Health management system and method of operation thereof

Abstract

Disclosed are a health management system and a method for operating the same. The health management system according to various embodiments can include a backend for frontend (BFF) server serving a front end, performing an interface function with a health management application installed in a user terminal, and receiving a service request from the health management application; a fast healthcare interoperability resource relay server converting the service request based on a fast healthcare interoperability resource (FHIR) protocol and generating a fast healthcare interoperability resource request, and transmitting the fast healthcare interoperability resource request to an internal network of a medical institution; and a fast healthcare interoperability resource server located in the internal network, performing a process on the fast healthcare interoperability resource request. In this case, the health management system is linked with an old system of the medical institution, and can perform a process based on health data of a user independently from the old system. According to such a system, the health management service can be comprehensively provided to the user in connection with the medical institution.

Description

Technical Field

[0001] This invention relates to technologies for monitoring and managing user health.

[0002] This patent is the result of a project supported by the Ministry of Trade, Industry and Energy of Korea (project management (professional) agency: Korea Institute for Industrial Technology Planning and Evaluation). The project's inherent number is 2410013887, the project number is 07892971, the research project name is Biotechnology Development (R&D), the research topic name is the development of a biosignal monitoring and real-time early warning system beacon for critically injured patients based on artificial intelligence (Biosiginal Emergency Advanced Critical Observation Network), and the research period is from April 1, 2025 to December 31, 2025. Background Technology

[0003] Recently, with the development of wearable devices and mobile healthcare technologies, various forms of health management applications (APPs) have been launched that can collect users' body information or health habit information and manage their health status based on this. These health management applications perform functions such as providing personalized feedback based on information such as weight, diet, sleep, and exercise input by users, or visualizing the user's health status using relevant information.

[0004] However, most existing health management applications do not integrate with the legacy systems of healthcare institutions, making it impossible to comprehensively verify user medical information managed by the institutions (e.g., prescription information, blood test results, etc.). Furthermore, the user health information collected through these applications is difficult to apply to healthcare institutions. Therefore, existing health management applications have significant limitations in providing users with comprehensive and holistic health management services.

[0005] Existing technical documents

[0006] Patent documents

[0007] Korean Patent Publication No. 10-2014-0080143 (published on June 30, 2014) Summary of the Invention

[0008] Technical issues

[0009] The various technical objectives addressed by the embodiments of the present invention are to provide a system and its working method capable of providing users with comprehensive and integrated health management services.

[0010] Specifically, the technical objective addressed by various embodiments of the present invention is to provide a structure and operating method for a health management system that can effectively link with the legacy system of a medical institution.

[0011] Furthermore, another technical objective achieved through various embodiments of the present invention is to provide a system and its working method capable of predicting health abnormalities at an early stage based on a user's health data and supporting proactive management measures.

[0012] The technical objectives of this invention are not limited to those mentioned above. Those skilled in the art to which this invention pertains can clearly understand other technical objectives not mentioned through the following description.

[0013] Technical solution

[0014] To achieve the technical objectives described above, the health management system of various embodiments of the present invention may include: a backend-for-frontend (BFF) server that serves the frontend, executes interface functions with a health management application located on a user terminal, and receives service requests from the health management application; a Fast Healthcare Interoperability Resource Relay Server that converts the service requests based on the Fast Healthcare Interoperability Resource (FHIR) protocol and generates Fast Healthcare Interoperability Resource Requests, and transmits the Fast Healthcare Interoperability Resource Requests to the internal network of the medical institution; and a Fast Healthcare Interoperability Resource Server located in the internal network that processes the Fast Healthcare Interoperability Resource Requests. In this case, the health management system is linked with the medical institution's legacy system, and the health management system can independently perform processing based on the user's health data.

[0015] In several embodiments, the legacy system may include an electronic medical record (EMR) system.

[0016] In several embodiments, the health management system provides at least a portion of the health data to the legacy system and can receive the user's medical information from the legacy system.

[0017] In several embodiments, the rapid medical interoperability resource relay server can transmit the rapid medical interoperability resource request to the rapid medical interoperability resource server through a gateway server, the gateway server being used to control access to the internal network.

[0018] In several embodiments, the service request includes a confirmation request to determine whether the user is a registered patient at the medical institution. The confirmation request includes the user's personal identification information and is executed based on a membership registration application through the health management application. The rapid medical interoperability resource request includes a rapid medical interoperability resource confirmation request corresponding to the confirmation request. The rapid medical interoperability resource server uses the personal identification information to confirm whether the user is a registered patient through the legacy system and can send a rapid medical interoperability resource response including the confirmation result back to the rapid medical interoperability resource relay server.

[0019] In several embodiments, the service request includes a prescription information query request, the query request includes patient identification information registered at the medical institution, the rapid medical interoperability resource request includes a rapid medical interoperability resource query request corresponding to the query request, the rapid medical interoperability resource server uses the patient identification information to obtain the prescription information through the legacy system, and can send a rapid medical interoperability resource response including the prescription information back to the rapid medical interoperability resource relay server.

[0020] In several embodiments, the service request includes a health record submission request, the rapid medical interoperability resource request includes a rapid medical interoperability resource submission request corresponding to the submission request, the rapid medical interoperability resource server stores the health record in a legacy system-accessible database, and can send a rapid medical interoperability resource response including the storage result back to the rapid medical interoperability resource relay server.

[0021] In several embodiments, the health data includes the user's body composition analysis information and health records submitted by the user through the health management application. The body composition analysis information is obtained from a body composition analysis server included in the legacy system. The health management system also includes an internal integration server that provides a visual representation of the health data through a control panel that can be accessed through the user interface (UI) of the electronic medical record (EMR) system included in the legacy system.

[0022] In several embodiments, the health record includes health habit information and weight information, and the control panel may include: a first interface displaying the user's basic information; a second interface displaying the health habit information, the second interface also displaying an assessment score calculated based on the health habit information; and a third interface displaying body information including the weight information and the body composition analysis information.

[0023] In several embodiments, the internal integration server predicts the user's success rate toward health goals by analyzing the health records, and the predicted success rate can be additionally provided through the control panel.

[0024] In several embodiments, the internal integration server predicts the success rate using a machine learning model. The health record includes health habit information and weight information. The machine learning model may include: a habit encoder that encodes the health habit information to generate a habit embedding; a will encoder that encodes the user's will information to generate a will embedding, the will information including at least one of the user's weight changes, the frequency of spontaneous submission of the health habit information, and the frequency of changes in the health goals; an environment encoder that encodes the user's living environment information to generate an environment embedding; an aggregator that aggregates the habit embedding, the will embedding, and the environment embedding to generate a comprehensive embedding; and a predictor that predicts the success rate based on the comprehensive embedding.

[0025] In several embodiments, the health management system further includes an internal integration server that analyzes the health data using a machine learning model to predict the risk of a user experiencing a health abnormality. The machine learning model may include: an encoder that generates variable embeddings corresponding to predetermined health variables; an interaction encoder that generates interaction embeddings corresponding to predetermined pairs of health variables based on the variable embeddings; a variable weighting estimator that estimates user-specific variable weights for the health variables based on the user's clinical information obtained from the legacy system; an interaction weighting estimator that estimates user-specific interaction weights for the pairs of health variables based on the clinical information; an aggregator that generates a composite embedding by aggregating the variable embeddings reflecting the variable weights and the interaction embeddings reflecting the interaction weights; and a predictor that predicts the risk level based on the composite embedding.

[0026] In several embodiments, when the predicted risk level is above a threshold and the preset waiting time expires, the internal integration server provides a notification related to the health anomaly. Before the waiting time expires, if the user is detected to have experienced an event that increases the risk of the health anomaly, the internal integration server immediately provides the notification. Before the waiting time expires, if the user is detected to have experienced an event that decreases the risk of the health anomaly, the internal integration server may reset the waiting time and re-predict the risk level using the machine learning model.

[0027] The effects of the invention

[0028] According to various embodiments of the present invention, the health management system can work in conjunction with the legacy system of a medical institution to provide users with general or comprehensive health management services. Furthermore, the health management system and the legacy system independently perform health data processing (e.g., analysis, monitoring, etc.), thus, without affecting the structure or performance of the legacy system, the health management system can reflect (or integrate) the user's health-related information to the legacy system.

[0029] Furthermore, by providing users' health-related information to healthcare professionals through the control panel, the health management system can further and more effectively provide users with comprehensive health management services.

[0030] Furthermore, the health management system includes a backend for frontend (BFF) server, a fast healthcare interoperability resource (FHIR) relay server, a gateway server, a fast healthcare interoperability resource server, and an internal integration server. The fast healthcare interoperability resource server and the internal integration server are located in the medical institution's internal network, thus enabling effective linkage between the health management system and the medical institution's legacy systems.

[0031] Furthermore, by comprehensively analyzing users' health variable data and clinical information through machine learning models, the risk level of a user developing health abnormalities can be accurately predicted. Moreover, if the risk level exceeds a certain threshold, notifications can be provided to users and / or healthcare professionals to proactively prevent health abnormalities from occurring.

[0032] The effects based on the technical concept of this invention are not limited to those mentioned above. Those skilled in the art to which this invention pertains can clearly understand other effects not mentioned through the following description. Attached Figure Description

[0033] Figure 1 This is an illustrative diagram illustrating the working environment of a health management system for various embodiments of the present invention.

[0034] Figure 2 This is an illustrative block diagram used to explain the internal structure and operation of a health management system according to several embodiments of the present invention.

[0035] Figures 3 to 5 This is an illustrative flowchart to show the working method of a health management system according to several embodiments of the present invention.

[0036] Figure 6 This is an illustrative diagram illustrating the screen structure of a control panel for explaining various embodiments of the present invention.

[0037] Figures 7a to 7d An example diagram showing an instance of the control panel.

[0038] Figure 8 This is an example diagram illustrating the input and output data of a machine learning model for predicting health anomalies according to several embodiments of the present invention.

[0039] Figure 9 The diagram illustrates the structure and operation of machine learning models in various embodiments of the present invention.

[0040] Figure 10 This is an illustrative diagram illustrating a model training method based on multi-task learning, which is used to explain several embodiments of the present invention.

[0041] Figure 11 This is an illustrative diagram used to illustrate the structure and operation of a machine learning model for predicting the success rate of health goals in several embodiments of the present invention.

[0042] Figure 12 An illustrative diagram is provided to show a computing device capable of implementing a health management system and / or its structural elements that are among the various embodiments of the present invention. Detailed Implementation

[0043] Hereinafter, several embodiments of the present invention will be described in detail with reference to the accompanying drawings. The advantages, features, and implementation methods of the present invention can be found by referring to the accompanying drawings. Figure 1 The detailed embodiments become clear. However, the technical concept of the present invention is not limited to the following embodiments and can be implemented through different methods. The following embodiments are only used to complete the technical concept of the present invention so that those skilled in the art can fully understand the scope of the present invention. The technical concept of the present invention is limited only to the scope of the claims.

[0044] In describing various embodiments of the present invention, detailed descriptions of known structures or functions will be omitted when it is determined that such descriptions may obscure the essence of the invention.

[0045] Unless otherwise defined, the terminology (including technical and scientific terms) used in the following embodiments is intended to be understood by one of ordinary skill in the art to which this invention pertains, and may vary depending on the intent, practice, or emergence of new technologies of one of ordinary skill in the art. The terminology used in this specification is for illustrative purposes only and does not limit the scope of the invention.

[0046] In the following embodiments, unless explicitly specified in the context as singular, the singular expressions used include plural meanings. Furthermore, unless explicitly specified in the context as plural, the plural expressions include singular meanings.

[0047] Furthermore, in the following embodiments, the terms “first,” “second,” “A,” “B,” “(a),” “(b),” etc., are used only to distinguish one structural element from another, and the terms do not limit the nature, importance, or order of the corresponding structural elements.

[0048] In the following embodiments, the structural elements described using terms such as "unit," "module," "block," and "device" and the functional blocks shown in the figures can be implemented in software, hardware, or a combination thereof. For example, software may include machine code, firmware, embedded code (or software), application software, or a combination thereof. Hardware may include, for example, electrical circuits, electronic circuits, processors, computers, integrated circuits, integrated circuit cores, manual devices, or a combination thereof. As more specific examples, "unit," "module," etc., may include structural elements such as software structural elements, object-oriented software structural elements, class structural elements, and task structural elements, as well as processes, functions, attributes, programs, subroutines, program code segments, drivers, firmware, microcode, circuits, data, databases, data structures, lists, arrays, and variables.

[0049] Hereinafter, several embodiments of the present invention will be described in detail with reference to the accompanying drawings.

[0050] Figure 1 This is an illustrative diagram illustrating the working environment of the health management system 10 of various embodiments of the present invention.

[0051] like Figure 1 As shown, the health management system 10 in this embodiment is a computing device / system equipped with health management functions for one or more users, and can work in conjunction with the user terminal 11 and the legacy system 12 of the medical institution. However, the scope of the invention is not limited thereto. To support health management for multiple users, the health management system 10 can also be linked with multiple user terminals.

[0052] exist Figure 1 In the illustrated work environment, the health management system 10 collects users' health data (e.g., health habits, weight, etc.) through a health management application set on the user terminal 11, and can analyze it to execute health management functions for the corresponding users (e.g., health abnormality notifications, etc.).

[0053] Furthermore, the health management system 10 can provide (i.e., can share with the medical institution) users' health-related information (e.g., health data and / or its analysis results) to the medical institution's legacy system 12. In this case, the health management system 10 performs health data-based processing independently of the legacy system 12, thus reflecting (or integrating) users' health-related information to the legacy system 12 without affecting its structure or performance. The health management system 10 can also visualize users' health-related information through a control panel, as described below. Figures 6 to 7d Please provide further explanation.

[0054] Furthermore, the health management system 10 obtains users' medical information (e.g., various clinical information) from the legacy system 12, and can analyze the corresponding users' health data and medical information to perform health management tailored to the users. For example, the health management system 10 uses a machine learning model to predict the risk of health abnormalities occurring within a specific time frame (e.g., 24 hours, 6 hours, etc.) in advance, and can provide users with notifications for health management based on this. Such notifications can be provided through a health management application installed on the user terminal 11 (e.g., when the user is a registered patient at a medical institution, a notification can be provided to the attending medical staff). As another example, the health management system 10 can detect health abnormalities in real time or on a scheduled basis using a machine learning model.

[0055] Later, refer to Figure 2 The internal structure and operation of the health management system 10 are explained in further detail.

[0056] Depending on the circumstances, the health management system 10 may also be named "health management platform", "medical system / platform", "comprehensive health management service system / platform", etc.

[0057] User terminal 11 refers to a user-side device (i.e., a computing device) that can be linked with the health management system 10. A health management application linked with the health management system 10 can be installed on user terminal 11, and users can receive health management functions (e.g., submitting health records and receiving health abnormality notifications) through the corresponding application.

[0058] For example, the user terminal 11 can be a device in various forms such as a smartphone, tablet, desktop computer, smartwatch or other wearable device, but is not limited to these.

[0059] Legacy system 12 refers to the pre-built and operational information system of the medical institution. Legacy system 12 can obtain users' health-related information (e.g., health data, analysis results, etc.) from health management system 10, and can also provide corresponding users' medical information to health management system 10.

[0060] For example, medical information can include both clinical and non-clinical information. Clinical information, for example, can include patient type (e.g., pre-operative, post-operative, emergency room patient), demographic information, diagnostic information (e.g., diagnosis name), prescription information, examination results (e.g., blood test results, physical examination results), surgical information, medication history, medical history, body composition analysis information (e.g., information measured using dual-energy X-ray absorptiometry (DEXA) or bioelectrical impedance analysis (BIA) techniques), but is not limited to these. Non-clinical information, for example, can include administrative information, insurance information, etc., but is not limited to these.

[0061] For example, legacy system 12 could be an Electronic Medical Record (EMR), a Hospital Information System (HIS), an Order Communication System (OCS), a Picture Archiving and Communication System (PACS), a Laboratory Information System (LIS), etc., but is not limited to these. For example, within the scope of legacy system 12, it could include a body composition analysis server located on the internal network of a medical institution. The illustrated systems are typically built on a closed architecture or non-standard interfaces, therefore, direct interaction with external systems is often limited.

[0062] like Figure 1 As shown, user terminal 11, health management system 10, etc., can communicate via a network. The network can be any type of wired / wireless network, such as a local area network (LAN), a wide area network (WAN), a mobile radio communication network, or a wireless broadband internet (Wibro).

[0063] Above, refer to Figure 1 An exemplary working environment for a health management system 10 according to several embodiments of the present invention has been described. Hereinafter, reference will be made to... Figure 2 Describe the internal structure and operation of the health management system 10.

[0064] Figure 2 This is an illustrative diagram used to explain the internal structure and operation of the health management system 10 according to several embodiments of the present invention.

[0065] like Figure 2 As shown, in several embodiments, the health management system 10 may include a backend-for-frontend (BFF) server 21 serving the frontend, a rapid medical interoperability resource relay server 22, a gateway server 23, a rapid medical interoperability resource server 24, and an internal integration server 25. The rapid medical interoperability resource server 24, the internal integration server 25, and the legacy system 12 may be located within the medical institution's internal network. For internal network security, the gateway server 23 may be located within a demilitarized zone (DMZ). In this configuration, effective linkage between the health management system 10 and the legacy system 12 can be achieved. The operation of each server is briefly described below.

[0066] The backend server 21 serving the front end refers to a server that provides interface functionality with the health management application installed on the user terminal 11. For example, the backend server 21 serving the front end receives service requests (e.g., application programming interface requests) from the health management application and can send (or transmit / send, etc.) a service response to the health management application, including its processing results. For example, the service request can be executed via an application programming interface call; in this case, the backend server 21 serving the front end can be equivalent to an application programming interface server. However, the scope of the invention is not limited thereto. Later, refer to... Figures 3 to 5 This section describes a specific example of a service request. The backend server 21, which serves the frontend, may transmit (or send) a corresponding request to the Rapid Medical Interoperability Resource Relay Server 22 in order to process a specific service request.

[0067] Next, the Rapid Medical Interoperability Resource Relay Server 22 refers to a server that performs processing based on the Rapid Medical Interoperability Resource Protocol. Specifically, the Rapid Medical Interoperability Resource Relay Server 22 receives service requests from the backend server 21 serving the frontend and can convert them into Rapid Medical Interoperability Resource requests (e.g., application programming interface requests) based on the Rapid Medical Interoperability Resource Protocol. Subsequently, the Rapid Medical Interoperability Resource Relay Server 22 transmits the Rapid Medical Interoperability Resource request to the Rapid Medical Interoperability Resource Server 24 on the internal network through the gateway server 23, and can receive the response to the Rapid Medical Interoperability Resource request (i.e., the Rapid Medical Interoperability Resource response) from the Rapid Medical Interoperability Resource Server 24. Then, the Rapid Medical Interoperability Resource Relay Server 22 can convert the Rapid Medical Interoperability Resource response into a service response format that can be read by the backend server 21 serving the frontend and transmit (or feed back) it to the backend server 21 serving the frontend. The Rapid Medical Interoperability Resource Relay Server 22 can internally act as a Rapid Medical Interoperability Resource client.

[0068] Subsequently, gateway server 23 refers to the server responsible for gateway functions, relaying communications with the outside world while maintaining internal network security. For example, gateway server 23 performs access control on the medical institution's internal network and can transmit (or send) rapid medical interoperability resource requests received from rapid medical interoperability resource relay server 22 to rapid medical interoperability resource server 24. Furthermore, gateway server 23 can transmit (or send back) rapid medical interoperability resource responses received from rapid medical interoperability resource server 24 to rapid medical interoperability resource relay server 22.

[0069] Then, the Rapid Medical Interoperability Resource Server 24 refers to a server used to read and process Rapid Medical Interoperability Resource requests. Specifically, the Rapid Medical Interoperability Resource Server 24 can read and process Rapid Medical Interoperability Resource requests received from the Gateway Server 23, generate its Rapid Medical Interoperability Resource response, and feed it back (or transmit) it to the Gateway Server 23. Depending on the circumstances, the Rapid Medical Interoperability Resource Server 24 may also provide (or share) the user health records included in the Rapid Medical Interoperability Resource request to the internal integration server 25 and / or legacy system 12. Providing health records to a specific server and / or system includes not only the concept of transmitting the corresponding health records, but also the concept of storing the corresponding health records in a database (i.e., a repository) accessible to the corresponding server and / or system.

[0070] Finally, the internal integration server 25 refers to a server that comprehensively analyzes and manages users' health data, medical information / data, etc. For example, the internal integration server 25 can receive users' medical information from the legacy system 12, and can also obtain users' health data from the Rapid Medical Interoperability Resource Server 24 (e.g., obtain health data from the database of the Rapid Medical Interoperability Resource Server 24). The internal integration server 25 can comprehensively manage the health data and medical data of multiple users.

[0071] Furthermore, the internal integration server 25 can provide users and / or healthcare personnel with users' health-related information (e.g., health data, analysis results, etc.) via a control panel. For example, the internal integration server 25 can visually provide users' health data to healthcare personnel utilizing legacy systems (e.g., electronic medical record systems) 12; for this, please refer to [link to relevant documentation]. Figures 6 to 7d The internally integrated server 25 can also provide comprehensive health-related information for multiple users through the control panel.

[0072] Furthermore, the internal integration server 25 uses machine learning models to predict users' health abnormalities at an early stage, and can provide notifications to users and / or healthcare personnel based on its predictions. More details will follow later. Figures 8 to 10 Please provide further explanation.

[0073] Furthermore, the internal integration server 25 can predict the user's success rate for achieving health goals (i.e., the projected success rate) through machine learning models, as will be discussed later. Figure 11 Please provide further explanation.

[0074] The health management system 10 or its structural elements 21, 22, 23, 24, and 25 can be implemented by at least one computing device. For example, all the functions of the internal integration server 25 can be implemented by one computing device, the first function of the internal integration server 25 can be implemented by a first computing device, and the second function can be implemented by a second computing device. Alternatively, specific functions of the internal integration server 25 can also be implemented by multiple computing devices.

[0075] The computing device may include any device with computing (or processing) capabilities, as will be discussed later. Figure 12 An example of the described device is given. A computing device, as an integration of various structural elements (e.g., memory, processor, etc.) interacting with each other, may also be referred to as a "computing system," depending on the context. Of course, the concept of the term "computing system" can also include an integration of multiple computing devices interacting with each other.

[0076] Above, refer to Figure 2 The internal structure and operation of the health management system 10 are described below. (Refer to the following...) Figure 3 The working method of the health management system 10 is described in detail.

[0077] For ease of understanding, the following will be explained... Figure 1 In the environment shown, health management system 10 is used as an example. Figure 2 The structure shown will be explained. For ease of explanation, the health management system 10 will be referred to simply as the "system".

[0078] For example, the working method of system 10 can execute processes such as member registration, prescription information query, and health record submission. The following details the working method of the health management system 10 related to each process.

[0079] Figure 3 The following is an example flowchart illustrating the member registration process of various embodiments of the present invention.

[0080] like Figure 3 As shown, in this process, members of the health management system (or health management service) 10 are registered through user terminal 11.

[0081] As shown in the figure, this process can begin with step S31, where the user terminal 11 receives a member registration application. For example, a user can apply for membership registration through the health management application on terminal 11 and enter personal identification information (e.g., name, date of birth, gender, mobile phone number, etc.).

[0082] It should be understood that steps S32 to S37, as described below, refer to the steps of providing comprehensive health management services by contacting a medical institution when the user is a patient registered with the medical institution.

[0083] In step S32, the user terminal 11 transmits a service request (hereinafter referred to as the "confirmation request") to the backend server 21 serving the front end to confirm whether the corresponding user is a patient registered at the medical institution. For example, the health management application of the user terminal 11 generates a confirmation request including the personal identification information of the corresponding user, which can be transmitted to the backend server 21 serving the front end.

[0084] In step S33, the confirmation request received by the backend server 21 serving the front end is transmitted to the rapid medical interoperability resource relay server 22.

[0085] In step S34, the confirmation request received by the Rapid Medical Interoperability Resource Relay Server 22 is converted based on the Rapid Medical Interoperability Resource Protocol and transmitted to the Gateway Server 23. That is, the Rapid Medical Interoperability Resource Relay Server 22 converts the corresponding confirmation request to generate a Rapid Medical Interoperability Resource Confirmation Request, and transmits it to the Rapid Medical Interoperability Resource Server 24 in the internal network through the Mesh Server 23.

[0086] In step S35, the gateway server 23 transmits a rapid medical interoperability resource confirmation request to the rapid medical interoperability resource server 24.

[0087] In step S36, the rapid medical interoperability resource server 24 interacts with the legacy system (e.g., electronic medical record system) 12 to process rapid medical interoperability resource confirmation requests by verifying whether the user is a registered patient. For example, the rapid medical interoperability resource server 24 can use the personally identifiable information included in the corresponding rapid medical interoperability resource confirmation request to query the patient registration number managed by the legacy system 12. This process can be performed through the application programming interface provided by the legacy system 12, but the scope of the invention is not limited thereto.

[0088] In step S37, the rapid medical interoperability resource server 24 provides a confirmation result. Specifically, the rapid medical interoperability resource server 24 can generate and provide a rapid medical interoperability resource response that includes the confirmation result (e.g., whether the patient is registered, patient registration number, etc.). The corresponding rapid medical interoperability resource response can be transmitted to the rapid medical interoperability resource relay server 22 via the gateway server 23. The rapid medical interoperability resource relay server 22 can convert it into a service response that can be read by the backend server 21 serving the front end and transmit it to the backend server 21 serving the front end. Thus, the backend server 21 serving the front end can transmit it to the user terminal 11.

[0089] In step S38, the user terminal 11 can display (or provide) the result of the member registration application (e.g., success, failure, patient registration number, etc.) to the user.

[0090] Figure 4 The following is an example flowchart illustrating the prescription information query process of various embodiments of the present invention.

[0091] like Figure 4 As shown, in this process, the user queries the processing information managed by the medical institution through terminal 11. This step can be performed when the user is a patient registered with the medical institution.

[0092] As shown in the figure, this process may begin with step S41, where the user terminal 11 receives a prescription information query request. For example, the user may request a prescription information query through the health management application on terminal 11 (e.g., by clicking the query button, etc.).

[0093] In step S42, the user terminal 11 transmits a service request (hereinafter referred to as the "query request") for querying prescription information to the backend server 21 serving the frontend. For example, the health management application of the user terminal 11 generates a query request that includes the patient identification information of the corresponding user (e.g., patient registration number, etc.) and can transmit it to the backend server 21 serving the frontend.

[0094] In step S43, the query request received by the backend server 21 serving the front end is transmitted to the rapid medical interoperability resource relay server 22.

[0095] In step S44, the query request received by the Rapid Medical Interoperability Resource Relay Server 22 is converted based on the Rapid Medical Interoperability Resource Protocol and transmitted to the Gateway Server 24. That is, the Rapid Medical Interoperability Resource Relay Server 22 converts the corresponding query request to generate a Rapid Medical Interoperability Resource Query Request, and transmits it to the Rapid Medical Interoperability Resource Server 24 on the internal network through the Gateway Server 24.

[0096] In step S45, the gateway server 24 transmits a rapid medical interoperability resource query request to the rapid medical interoperability resource server 24.

[0097] In step S46, the rapid medical interoperability resource server 24 interacts with the legacy system (e.g., the electronic medical record system) 12 to process rapid medical interoperability resource query requests by querying user prescription information. For example, the rapid medical interoperability resource server 24 can use the patient identification information included in the corresponding rapid medical interoperability resource query request to query user prescription information managed by the legacy system 12. This process can be executed through the application programming interface provided by the legacy system 12; however, the scope of the invention is not limited thereto.

[0098] In step S47, the rapid medical interoperability resource server 24 returns the query result. Specifically, the rapid medical interoperability resource server 24 can generate and return a rapid medical interoperability resource response including the query result (e.g., success, failure, queried prescription information, etc.). The corresponding rapid medical interoperability resource response can be transmitted to the rapid medical interoperability resource relay server 22 through the gateway server 24. The rapid medical interoperability resource relay server 22 can convert it into a service response that can be read by the backend server 21 serving the front end and transmit it to the backend server 21 serving the front end. Thus, the backend server 21 serving the front end can transmit it to the user terminal 11.

[0099] In step S48, the user terminal 11 displays (or provides) the prescription information query results to the user.

[0100] Figure 5 An exemplary flowchart illustrating the health record submission process of various embodiments of the present invention is provided.

[0101] like Figure 5 As shown, in this process, the user submits a health record through terminal 11 in order to receive health management services. For example, the health record may include information related to the user's practiced (and / or unpracticed) health habits (e.g., habits related to exercise, diet, sleep, etc.) and weight information, but is not limited to these.

[0102] As shown in the figure, this process may begin with step S51, where the user terminal 11 receives the health record. For example, the user can enter the health record and request its submission through the health management application on terminal 11 (e.g., enter the health record and click the submit button, etc.).

[0103] In step S52, the user terminal 11 transmits a service request (hereinafter referred to as the "submission request") for submitting health records to the backend server 21 serving the frontend. For example, the health management application of the user terminal 11 generates a submission request including the corresponding health records entered, and can transmit it to the backend server 21 serving the frontend.

[0104] In step S53, the submission request received by the backend server 21 serving the front end is transmitted to the rapid medical interoperability resource relay server 22.

[0105] In step S54, the submission request received by the Rapid Medical Interoperability Resource Relay Server 22 is converted based on the Rapid Medical Interoperability Resource Protocol and transmitted to the Gateway Server 23. That is, the Rapid Medical Interoperability Resource Relay Server 22 converts the corresponding submission request to generate a Rapid Medical Interoperability Resource Submission Request, which is then transmitted to the Rapid Medical Interoperability Resource Server 24 on the internal network via the Gateway Server 23.

[0106] In step S55, the gateway server 23 transmits a rapid medical interoperability resource submission request to the rapid medical interoperability resource server 24.

[0107] In step S56, the rapid medical interoperability resource server 24 stores health records and processes rapid medical interoperability resource submission requests. For example, the rapid medical interoperability resource server 24 may store the health records included in the corresponding rapid medical interoperability resource submission request in a database accessible to the internal integration server 25. Alternatively, the rapid medical interoperability resource server 24 may store the health records in the corresponding database while simultaneously transmitting them to the internal integration server 25.

[0108] In step S57, the rapid medical interoperability resource server 24 provides feedback on the submission result. Specifically, the rapid medical interoperability resource server 24 can generate and provide a rapid medical interoperability resource response that includes the submission result (e.g., success, failure, etc.). The corresponding rapid medical interoperability resource response can be transmitted to the rapid medical interoperability resource relay server 22 via the gateway server 23. The rapid medical interoperability resource relay server 22 can convert it into a service response in a format accessible to the backend server 21 serving the frontend and transmit it to the backend server 21 serving the frontend. Thus, the backend server 21 serving the frontend can transmit it to the user terminal 11.

[0109] In step S58, the user terminal 11 displays the health record submission result to the user.

[0110] It should be understood that steps S59-1 to S59-3 described later can be performed as needed, without regard to the order.

[0111] In step S59-1, the internal integration server 25 retrieves the user's health records from the Rapid Medical Interoperability Resource Server 24. For example, the internal integration server 25 may periodically or non-periodically retrieve the corresponding user's health records from the repository of the Rapid Medical Interoperability Resource Server 24.

[0112] In step S59-2, legacy system 12 retrieves the user's health records from the Rapid Medical Interoperability Resource Server 24. For example, legacy system 12 may periodically or non-periodically retrieve the corresponding user's health records from the repository of the Rapid Medical Interoperability Resource Server 24.

[0113] In step S59-3, the internal integration server 25 may provide the legacy system 12 with health data (or health-related information including analysis results) of one or more users. For example, the health data may include the user's health records, body composition analysis information, etc., and the body composition analysis information may be obtained from the body composition analysis server included in the legacy system 12. However, the scope of the present invention is not limited thereto.

[0114] For example, in response to a request from a legacy system (e.g., an electronic health record system) 12, the internal integration server 25 can provide a control panel including user health data. As a more specific example, when healthcare professionals request to view a user's (i.e., a registered patient's) health data through the user interface (UI) of the electronic health record system 12 (i.e., the user interface of the electronic health record client), a control panel (i.e., an access control panel accessible through the electronic health record system's user interface) can be provided to the healthcare professional through the corresponding user interface. See below. Figures 6 to 7d Explain the screen structure of the control panel, etc.

[0115] Depending on the situation, the control panel may also be named "Monitoring Interface", "Monitoring Screen", "Monitoring Board", "Monitoring User Interface (UI)", "Monitoring Graphical User Interface (GUI)", etc.

[0116] The control panel can be designed in various forms, and its specific form may vary depending on the implementation. See below for further details. Figures 6 to 7d Provide a specific design example of the control surface.

[0117] On the other hand, the internal integration server 25 can perform a variety of functions for user health management based on user health records obtained (or collected) from the rapid medical interoperability resource server 24 and medical information obtained from the legacy system 12.

[0118] For example, the internal integration server 25 can detect health abnormalities (e.g., drastic weight changes, illness occurrence, etc.) in real time or on a scheduled basis by analyzing health records (e.g., a series of health records). The internal integration server 25 can detect these health abnormalities using machine learning models or other methods.

[0119] As another example, the internal integration server 25 utilizes machine learning models to comprehensively analyze users' health variable data and medical information (e.g., clinical information), enabling early prediction of various information related to health abnormalities. For instance, the internal integration server 25 can use machine learning models to predict the risk (i.e., probability) of a health abnormality occurring within a specific time / period (e.g., 24 hours) from the current point in time, and can also predict the time of occurrence (i.e., the expected time of occurrence), type, cause, and duration of the health abnormality. See later... Figures 8 to 10 This example is explained in detail. For reference, health variable values ​​can be extracted directly from health records or exported by analyzing the corresponding records.

[0120] As another example, when the predicted risk level meets preset conditions (e.g., above a threshold), the internal integration server 25 can provide a notification to the user and / or healthcare personnel (e.g., the healthcare personnel responsible for treating the user). That is, a notification message can be sent to the user and / or healthcare personnel. For example, the notification message can be sent to the corresponding user through a health management application installed on the user terminal 11; however, the scope of the invention is not limited thereto. For example, the notification message may include the type of health abnormality, risk level, expected time of occurrence, cause of occurrence, expected duration, suggested actions, or health management guidelines, but is not limited thereto.

[0121] In the example described, the specific method for providing the notification can be designed in a variety of ways.

[0122] For example, when the predicted risk level exceeds a threshold, the internal integration server 25 can provide an immediate notification.

[0123] As another example, when the predicted risk level exceeds a threshold, the internal integration server 25 can predict the duration of the health anomaly using a machine learning model. Then, if the predicted duration exceeds a baseline value, the internal integration server 25 can provide an immediate notification. Conversely, in the opposite case, the internal integration server 25 can retain the notification.

[0124] As another example, when the predicted risk level exceeds a threshold, the internal integration server 25 retains the notification until the waiting timer expires, monitoring the user's health-related events (e.g., diet, exercise, sleep, medication administration, etc.). Furthermore, if an event increasing health risk (i.e., the risk of health abnormalities) is detected before the waiting timer expires, the internal integration server 25 can immediately provide a notification. Conversely, if an event decreasing health risk is detected, the internal integration server 25 can reset the waiting timer and re-predict the risk level using a machine learning model (i.e., re-perform the prediction based on newly collected health variable data, etc.). This approach not only reduces the inconvenience caused to users by unnecessary frequent notifications but also helps address the problem of declining notification effectiveness. For reference, even for the same event, depending on the user's state, it can be judged as an event increasing health risk or, conversely, as an event decreasing health risk (e.g., if the user is fasting, exercise can be judged as an event increasing health risk). Furthermore, the end time of the waiting timer can be dynamically set. For example, the corresponding end time can be dynamically set based on the predicted risk level of the health abnormality and the expected time of occurrence (for example, if the predicted risk level is low and the expected time of occurrence is relatively long, a longer corresponding end time can be set).

[0125] As another embodiment, when the predicted risk level is above a threshold, the internal integration server 25 can predict the time of occurrence of a health abnormality using a machine learning model. Furthermore, when the difference between the current time point and the predicted time point is less than a baseline value, the internal integration server 25 can immediately provide a notification. When the corresponding difference is above the baseline value, the internal integration server 25 retains the notification until the timer expires, and can monitor the user's health-related events (e.g., diet, exercise, sleep, medication administration, etc.). Subsequently, when an event that reduces health risk (i.e., the risk of health abnormality occurring) is detected (e.g., exercise, medication administration, etc.), the internal integration server 25 can re-predict the risk level using a machine learning model. Conversely, when an event that increases health risk is detected (e.g., high-calorie diet, failure to take medication at a predetermined time, exercise on an empty stomach, etc.), the internal integration server 25 can immediately provide a notification.

[0126] For reference, the term "notification" can also be used interchangeably with expressions such as "reminder," "alarm," and "warning," depending on the context.

[0127] Above, refer to Figures 3 to 5 The operation of system 10 according to several embodiments of the present invention has been described. As described above, system 10, which performs health management, works in conjunction with the legacy system 12 of the medical institution to provide health management services to users in a general or comprehensive manner. Moreover, system 10 and the legacy system 12 independently perform health data processing (e.g., analysis, monitoring, etc.), so that user health-related information can be reflected (or integrated) to the legacy system 12 without affecting its structure or performance.

[0128] Furthermore, by providing users' health-related information to healthcare professionals through the control panel, System 10 can further and more effectively provide users with comprehensive health management services.

[0129] Furthermore, by comprehensively analyzing users' health variable data and clinical information through machine learning models, the risk level of a user developing health abnormalities can be accurately predicted. Moreover, if the risk level exceeds a certain threshold, notifications can be provided to the user and / or healthcare personnel to proactively prevent health abnormalities from occurring.

[0130] The following is for reference Figures 6 to 7d This section describes a design example for control panel 61.

[0131] Figure 6 This is an example diagram illustrating the screen structure of the control panel 61.

[0132] like Figure 6 As shown, control panel 61 may include multiple interfaces (62 to 66 or regions).

[0133] The first interface 62 is for inputting search criteria. For example, search criteria can be the period, user (patient) name, ward number, etc., but are not limited to these. Figure 7a Examples of a first interface 62 and a second interface 63 are shown. The first interface 62 shows the case where period information is entered as a search criterion (see “Period Input Section”). Interfaces 64 and 65 display information that matches the search criteria.

[0134] Next, the second interface 63 is an interface for displaying basic user information (such as personal information). Figure 7a The second interface 62 shows the patient's registration number, user name, date of birth, etc.

[0135] Subsequently, the third interface 64 is used to display health habit information. For example, the third interface 64 can display habit information related to sleep, diet, exercise, drinking, etc. (e.g., time point, consumption time, number / frequency, etc.) and assessment scores based on this, and can also visualize the trend of changes in the assessment scores. Figure 7b The third interface 64 shows the trend of assessment scores based on daily health habit information and the passage of time (refer to the "curve").

[0136] Next, the fourth interface 65 is used to display body information. In addition to the weight information submitted by the user, this body information may also include body composition analysis information, etc. For an example of the fourth interface 65, please refer to [link / reference needed]. Figure 7c . Figure 7c The fourth interface 65 shows the display of daily weight information, daily body composition analysis information, weight change curves, etc.

[0137] Depending on the circumstances, control panel 61 may also include a fifth interface (not shown) for displaying the projected success rate related to health goals set by the user and / or healthcare professionals (e.g., weight loss, visceral fat reduction, etc.). For example, the fifth interface (not shown) may be displayed upon receiving user input to button objects on control panel 61; however, the scope of the invention is not limited thereto. For examples of the fifth interface, please refer to... Figure 7d . Figure 7d The fifth interface shows the trend of the expected success rate based on the daily expected success rate and the change in the expected success rate over time (see “Curve”).

[0138] For example, when the health goal is to lose weight, the expected success rate can be calculated based on weight changes over 14 days, the number of health record submissions, and whether weight has been lost, but it is not limited to these factors.

[0139] In several embodiments, the internal integration server 25 may utilize machine learning models to accurately predict the success rate (i.e., the projected success rate) of health goals, as will be discussed later. Figure 11 Please provide a detailed explanation.

[0140] Above, refer to Figures 6 to 7d A design example of control panel 61 is illustrated. The control panel 61 can visually provide users with a variety of health-related information, and can be effectively used to monitor and manage users' health status.

[0141] The following is for reference Figures 8 to 10 This section describes the instance structure and training method of a machine learning model 80 used to predict health abnormalities.

[0142] Figure 8An illustrative diagram is provided to show the input and output data of a machine learning model 80 according to several embodiments of the present invention.

[0143] like Figure 8 As shown, machine learning model 80 receives users' demographic information, diagnostic information, prescription information, medication history, disease history, and other clinical (or medical) information, as well as health variable data. It can predict and output health abnormalities at an early stage. That is, machine learning model 80 can predict information related to health abnormalities by comprehensively analyzing users' clinical information and health variable data. In this way, when considering the clinical information of individual users, predictions related to health abnormalities can be performed in a more precise and personalized manner.

[0144] Health variable data may include a variety of health variable values, such as weight, body mass index (BMI), waist circumference, body fat percentage, muscle mass, blood pressure, heart rate, blood sugar, sleep duration, meal time, meal frequency, exercise time, exercise frequency, alcohol consumption frequency, smoking frequency, stress index, or combinations thereof, but are not limited to these.

[0145] For example, health abnormality information may include the type of health abnormality (e.g., drastic weight changes, abnormal blood pressure, abnormal blood sugar, occurrence of disease, etc.), whether the health abnormality has occurred, the expected time of occurrence, the duration of the health abnormality, the cause of occurrence, or a combination thereof, but is not limited to these.

[0146] The machine learning model 80 can be designed and implemented in various ways. For example, the machine learning model 80 can be designed and implemented based on various types of models / algorithms such as decision tree, random forest, eXtreme Gradient Boosting (XGBoost), Light Gradient Boosting Machine (LightGBM), Categorical Boosting (CatBoost), Support Vector Machine (SVM), logistic regression, Adaptive Boosting (AdaBoost), and neural network. Its specific type / form / structure can vary depending on the task of the machine learning model 80. Those skilled in the art to which this invention pertains are familiar with the basic structure and training methods of exemplary models; therefore, their description will be omitted.

[0147] The machine learning model 80 can be trained using a dataset of normal individuals and patients (i.e., individuals with health abnormalities) in a supervised learning manner. Those skilled in the art to which this invention pertains are familiar with training based on supervised learning methods; therefore, its description will be omitted. For reference, in supervised learning, the details of the labels may vary depending on the task of the machine learning model 80.

[0148] Figure 9 This is an illustrative diagram used to explain the structure and operation of the machine learning model 80 in several embodiments of the present invention. Figure 9 The structure of a machine learning model 80 provided for performing individual user-specific predictions is shown, and the corresponding model 80 is implemented based on a neural network.

[0149] like Figure 9 As shown, in several embodiments, the machine learning model 80 may include an encoder 91, a variable weighted estimator 92, an interactive weighted estimator 93, an interactive encoder 94, an aggregator 95, and a predictor 96.

[0150] Encoder 91 refers to a neural network module that encodes variable data 97 and generates embeddings. Variable data 97 can consist of various health variable values ​​or be equivalent to time-series data. For example, encoder 91 can receive each health variable value and generate its corresponding variable embedding (e.g., embedding vector).

[0151] Encoder 91 can be implemented as any type / morphology of neural network. For example, encoder 91 can be implemented based on a multi-layer perceptron (MLP) (or a fully-connected layer) or through other types of neural networks.

[0152] Next, the variable weighting estimator 92 is a module that estimates the weighted values ​​assigned to each health variable based on the user's personal characteristic information 98-1. The personal characteristic information 98-1 may include the user's clinical information and other characteristic information. Furthermore, the personal characteristic information 98-1 can be encoded in an appropriate form and embedded in the variable weighting estimator 92. For example, the variable weighting estimator 92 can estimate and output the weighted values ​​(i.e., variable weights) of each health variable specific to the individual user by analyzing the user's clinical information 98-1. It should be understood that the variable weights refer to a vector used to reflect the degree to which each health variable affects health abnormalities varies from user to user.

[0153] Depending on the circumstances, the variable weighting estimator 92 can receive the representative vector (e.g., center vector, weighted average vector, etc.) of the cluster to which the user's personal characteristic information 98-1 belongs. This technique is also applicable to the interactive weighting estimator 93.

[0154] The variable weighted value estimator 92 can be implemented in any form. For example, the variable weighted value estimator 92 can be implemented based on a neural network module such as a multiple perceptron (MLP), or it can be implemented by a module that performs attention operations.

[0155] Next, the interaction weighting estimator 93 is a module that estimates the interaction weights of predetermined pairs of health variables based on the user's personal characteristic information 98-1. For example, the interaction weighting estimator 93 can estimate and output the weighted values ​​(i.e., interaction weights) of each pair of health variables specific to the individual user by analyzing the user's clinical information 98-1. It should be understood that the interaction weights refer to a vector used to reflect the degree of influence of the interaction between health variables on health abnormalities that varies from user to user. The interaction weighting estimator 93 can be implemented in a similar manner to the variable weighting estimator 92.

[0156] Subsequently, the interactive encoder 94 is a module for generating interactive embeddings corresponding to the pairs of health variables. For example, the interactive encoder 94 can generate interactive embeddings corresponding to the predetermined pairs of health variables based on the variable embeddings output by the encoder 91.

[0157] The interactive encoder 94 can be implemented in any form. For example, the interactive encoder 94 can be implemented based on a neural network module such as a multiple perceptron (MLP), or it can be implemented as a module that performs operations such as averaging, concatenation, and element-wise multiplication on the embeddings of two variables.

[0158] Then, aggregator 95 refers to the module that aggregates variable embeddings and interaction embeddings reflecting weighted values ​​to generate a composite embedding. In other words, aggregator 95 can aggregate variable embeddings reflecting variable weighted values ​​and interaction embeddings reflecting interaction weighted values ​​to generate a composite embedding.

[0159] Aggregator 95 can be implemented in any form. For example, aggregation region 95 can be implemented based on neural network modules such as multiple perceptrons (MLP), or it can be implemented in other ways.

[0160] In several embodiments, as shown in the figure, the aggregator 95 may also receive user health-related event information 98-2. The health-related event information 98-2 may be encoded in an appropriate format and embedded in the aggregator 95. In this case, the predictor 96 may consider the user's health-related events (e.g., exercise, sleep, diet, medication administration, etc.) to predict health abnormality information 99.

[0161] Finally, predictor 96 refers to a neural network module that predicts health abnormality-related information 99 based on comprehensive embedding. That is, predictor 96 can predict the health abnormality information 99 of an individual user by comprehensively analyzing various information included in the comprehensive embedding (e.g., health variable values ​​and various clinical information). For example, the predictor 96 can be implemented based on a multiple perceptron (MLP), but the scope of the invention is not limited thereto.

[0162] For example, predictor 96 can output the risk (i.e., probability) of a user having a health abnormality within (or after) a specific time / period (e.g., 6 hours, 12 hours, 1 day, 3 days, etc.) (e.g., outputting a confidence score related to whether a health abnormality occurs), or it can output the type of health abnormality, cause of occurrence, expected time of occurrence, duration, etc. (e.g., outputting confidence scores according to cause of occurrence).

[0163] On the other hand, in several embodiments, a machine learning model 80 can be constructed to further accurately predict health abnormality information using multi-task learning techniques. For such embodiments, the following refers to… Figure 10 Please provide an explanation.

[0164] Figure 10 The diagram illustrates the model training method based on multi-task learning, which is used to explain several embodiments of the present invention. Hereinafter, for the purpose of clearly explaining the present invention, "Predictor A," "Predictor B," etc., will be named "First Predictor," "Second Predictor," etc., and this naming convention also applies to other items using similar names.

[0165] like Figure 10 As shown, system 10 can perform multi-task learning using multiple predictors (e.g., 101, 102) connected to machine learning model 80. Figure 10 The illustration shows an aggregator 95 of a machine learning model 80 connected to two predictors 101 and 102; however, the scope of the invention is not limited thereto. For example, the aggregator 95 may also be connected to a third predictor (not shown). At least one of the predictors 101 and 102 may represent the execution of the target task. Figure 9 Predictor 96.

[0166] The first predictor 101 is a neural network module for predicting first health abnormality information (hereinafter referred to as "first information") 103, and the second predictor 102 is a neural network module for predicting second health abnormality information (hereinafter referred to as "second information") 104 that is different from the first information (refer to examples of the various types of information below). That is, the first predictor 101 and the second predictor 102 perform different tasks based on a comprehensive embedding (i.e., the task of predicting the first information and the task of predicting the second information).

[0167] The types and number of tasks used for multi-task learning can be defined and designed in various ways.

[0168] For example, machine learning model 80 can be trained using a first task to predict the risk of a health abnormality occurring within a specific time period and a second task to predict the cause of the health abnormality (i.e., the first information 103 is the risk of a health abnormality occurring within a specific time period, and the second information is the cause of the health abnormality). Specifically, the parameters of the first predictor 101, aggregator 95, encoders 91, 94, and / or weighted estimators 92, 93 are updated based on the prediction loss of the first task, and the parameters of the second predictor 102, aggregator 95, encoders 91, 94, and / or weighted estimators 92, 93 are updated based on the prediction loss of the second task.

[0169] As another example, the machine learning model 80 can be trained via a first task for predicting the risk of a health abnormality occurring in a first time period and a second task for predicting the risk of a health abnormality occurring in a second time period different from the first time period (i.e., the first information is the risk of a health abnormality occurring in the first time period, and the second information is the risk of a health abnormality occurring in the second time period). As a more specific example, the machine learning model 80 can be trained by performing a first task for predicting the risk of a health abnormality occurring within 24 hours via a first predictor 101 and a second task for predicting the risk of a health abnormality occurring within 48 hours via a second predictor 102.

[0170] As another example, machine learning model 80 can be trained using a first task for predicting the risk of a health abnormality occurring within a specific time period and a second task for predicting the time of occurrence of the health abnormality (i.e., the first information is the risk of a health abnormality occurring within a specific time period, and the second information is the expected time of occurrence of the health abnormality). As a more specific example, machine learning model 80 can be trained by performing a first task, predicting the risk of a health abnormality occurring within 24 hours, via a first predictor 101, and a second task, predicting the time of occurrence of the health abnormality, via a second predictor 102, which is equivalent to a regression head.

[0171] As another example, the machine learning model 80 can be trained by a first task for predicting the risk of health abnormalities occurring within a specific time period and a second task for predicting the number of times (or duration, etc.) health abnormalities occurring within a specific time period (i.e., the first information is the risk of health abnormalities occurring within a specific time period, and the second information is the number of times health abnormalities occur within a specific time period).

[0172] As another example, machine learning models can also be trained using tasks such as predicting the cumulative time of health abnormalities within a specific time period (or time frame), predicting the values ​​of health variables after a specific time, and predicting health-related events that affect health data.

[0173] As yet another example, machine learning model 80 can also be trained based on various combinations of the examples described.

[0174] If the machine learning model 80 is successfully trained, the system 10 can use it to predict users' health abnormalities and provide notifications. In this case, the health abnormality information predicted by the machine learning model 80 can be used to determine whether to provide a notification, the method of notification, and the content of the notification message. For example, the risk level of the health abnormality, the expected time of occurrence, and the duration can be used to determine whether to provide a notification, the method of notification, and / or the health management guidelines included in the notification message.

[0175] In several embodiments, system 10 can derive a health management guideline suitable for the user through machine learning model 80. Specifically, if the predicted risk of a health abnormality is above a threshold, system 10 can modify the health-related event information 98-2 and repeatedly perform risk prediction. Then, system 10 can determine a health management guideline suitable for the user based on the health-related event information 98-2 that minimizes the predicted risk. For example, if the health-related event is "exercise," system 10 can determine "exercise" as the corresponding user's health management guideline if the predicted risk is minimized.

[0176] Above, refer to Figures 8 to 10 The input and output data, illustrative detailed structure, and training methods of the machine learning model 80 according to several embodiments of the present invention are described. As described above, by comprehensively analyzing a user's health variable data and individual characteristic information (e.g., clinical information) through the machine learning model 80, the risk level of a corresponding user developing health abnormalities can be accurately predicted. Moreover, when the risk level is above a threshold, notifications can be provided to the user and / or medical personnel to prevent the user from developing health abnormalities in advance.

[0177] The following is for reference Figure 11 This describes a machine learning model 110 used to predict the success rate of health goals.

[0178] Figure 11 This is an illustrative diagram used to explain the structure and operation of the machine learning model 110 in several embodiments of the present invention.

[0179] like Figure 11 As shown, the machine learning model 110 may include a habit encoder 111, a will encoder 112, an environment encoder 113, an aggregator 114, and a predictor 115.

[0180] The habit encoder 111 refers to a neural network module that encodes the user's health habit information 116 and generates habit embeddings. As described above, for example, the health habit information 116 may include information related to health habits that the user practices (and / or does not practice) (e.g., habits related to exercise, diet, sleep, drinking, etc.) (e.g., time point, consumption time, number / frequency, etc.), or it may be equivalent to time-series data (e.g., daily health habit information, etc.).

[0181] The habit encoder 111 can be implemented using any type / morphology of neural network. For example, the habit encoder 111 can be implemented based on a multiple perceptron, or it can be implemented using other types of neural networks (e.g., recurrent neural networks (RNNs) or the Transformer family of neural networks used for encoding time-series data).

[0182] Next, the will encoder 112 refers to a neural network module that encodes the user's will information 117 and generates a will embedding. Here, the will information 117 refers to information that directly or indirectly represents the user's will to achieve health goals. For example, it may include, but is not limited to, weight changes, the frequency of spontaneous submission of health habit information, the frequency of changes in health goals, the duration of health goal maintenance (e.g., average maintenance period), goal interruption rate, user-input self-motivation scores, or combinations thereof.

[0183] For example, the will encoder 112 can be implemented in a similar manner / form to the habit encoder 111.

[0184] Subsequently, the environmental encoder 113 refers to a neural network module that encodes the user's living environment information 118 and generates an environmental embedding. The living environment information 118 refers to information related to external factors that hinder and / or benefit the achievement of health goals. For example, it may include information related to the sleep environment (e.g., temperature, humidity, noise, etc. of the residence), the exercise environment (e.g., temperature, humidity, distance from the exercise venue, etc.), the external environment (e.g., weather, dust, etc.), the office environment (e.g., work style, office hours, etc.), and the tempting environment (e.g., the density of restaurants, convenience stores, dessert cafes, etc. located near the residence or within the activity radius), but is not limited to these.

[0185] For example, the environmental encoder 113 can be implemented in a similar manner / form to the habit encoder 111.

[0186] Then, aggregator 114 is a module that generates a comprehensive embedding by aggregating habitual embeddings, volitional embeddings, and environmental embeddings. In this case, aggregator 114 can also generate a comprehensive embedding by aggregating health goal information 119-1 set by users and / or healthcare professionals. Health goal information 119-1 can be encoded into an appropriate format and input into aggregator 114. For aggregator 114, please refer to... Figure 9 Aggregator 95.

[0187] In several embodiments, aggregator 114 may also receive personal characteristic information of the user (not shown). This personal characteristic information may be encoded in an appropriate format and input into aggregator 114. In this case, predictor 115 may also consider user characteristics to perform a personalized prediction of success rate 119-2.

[0188] Finally, predictor 115 refers to a neural network module that predicts the success rate 119-2 of a user's health goal based on comprehensive embedding. That is, predictor 15 can predict the success rate 119-2 of the health goal by comprehensively analyzing various information included in the comprehensive embedding (e.g., information related to health habits, willpower, living environment, etc.). For example, the predictor 115 can be implemented based on a multiple perceptron, but the scope of the invention is not limited thereto.

[0189] The machine learning model 110 can be trained using supervised learning. That is, whether a user has previously achieved (or reached) health goals can be used as label information to train the machine learning model 110.

[0190] Above, refer to Figure 11 The example structure and operation of machine learning model 110 for accurately predicting the success rate of health goals are illustrated below. (Refer to the following...) Figure 12 An exemplary computing device 120 is described that can realize the system 10 or its structural elements 21, 22, 23, 24, 25.

[0191] Figure 12 A hardware structure diagram illustrating an example of a computing device 120.

[0192] like Figure 12 As shown, the computing device 120 may include one or more processors 121, a bus 123, a communication interface 124, a memory 122 for loading a computer program 126 executed by the processor 121, and a memory 125 for storing the computer program 126. However, Figure 12Only structural elements relevant to embodiments of the present invention are shown. Therefore, those skilled in the art to which this invention pertains should understand that, in addition to... Figure 12 In addition to the structural elements shown, other general structural elements may also be included. That is, besides... Figure 12 In addition to the structural elements shown, the computing device 120 may also include various other structural elements. Furthermore, depending on the circumstances, the computing device 120 may be omitted. Figure 12 The structural elements shown are some of the structural elements. The following describes each structural element of the computing device 120.

[0193] Processor 121 can control the overall operation of various structures of computing device 120. Processor 121 may include at least one of a central processing unit (CPU), microprocessor (MPU), microcontroller (MCU), graphics processing unit (GPU), neural network processing unit (NPU), tensor processing unit (TPU), vision processing unit (VPU), accelerated processing unit (APU), or any type of processor known in the art to which this invention pertains. Furthermore, processor 121 can perform operations related to at least one application or program for performing a specific task / step / method. Computing device 120 may include more than one processor.

[0194] Next, memory 122 can store various data, instructions, and / or information. Memory 122 can load computer program 126 from memory 125 to perform specific tasks / steps / methods. Memory 122 can be volatile memory such as random access memory; however, the scope of the present invention is not limited thereto.

[0195] Subsequently, bus 123 can provide communication functions between the structural elements of computing device 120. Bus 123 can be of various forms, such as address bus, data bus, and control bus.

[0196] Then, the communication interface 124 can support wired and wireless communication of the computing device 120. To this end, the communication interface 94 may include a communication module known in the art to which this invention pertains (e.g., a radio frequency module that supports wireless communication) and can support multiple communication methods.

[0197] Next, memory 125 may non-temporarily store one or more computer programs 126. Memory 125 may include read-only memory (ROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other non-volatile memory, hard disk, removable disk, or any form of computer-readable recording medium known in the art to which this invention pertains.

[0198] Subsequently, the computer program 126 may include instructions that, when loaded into memory 122, cause the processor 121 to perform specific work / steps / methods. That is, by executing the loaded instructions, the processor 121 can perform specific work / steps / methods.

[0199] For example, computer program 126 may include instructions for performing the following tasks: acquiring a user's health data; and providing the corresponding health data via a control panel. In this case, the internal integrated server 25 of various embodiments of the present invention may be implemented via computing device 120.

[0200] As another example, computer program 126 may include functions for performing the reference... Figures 1 to 11 Instructions for at least a portion of the work / steps / methods described. As shown in the figures, the system 10 and / or its structural elements (21 to 25) of various embodiments of the present invention can be implemented via computing device 120.

[0201] On the other hand, in several embodiments, Figure 12 The computing device 120 shown refers to a virtual machine implemented based on cloud technology. For example, the computing device 120 can be a virtual machine running on one or more physical servers included in a server farm. In this case, Figure 12 At least a portion of the processor 121, memory 122, and memory 125 shown may be virtual hardware, and the communication interface 124 may be a virtual network component such as a virtual switch.

[0202] Above, refer to Figure 12 An exemplary computing device 120 is described, which can implement the system 10 of various embodiments of the present invention.

[0203] The following is for reference Figures 1 to 12This document describes several embodiments of the present invention and their effects. The effects of the present invention are not limited to those mentioned above; those skilled in the art will clearly understand other effects not mentioned through the following description.

[0204] Furthermore, although multiple structural elements are combined into one or more in the above embodiments, this does not mean that the technical concept of the present invention is limited to this embodiment. That is, as long as it is within the scope of the purpose of the technical concept of the present invention, all structural elements can also be selectively combined into more than one for operation.

[0205] The technical concept of the present invention described above can be implemented by computer-readable code on a computer-readable recording medium. The computer program recorded on the computer-readable recording medium can be transmitted to other computing devices via networks such as the Internet and installed on the respective computing devices, thereby becoming usable by those devices.

[0206] Although multiple tasks are shown in a specific order in the accompanying drawings, it should not be construed as meaning that the tasks must be performed in the shown order or sequential order, or that all the shown tasks must be performed to obtain the desired result. In certain situations, multitasking and parallel processing may also be advantageous. While several embodiments of the invention have been described above with reference to the accompanying drawings, those skilled in the art should understand that the technical concept of the invention can also be implemented through other specific embodiments without changing its technical idea or essential features. Therefore, the embodiments described above are merely examples at all levels and should not be construed as limiting. The scope of protection of this invention should be interpreted based on the scope of the invention claims, and all technical ideas within the same scope fall within the scope of the invention claims as defined by the technical concept of this invention.

Claims

1. A health management system, characterized in that, include: A backend server that serves the front end executes the interface functions set in the health management application on the user terminal and receives service requests from the health management application; A rapid medical interoperability resource relay server, based on the rapid medical interoperability resource protocol, transforms the service request and generates a rapid medical interoperability resource request, and transmits the rapid medical interoperability resource request to the medical institution's internal network; and A rapid medical interoperability resource server, located on the internal network, processes requests for the rapid medical interoperability resources. The health management system is linked with the medical institution's existing system. The health management system performs processing based on the user's health data independently of the old system.

2. The health management system according to claim 1, characterized in that, The old system included an electronic medical record system.

3. The health management system according to claim 1, characterized in that, The health management system provides at least a portion of the health data to the legacy system. Receive the user's medical information from the old system.

4. The health management system according to claim 1, characterized in that, The rapid medical interoperability resource relay server transmits the rapid medical interoperability resource request to the rapid medical interoperability resource server through the gateway server, and the gateway server is used to control access to the internal network.

5. The health management system according to claim 1, characterized in that, The service request includes a confirmation request to determine whether the person is a patient registered at the medical institution. The confirmation request includes the user's personal identification information and is executed based on a membership registration application submitted through the health management application. The rapid medical interoperability resource request includes a rapid medical interoperability resource confirmation request corresponding to the confirmation request. The rapid medical interoperability resource server uses the personal identification information to verify whether the user is a registered patient through the legacy system. The Rapid Medical Interoperability Resource Response, including confirmation results, is sent back to the Rapid Medical Interoperability Resource Relay Server.

6. The health management system according to claim 1, characterized in that, The service request includes a request to query prescription information. The query request includes patient identification information registered at the medical institution. The rapid medical interoperability resource request includes a rapid medical interoperability resource query request corresponding to the query request. The rapid medical interoperability resource server uses the patient identification information to obtain the prescription information through the legacy system. The rapid medical interoperability resource response, including the prescription information, is sent back to the rapid medical interoperability resource relay server.

7. The health management system according to claim 1, characterized in that, The service request includes a request to submit health records. The rapid medical interoperability resource request includes a rapid medical interoperability resource submission request corresponding to the submission request. The rapid medical interoperability resource server stores the health records in a database accessible to the legacy system. The Rapid Medical Interoperability Resource Response, including the stored results, is fed back to the Rapid Medical Interoperability Resource Relay Server.

8. The health management system according to claim 1, characterized in that, The health data includes the user's body composition analysis information and the health records submitted by the user through the health management application. The body composition analysis information is obtained from the body composition analysis server included in the legacy system. The health management system also includes an internal integration server that provides visualized health data through a control panel. The control panel can be accessed through the user interface of the electronic medical record system included in the legacy system.

9. The health management system according to claim 8, characterized in that, The health record includes information on health habits and weight. The control panel includes: The first interface displays the user's basic information; The second interface displays the health habit information, and also displays an assessment score calculated based on the health habit information; and The third interface displays body information, including the weight information and the body composition analysis information.

10. The health management system according to claim 8, characterized in that, The internal integration server analyzes the health records to predict the user's success rate in achieving their health goals. The predicted success rate is additionally provided through the control panel.

11. The health management system according to claim 10, characterized in that, The internal integration server predicts the success rate using a machine learning model. The health record includes information on health habits and weight. The machine learning model includes: A habit encoder encodes the health habit information to generate habit embeddings; A will encoder encodes the user's will information to generate a will embedding, the will information including at least one of the user's weight changes, the frequency of spontaneous submission of health habit information, and the frequency of change of health goals; An environment encoder encodes the user's living environment information to generate an environment embedding; An aggregator generates a comprehensive embedding by aggregating the habitual embedding, the will embedding, and the environment embedding; and The predictor predicts the success rate based on the comprehensive embedding.

12. The health management system according to claim 1, characterized in that, It also includes an internally integrated server that uses machine learning models to analyze the health data to predict the user's risk of developing health abnormalities. The machine learning model includes: The encoder generates variable embeddings corresponding to predefined health variables; An interactive encoder generates interactive embeddings corresponding to predetermined health variable pairs based on the variable embeddings. A variable weighting estimator estimates user-specific variable weights for the health variables based on the user's clinical information obtained from the legacy system. An interactive weighted value estimator, based on the clinical information, estimates user-specific interactive weighted values ​​for each pair of health variables; An aggregator generates a composite embedding by aggregating the variable embeddings reflecting the weighted values ​​of the said variables and the interaction embeddings reflecting the interaction weighted values; and The predictor predicts the risk level based on the comprehensive embedding.

13. The health management system according to claim 12, characterized in that, When the predicted risk level exceeds the threshold and the preset waiting time expires, the internal integration server provides a notification related to the health anomaly. Before the waiting timer expires, if an event is detected that increases the risk of the user experiencing the health abnormality, the internal integration server immediately provides the notification. Before the waiting timer expires, if an event is detected that reduces the risk of the user's health abnormality, the internal integration server resets the waiting timer and re-predicts the risk level using the machine learning model.

Images