prevention behaviors on similar groups to provide optimized recommendations to users.

In order to update the prediction model based on the effect, the similarity of the users is evaluated for the users whose evaluation model is not accurate while evaluating the effect of the prediction model, and the accuracy of the prediction model is improved by reanalyzing the cloud data. It is possible to build a sustainable accuracy improvement system that can improve the accuracy for each subgroup.

Referring to the data-based preventive medical information providing process according to the present invention step by step.

4 is a flowchart illustrating a process of providing data-based preventive medical information according to the present invention.

In order to provide data-based preventive medical information, when the selection of the diseases of interest is made (S401), identity authentication and user cloud information disclosure scope approval (S402), and disease-specific predictive model risk score calculation (S403).

Subsequently, disease risk scores and supporting papers are provided, individual or similar group scores, and prediction accuracy are provided. (S404)

If the selected diseases are not made in question, provide customized prevention-related feeds for diseases of interest and diseases not selected (S411), collect user response information data (S412), and retain other than demographic information such as gender and age. Analyze health promotion propensity and set similar groups by analyzing data and feed response information (S413).

Similar groups provide disease prediction models with high interest and provide results when it is possible to calculate the data they have, and provide additional tests and accuracy of the corresponding prediction models if additional information is needed (S405).

Subsequently, it provides health promotion and prevention information about diseases with high predictive model risk scores, and provides success rate, risk score reduction rate, and prevention effect evaluation results for each information (S406).

Here, the information that can increase the success rate of the related information and the person with a high or low success rate may be provided together.

Referring to the disease prediction model update process by providing and evaluating data-based preventive medical information according to the present invention step by step.

5 is a flowchart illustrating a disease prediction model update process by providing and evaluating data-based preventive medical information according to the present invention.

First, among the recommended health promotion and prevention-related information, the user selects a method to be implemented (S501).

Here, you can add a method that you have not recommended, but you are running, and you can choose whether the method can be disclosed to other users.

In order to confirm that the user is performing by selecting actual numerical monitoring, taking medicine or exercising, collecting survey information, testing equipment, wearable device information, certification of fellowships to be executed together, and certification of experts to assist the implementation (coach, health) Collect data using methods such as medical personnel. (S502)

Then, the average success rate of the group showing a pattern similar to the user in the process and method of performing the action is provided (S503).

Here, the methods selected by the subgroup having a higher success rate are also provided, and it is also possible to provide the average success rate and the selection methods of the groups having different patterns.

By using the updated user cloud data (S504), using the updated user cloud information and pattern information of behavioral performance, such as during specific health promotion and preventive actions, the predictive model risk score is recalculated and disease occurrence and death information is used. And update the result indicator.

Apparatus and method for providing and evaluating data-based preventive medical information according to the present invention as described above are effective disease and health by calculating disease risks using disease prediction models and recommending drugs and behaviors for disease prevention and health promotion. Management is made possible.

In addition, it accumulates various types of bioinformation data in the medical field and uses it for the public interest to realize a data-based medical paradigm.

It will be understood that the present invention is implemented in a modified form without departing from the essential features of the present invention as described above.

Therefore, the described embodiments should be considered in descriptive sense only and not for purposes of limitation, and the scope of the present invention is shown in the claims rather than the foregoing description, and all differences within the equivalent scope are included in the present invention. It should be interpreted.

10. Data collection unit 20. Disease risk calculation and recommendation unit
30. Behavior change monitoring unit 40. Disease risk and occurrence monitoring unit
50. Prediction model and effect evaluation unit 60. Prediction model update unit

Claims (13)

If you input address information or activity information depending on whether the user has approved the information, personal information, activity information and medical information of users who can identify the air or environmental pollutant exposure information, health information entered in the app or the web, A data collection unit storing health examination information and exercise information in a cloud for a user, and separately recording the types of records and the reliability and objectivity of the records with respect to the records generated by data collection;
Compare and analyze the results of applying disease prediction models, provide evidence, apply the necessary parameters to various published prediction models using approved health information, calculate disease risk scores, and evaluate user's health habits online and online. Analyze user's propensity and behavioral propensity using media response information, and monitor the percentage of people who are actually performing in all or similar groups on health promotion information of medication or behavior, success rate of achievement of achievement goal, and reduction of risk score Results The disease risk calculation and recommendation unit provides a result of evaluating the preventive effect, calculates disease risk, and recommends drugs and behaviors for disease prevention and health promotion;
We use the established disease prediction model to verify and monitor changes in user behavior of disease prevention, health promotion medications, and recommended behavior enforcement items, and provide the average success rate of group achievement of groups whose behavior and pattern results show similar patterns to users. A behavior change monitoring unit providing methods selected by the subgroup having a higher success rate and providing average success rates and selection methods of different groups with different patterns;
It checks whether the disease risk score decreases due to the change of user behavior, monitors the actual disease occurrence, and uses the updated user cloud information and the pattern information of behavior performance while performing specific health promotion and preventive actions, and uses the predicted model risk score. Recurrent disease risk and occurrence monitoring;
Prediction model and effect evaluation unit for evaluating the effects of disease prediction model and disease prevention, health promoting agent, behavioral effect;
The disease prediction model is updated according to the effect evaluation results of the predictive model and the effect evaluation unit, and the users who are judged that the predictive model is not highly accurate while evaluating the effect of the predictive model are classified, and the similarity of the users is evaluated. Apparatus for providing and evaluating data-based preventive medical information comprising a; prediction model update unit for re-analyzing using the cloud data to improve the accuracy of the prediction model for the corresponding group and all users. The method of claim 1, wherein the data collection unit,
Store information about participation in health-related research, consent to use of information, whether or not you have health care insurance, and whether
Apparatus for providing and evaluating data-based preventive medical information, characterized in that to analyze the health-related propensity of the user or group according to the user's approval of information. According to claim 1, During the data collection process in the data collection unit,
Identity verification is done by the user, expert in data generation or inspection company, personal information of questionnaire response or activity information is done through user identity verification, and certification in the process of being stored in the health examination data cloud , Through the inspection company certification,
The certification in the process of being stored in the healthcare information data cloud is the certification of the healthcare institution, and the certification in the process of being stored in the insurance information data cloud is the certification of the private insurance institution, the participation in the medical research and the scope of consent. The certification is a device for providing and evaluating data-based preventive medical information, characterized in that the certification of the healthcare research institute, the data input directly through the app / web / wearable device or the device information received through the device manufacturer certification. The method of claim 1 or 2, wherein the information of the user cloud by data collection in the data collection unit records only meta information,
Apparatus for providing and evaluating preventive medical information based on data, comprising applying a block chain technology to track forgery and tampering with health-related information in a data collection step. According to claim 1, Disease risk calculation and recommendation unit,
Providing data-based preventive medical information to the user by setting the publicly available range to use the type of health-related information owned by the user, disease information of interest, information on using the predictive model, and the result of using the predictive model for analysis by group And apparatus for evaluation. delete According to claim 1, Behavior change monitoring unit,
Collects data about whether the user is choosing and implementing from the recommended health promotion and prevention information, and whether
Apparatus for providing and evaluating data-based preventive medical information, characterized in that to provide a change pattern and reliability assessment over time whether to perform the action. delete delete According to claim 1, Disease risk and occurrence monitoring unit,
Apparatus for providing and evaluating data-based preventive medical information, characterized in that for updating the outcome indicators of disease occurrence and death information, and repeating the risk score re-attribution when the cloud information of the user is updated over a certain percentage. The method of claim 1, wherein the prediction model and the effect evaluation unit,
Compare and analyze the accuracy of predictive models by comparing the risk scores of different disease prediction models with actual disease occurrence and death data.
Lists predictive models with high user interest or goal achievement rate by health promotion and disease prevention behavior, and changes in risk score, rate of achievement of health promotion and disease prevention behavior, actual health promotion indicators, and disease and death reduction indicators. Apparatus for providing and evaluating data-based preventive medical information, characterized in that for evaluating the information. delete If the data collection unit collects the data according to the user's information approval and inputs address information or activity information, the personal information, activity information, medical information, and apps of users who can identify the air pollutant or environmental pollutant exposure information accordingly The health information, health examination information, and exercise information input from the web or the web are stored in the cloud for the user, and the records generated by data collection are classified by the type of record, reliability of information, and objectivity. A data collection step of recording and constructing disease prediction models;
The disease risk calculation and recommendation department compares and analyzes the results of applying the disease prediction models, provides approved disease information, calculates the disease risk scores by applying the necessary variables to various published prediction models. Analyze the user's disposition and behavioral disposition using the values of health habits and online media response information, and achieve the actual percentage and performance goals of people who are actually performing in all or similar groups on health promotion information of each drug or behavior Calculate disease risk by providing success rate, risk score reduction rate, and monitoring result evaluation results, identify disease risk factors at an adjustable individual level, calculate disease risk, recommend drugs and actions for disease prevention and health promotion, and Recommendation step;
Using the disease prediction model established by the Behavioral Change Monitoring Unit, we verify and monitor the changes in user behaviors of disease prevention, health promotion drugs, and recommended behaviors, and achieve the average goal of the group whose behavior and pattern results are similar to those of users. A behavior change monitoring step of providing a success rate, providing a method selected by a subgroup having a higher success rate, and providing an average success rate and selection methods of different groups with different patterns;
The disease risk and occurrence monitoring department checks whether the disease risk score decreases due to changes in user behavior, monitors the actual disease occurrence, and updates the user cloud information and pattern information of behavior performance during specific health promotion and preventive actions. Monitoring disease risk and incidence to regenerate predictive model risk scores;
Predictive model and effect evaluation step of evaluating the effect of the disease prediction model and predicting the effects of drugs and behaviors for disease prevention and health promotion in the predictive model and effect evaluation unit;
The prediction model updating unit updates the disease prediction model according to the results of the evaluation of the effects of the prediction model and the effect evaluation step, and evaluates the users whose predictive models are not accurate while evaluating the effectiveness of the prediction model, and then targets them. The predictive model update step of evaluating the similarity of each other and re-analyzing the cloud data to improve the accuracy of the predictive model for the corresponding group and the entire user; and repeating these steps. Method for provision and evaluation.

Images