Medical guideline and ai-based school health response system and method to which virtuous learning cycle is applied

Abstract

A medical guideline and artificial intelligence (AI)-based school health response system to which a virtuous learning cycle is applied, according to the present invention, comprises: an information input unit for receiving basic health information of students and school personnel, guardian contact information, and external institution information; a treatment guide unit for guiding basic medical treatment and AI-based individual specialized treatment on the basis of the input information; a role management unit for assigning role sharing to the school personnel and managing same; an external linking unit for linking in real time with external personnel and institutions according to a determination of the treatment guide unit or the assignment results of the role management unit; a situation response training unit, which manages all response records of actual and training situations so as to optimize the overall system; and a control unit, which feeds back all training results and response log data stored in a database to the treatment guide unit and the role management unit so as to be utilized as useful data.

Description

Medical Guidelines and AI-Based School Health Response System and Method Applying a Virtuous Cycle Learning Structure

[0001] The present invention relates to a school health response system, and more specifically, to a medical guideline and an AI-based school health response system and method that apply a virtuous cycle learning structure to improve medical response capabilities and safety within a school by providing existing medical guidelines and AI-based guidelines and behavioral procedures in stages, enabling all school staff to participate based on the health status of students and staff.

[0002] Generally, educational institutions such as schools face an environment where it is difficult to systematically manage the health of students and staff. In most cases, doctors are not stationed on-site, and health management duties are concentrated among school nurses and a select few staff members, making it difficult to respond consistently to various medical situations. Currently used Electronic Health Record (EHR) systems are primarily focused on managing basic patient information and diagnostic records, and lack real-time support capabilities for non-emergency and chronic conditions that frequently occur in schools, as well as for patients with borderline symptoms falling between emergency and non-emergency statuses.

[0003] Furthermore, the existing system lacks a response framework that considers the roles of students and staff, resulting in limitations where immediate cooperation or division of roles among staff cannot occur in the event of an actual emergency. Consequently, even minor symptoms are concentrated on health managers, such as school nurses, leading to increased workloads; conversely, in the absence of a health manager, unnecessary external transfers occur. On the other hand, since conditions are assessed based on the limited medical knowledge of individual staff members, problems arise where severe symptoms are not recognized early or appropriate treatment is delayed. Even if a role-based response system exists, it is limited to pre-defined, static role assignments; it currently completely lacks intelligent optimization capabilities that assign tasks to the optimal personnel based on real-time locations or objective proficiency verified through training.

[0004] Furthermore, existing school health management systems lack connectivity with external entities such as guardians, 119 emergency services, medical institutions, and pharmacies. Consequently, there are issues where records of a student's health status or treatment measures are not promptly conveyed to guardians, or where follow-up measures, such as telemedicine and pharmacy deliveries, are not smoothly coordinated when necessary.

[0005] Above all, existing technologies lack a virtuous learning cycle that accumulates data on actual response cases or training results and uses this feedback to refine the system's AI judgment or update the proficiency of faculty and staff. This leads to limitations where the system's response capabilities fail to evolve and stagnate, even when the same type of incident is repeated.

[0006] In summary, existing technologies are limited to simple medical record management or emergency response, failing to provide an integrated health management system specialized for the school environment. Accordingly, there is a need for a new system equipped with a role-based response structure optimized using real-time data (location, proficiency) while involving all school staff, a classification function for non-emergency, chronic, and borderline patients utilizing both medical guidelines and AI, a self-evolving capability through a virtuous learning cycle, and real-time connectivity with external organizations.

[0007] The present invention was created to improve upon the problems of the conventional technology described above, and aims to provide a medical guideline and an AI-based school health response system and method that apply a virtuous cycle learning structure, which can comprehensively resolve the limitations of static role division that fails to reflect the real-time location or objective proficiency of school staff, and the absence of a virtuous cycle learning structure in which the system cannot evolve autonomously by receiving feedback on actual response logs.

[0008] Another objective of the present invention is to provide a medical guideline and AI-based school health response system and method that applies a virtuous cycle learning structure, which dynamically optimizes roles by reflecting the real-time location information and proficiency of school staff, maximizes the reliability and speed of response by efficiently allocating response resources across the entire school, simultaneously provides a standard response based on medical guidelines and a customized response considering the patient's underlying conditions, reduces unnecessary external transfer of non-emergency patients, and supports even medical non-experts in performing accurate initial responses.

[0009] Another objective of the present invention is to provide a medical guideline and an AI-based school health response system and method that apply a virtuous cycle learning structure, which goes beyond response within the school to secure the golden time by transmitting emergency data to 119 and medical institutions in advance, transparently share information with guardians in real time, and provide one-stop follow-up measures of remote medical treatment and pharmacy delivery to patients with mild conditions.

[0010] Another objective of the present invention is to provide a medical guideline and an AI-based school health response system and method applying a virtuous cycle learning structure, which can implement a self-evolving optimization system in which the system's judgment accuracy and role assignment efficiency improve as the system is used by feeding back the above actual response and training logs.

[0011] To achieve the above objective, the medical guideline and AI-based school health response system applying a virtuous cycle learning structure according to the present invention is,

[0012] Information input section for receiving basic health information of students and faculty, periodic examination data, job duties and work locations of faculty, guardian contact information, and 119 / external medical institution / pharmacy information;

[0013] A treatment guidance unit that provides guidance on basic medical treatment and AI-based personalized treatment based on information entered through the above information input unit;

[0014] A role management unit that integrates information entered through the information input unit and the judgment of the treatment guidance unit to assign and manage role assignments to faculty and staff;

[0015] An external linkage unit that links in real-time with external personnel and institutions, including guardians, 119, medical institutions, and pharmacies, based on the judgment of the treatment guidance unit or the assignment result of the role management unit;

[0016] A situation response training department that provides medical situation response training programs to faculty and staff and optimizes the entire system by managing all response records from actual and training; and

[0017] It is characterized by including a control unit that controls the status check and operation of the information input unit, treatment guidance unit, role management unit, external linkage unit, and situation response training unit; feeds back all training results and response log data of all actual cases stored in the database (DB) to the treatment guidance unit to be used as re-training data to improve the accuracy of the machine learning model; and feeds back the response log data to the role management unit to be utilized as core data for automatically updating proficiency scores for each staff member based on objective response records.

[0018] Here, the information input unit can perform the role of a gateway that collects, verifies, standardizes, and manages all data necessary for the treatment guidance unit and the role management unit to operate accurately.

[0019] In addition, the information input unit can store the received information in a database (DB) by classifying it into static input data (Pre-registered Data), which is basic data essential for system operation, and dynamic input data (Real-time Incident Data), which is data entered in real-time by primary users, such as school staff, when actual medical situations occur within the school.

[0020] At this time, the static input data (Pre-registered Data) may include basic health information and periodic examination data (EHR) of students and faculty, the job duties, work location (default), and role proficiency (initial value) of faculty, guardian contact information and emergency messenger information, and external contact networks of 119, affiliated medical institutions, and pharmacies.

[0021] At this time, the dynamic input data (Real-time Incident Data) may also be data entered in real-time by school staff, who are the primary users, when an actual medical situation occurs within the school.

[0022] At this time, the information input unit may also be equipped with a function that, when the dynamic input data occurs, automatically queries static input data stored in the database (DB) and merges it in real time to generate an Integrated Situation Data Package.

[0023] In addition, the above information input unit may be equipped with functions for data encryption, access control, log recording, and destruction regarding the patient's basic information, biometric information, and sensitive information, in compliance with the personal information processing policy and relevant laws and regulations provided by the Personal Information Protection Commission.

[0024] In addition, the aforementioned treatment guidance unit analyzes the basically entered lifestyle health records, symptoms, and regular examination data to suggest self-treatment or simple management procedures within the school for non-emergency patients, provide personalized management guidelines for patients with chronic diseases, and for patients on the boundary between emergency and non-emergency patients, re-evaluate the risk level through additional observation or a simple survey to reduce unnecessary external transfers, while simultaneously enabling a rapid transition to external linkage through the aforementioned external linkage unit if necessary.

[0025] In addition, the treatment guidance unit may adopt a hybrid AI structure, and the hybrid AI structure may be configured to include a rule-based engine that immediately presents a standard protocol for a clear emergency situation by complying with predefined medical guidelines, and a personalization engine based on the rule-based engine, in which an artificial intelligence model analyzes the patient's individual health data and the currently input complex situation in real time to supplement and adjust the standard protocol.

[0026] In addition, the above treatment guide can sequentially (step-by-step) generate considerations for each stage and specific tasks to be performed necessary from the pre-hospital stage until the patient does not deteriorate and is ultimately safely treated or medically successful.

[0027] At this time, the generated task can be converted into easy-to-understand medical terms and intuitive action guidelines, taking into account the proficiency level of the primary user receiving the task.

[0028] In addition, the above role management unit can provide a solution suitable for the primary user based on the judgment result of the above treatment guidance unit.

[0029] At this time, if the above-mentioned primary user cannot resolve the issue alone or requires additional support, the above-mentioned role management department can automatically assign secondary response tasks and support roles to the most suitable staff(s) by considering the staff(s)' job characteristics, current location, and health management role proficiency in real time.

[0030] At this time, the assignment algorithm (a type of software program) for selecting the most suitable faculty member and assigning a role may also include: i) a step in which a role management department collects multidimensional data when a secondary response mission or auxiliary role occurs; ii) a step in which a predefined weight is applied based on the collected multidimensional data, or an optimization algorithm is executed to select the faculty member most suitable for the situation that occurred; and iii) a step in which a personalized secondary response mission or auxiliary role is immediately and automatically transmitted to the optimal faculty member selected through the optimization algorithm.

[0031] Here, in step i), the multidimensional data may include real-time location information collected from faculty / staff terminals, job characteristics of each faculty / staff member stored in advance through the information input unit, and role-specific proficiency scores updated in real-time from the response logs of the situation response training unit.

[0032] In addition, the external linkage unit can immediately send an automated first notification through various channels, including the application (App), text message (SMS), and email of a pre-registered guardian, when a medical situation occurs and the role management unit assigns the task of contacting a guardian to a specific staff member.

[0033] In addition, the external linkage unit can provide a one-touch reporting function to the 119 reporting staff member as soon as the treatment guidance unit determines that it is an emergency situation, and at the same time transmit an Emergency Data Package through an API linked to the 119 General Situation Room or the emergency room of the hospital to which the patient will be transferred.

[0034] At this time, the emergency data package may include static data including the patient's basic information, underlying diseases, and allergy information, and dynamic data including the time and place of the accident, initial symptoms, and details of measures taken at the school.

[0035] In addition, the external linkage unit can provide a secure API linkage to a remote medical service platform contracted with the school with the guardian's consent when the treatment guidance unit determines the patient to be mild or borderline and recommends remote medical treatment, and when a prescription is issued after remote medical treatment, immediately transmit the prescription data to the linked pharmacy system to manage follow-up measures leading to the delivery of the prescribed medicine.

[0036] In addition, the aforementioned situation response training department can support repetitive learning so that faculty and staff can utilize the system to respond quickly and accurately in real-life situations.

[0037] In addition, the situation response training program provided to faculty and staff by the aforementioned situation response training department may provide personalized simulations to help individual faculty and staff familiarize themselves with non-emergency / chronic disease management procedures or to master specific auxiliary roles that may be assigned to them by the aforementioned role management department in the future.

[0038] At this time, the above-mentioned situation response training program can also provide virtual scenarios for training role-based cooperative response.

[0039] Here, the above virtual scenario can set the location of patient occurrence and the severity of the situation as variables to induce faculty and staff closest to the location to participate in the training.

[0040] In addition, when group training begins, the above-mentioned situation response training unit can assign virtual primary users, secondary responders, and auxiliary roles to training participants in real time by using the assignment algorithm of the above-mentioned role management unit according to the scenario.

[0041] In addition, to achieve the above objective, the medical guideline and AI-based school health response method applying a virtuous cycle learning structure according to the present invention are,

[0042] As a medical guideline applying a virtuous cycle learning structure and an AI-based school health response method based on a medical guideline applying a virtuous cycle learning structure and an AI-based school health response system,

[0043] a) A step in which the information input unit receives basic health information of students and faculty / staff, periodic examination data, job duties and work locations of faculty / staff, guardian contact information, and 119 / external medical institution / pharmacy information;

[0044] b) A step in which the treatment guidance unit provides guidance on basic medical treatment and AI-based personalized treatment based on information entered through the information input unit;

[0045] c) A step in which the role management department integrates the information entered through the information input department and the judgment of the treatment guidance department to assign role assignments to faculty and staff;

[0046] d) A step in which the external linkage unit links in real-time with external personnel and institutions, including guardians, 119, medical institutions, and pharmacies, based on the judgment of the treatment guidance unit or the assignment result of the role management unit;

[0047] e) A step in which the Situation Response Training Department provides medical situation response training programs to faculty and staff and manages all response records of actual and training to optimize the entire system; and

[0048] f) It is characterized by including a step in which the control unit feeds back all training results and response log data of all actual cases stored in the database (DB) to the treatment guidance unit and role management unit so that they can be utilized as useful data.

[0049] Here, in step a) above, the information input unit can store the received information in a database (DB) by classifying it into static input data (Pre-registered Data), which is base data essential for system operation, and dynamic input data (Real-time Incident Data), which is data entered in real-time by primary users, such as school staff, when an actual medical situation occurs within the school.

[0050] At this time, the static input data (Pre-registered Data) may include basic health information and periodic examination data (EHR) of students and faculty, the job duties, work location (default), and role proficiency (initial value) of faculty, guardian contact information and emergency messenger information, and external contact networks of 119, affiliated medical institutions, and pharmacies.

[0051] At this time, the dynamic input data (Real-time Incident Data) may also be data entered in real-time by school staff, who are the primary users, when an actual medical situation occurs within the school.

[0052] At this time, the information input unit can also generate an Integrated Situation Data Package by automatically querying static input data stored in the database (DB) and merging it in real time when the dynamic input data occurs.

[0053] In addition, in step b) above, the treatment guidance unit basically analyzes the entered lifestyle health records, symptoms, and regular examination data to suggest self-treatment or simple management procedures within the school for non-emergency patients, provide personalized management guidelines for patients with chronic diseases, and for patients on the boundary between emergency and non-emergency patients, re-evaluate the risk level through additional observation or a simple survey to reduce unnecessary external transfer, while simultaneously enabling a rapid transition to external linkage through the external linkage unit if necessary.

[0054] In addition, in step c) above, the role management unit can provide a solution suitable for the primary user based on the judgment result of the treatment guidance unit.

[0055] At this time, if the above-mentioned primary user cannot resolve the issue alone or requires additional support, the above-mentioned role management department can automatically assign secondary response tasks and support roles to the most suitable staff(s) by considering the staff(s)' job characteristics, current location, and health management role proficiency in real time.

[0056] In addition, in step d) above, the external linkage unit can immediately send an automated first notification through various channels, including the application (App), text message (SMS), and email of a pre-registered guardian, when a medical situation occurs and the situation is severe enough for the role management unit to assign a guardian contact task to a specific staff member.

[0057] In addition, in step d) above, as soon as the treatment guidance unit determines that it is an emergency situation, the external linkage unit can provide a one-touch reporting function to the 119 reporting staff member and at the same time transmit an Emergency Data Package through an API linked to the 119 General Situation Room or the emergency room of the hospital to be transferred.

[0058] In addition, the situation response training program provided to the staff by the situation response training department in step e) above may provide personalized simulations for individual staff members to familiarize themselves with non-emergency / chronic disease management procedures or to master specific auxiliary roles that may be assigned to them by the role management department in the future.

[0059] In addition, in step f) above, the control unit may feed back the response log data to the treatment guidance unit to be used as re-training data to improve the accuracy of the machine learning model, and may feed back the response log data to the role management unit to be used as core data for automatically updating proficiency scores for each faculty member based on objective response records.

[0060] According to the present invention, the information input unit automatically merges and packages dynamic data (patient symptoms) entered in real-time by faculty members with static data stored in advance (student's underlying diseases, allergies, etc.) and transmits them to an AI engine. This fundamentally prevents human error, where faculty members may panic in emergency situations and miss the patient's key medical history, and enables the system to make the most rapid and accurate initial judgment by comprehensively considering the patient's underlying diseases.

[0061] Figure 1 is a schematic diagram showing the configuration of a medical guideline and an AI-based school health response system applying a virtuous cycle learning structure according to the present invention.

[0062] Figure 2 is a flowchart illustrating the implementation process of a medical guideline and an AI-based school health response method applying a virtuous cycle learning structure according to the present invention.

[0063] Embodiments of the present invention will be described in detail below with reference to the attached drawings.

[0064] FIG. 1 is a schematic diagram showing the configuration of a medical guideline and an AI-based school health response system applying a virtuous cycle learning structure according to an embodiment of the present invention.

[0065] Referring to FIG. 1, a medical guideline and AI-based school health response system (100) applying a virtuous cycle learning structure according to the present invention may be configured to include an information input unit (110), a treatment guidance unit (120), a role management unit (130), an external linkage unit (140), a situation response training unit (150), and a control unit (160).

[0066] The information input unit (110) receives basic health information of students and faculty, periodic examination data, job duties and work locations of faculty, guardian contact information, and 119 / external medical institution / pharmacy information. Here, the information input unit (110) can serve as a gateway that collects, verifies, standardizes, and manages all data necessary for the treatment guidance unit (120) and the role management unit (130) to operate correctly.

[0067] Additionally, the information input unit (110) can store the received information in a database (DB) (160b) by classifying it into static input data (Pre-registered Data), which is base data essential for system operation, and dynamic input data (Real-time Incident Data), which is data entered in real-time by primary users, such as school staff, when an actual medical situation occurs within the school. At this time, the static input data (Pre-registered Data) may include basic health information and periodic examination data (EHR) of students and school staff, the job duties, work location (default value), and role proficiency (initial value) of school staff, guardian contact information and emergency messenger information, and external contact networks of 119, affiliated medical institutions, and pharmacies. This is entered in advance through text, Excel file uploads, or administrator survey forms, and stored in the database (DB) (160b). The dynamic input data (Real-time Incident Data) may be data entered in real-time by primary users, such as school staff, when an actual medical situation occurs within the school. This can be rapidly entered not only through text input but also through voice input using STT (Speech-to-Text) considering usability in emergency situations, video streaming input that directly transmits the patient's condition (e.g., complexion, wound site), or standardized symptom checklist (questionnaire) forms.

[0068] In addition, the information input unit (110) may be equipped with a function to automatically retrieve static input data (e.g., "Student A, asthma underlying disease, allergy information") stored in the database (DB) (160b) and merge it in real-time to create an Integrated Situation Data Package when dynamic input data (e.g., "Student A, shortness of breath") occurs. Here, this Integrated Situation Data Package is immediately transmitted to the Treatment Guidance Unit (120) and the Role Management Unit (130), serving as the basis for rapid and accurate judgment and role assignment that considers not only simple symptoms but also the patient's underlying disease and the situation of the faculty and staff. Furthermore, the information input unit (110) may be equipped with a function to encrypt data, control access, log recording, and destroy the patient's basic information, biometric information, and sensitive information (including static and dynamic input data) in compliance with the personal information processing policy and relevant laws provided by the Personal Information Protection Commission. Through this, the legality and safety of the system can be secured by receiving periodic diagnoses of the personal information management level.

[0069] The treatment guidance unit (120) provides basic medical treatment and AI-based personalized treatment based on information entered through the information input unit (110). Here, the treatment guidance unit (120) analyzes the basic input lifestyle health records, symptoms, and regular examination data to suggest self-treatment or simple management procedures within the school for non-emergency patients, and provides personalized management guidelines (e.g., medication intake, lifestyle guidance) for patients with chronic diseases. For patients on the borderline between emergency and non-emergency patients, it re-evaluates the risk level through additional observation or a simple survey to reduce unnecessary external transfer, while simultaneously enabling rapid switching to external linkage through the external linkage unit if necessary.

[0070] Additionally, the treatment guidance unit (120) may adopt a hybrid AI structure, and the hybrid AI structure may be configured to include a rule-based engine that immediately presents a standard protocol for clear emergency situations (e.g., cardiac arrest, airway obstruction, anaphylaxis) in compliance with predefined medical guidelines, and a personalization engine based on the rule-based engine, in which an artificial intelligence model analyzes the patient's individual health data (e.g., underlying conditions such as diabetes, heart disease, allergy information) and currently input complex situations (e.g., symptoms, vital signs, etc.) in real time to supplement and adjust the standard protocol.

[0071] In addition, the above-mentioned treatment guide can sequentially generate specific tasks and considerations for each stage necessary to ensure safe treatment or medical care without the patient deteriorating during the pre-hospital phase (e.g., considering whether to administer sugar orally after checking consciousness in the case of a hypoglycemic patient) and specific tasks to be performed step-by-step. At this time, the generated tasks to be performed can be converted into easy-to-understand medical terms and intuitive behavioral guidelines, taking into account the proficiency level of the primary user receiving the task (e.g., school nurse, physical education teacher, general teacher, administrative staff).

[0072] The role management unit (130) integrates the information entered through the information input unit (110) and the judgment of the treatment guidance unit (120) to assign and manage roles to the staff. Here, the role management unit (130) as described above can provide a solution suitable for the primary user based on the judgment result of the treatment guidance unit (120). At this time, if the primary user cannot solve the problem alone or requires additional support, the role management unit (130) can automatically assign secondary response tasks and auxiliary roles to the most suitable staff(s) by considering the staff's job characteristics, current location, and proficiency in health management roles in real time.

[0073] At this time, the assignment algorithm (a type of software program) for selecting the most suitable staff member and assigning a role may also include: i) a step in which the role management unit (130) collects multidimensional data when a secondary response mission or auxiliary role occurs; ii) a step in which a predefined weight is applied based on the collected multidimensional data (e.g., distance from the location of occurrence 50%, proficiency 30%, job 20%, etc.), or an optimization algorithm is performed to select the most suitable staff member for the situation that occurred (e.g., AED request); and iii) a step in which a personalized secondary response mission or auxiliary role is immediately and automatically transmitted to the optimal staff member (one or more) selected through the optimization algorithm. Here, in step i), the multidimensional data may include real-time location information (GPS, WiFi positioning, or indoor beacon-based) collected from a staff terminal (smartphone, etc.), job characteristics of each staff member (health teacher, physical education teacher, administrative staff, etc.) stored in advance through the information input unit (110), and role-specific proficiency scores updated in real-time from the response log of the situation response training unit (150). Through steps i) to iii) as described above, a health teacher can receive assistance from other staff members, and an immediate response system can be implemented that enables the entire staff to cooperate efficiently even when the health teacher is absent.

[0074] The external linkage unit (140) links in real-time with external personnel and institutions, including guardians, 119, medical institutions, and pharmacies, based on the judgment of the treatment guidance unit (120) or the assignment result of the role management unit (130). Here, when a medical situation occurs, the external linkage unit (140) can immediately send an automated first notification through various channels, including the application (App), text message (SMS), and email of a pre-registered guardian, in cases where the situation is severe enough for the role management unit (130) to assign a guardian contact task to a specific staff member. For example, the first notification may include a standardized message such as "Your child is currently having a seizure and 119 has been called, and staff members are providing first aid," and, if necessary, provide a secure channel (e.g., a secure web link or an application screen) through which the guardian can check the student's current condition and the school's action history in real-time.

[0075] In addition, the external linkage unit (140) can provide a one-touch reporting function to the 119 reporting staff member as soon as the treatment guidance unit (120) determines that there is an emergency situation (e.g., need to call 119), and at the same time transmit an Emergency Data Package through an API linked to the 119 General Situation Room or the emergency room of the hospital to which the patient is to be transferred. At this time, the Emergency Data Package may include static data including the patient's basic information, underlying diseases (e.g., epilepsy, type 1 diabetes), and allergy information, and dynamic data including the time and place of the incident, the initial symptoms (e.g., duration of seizure), and details of measures taken at the school (e.g., securing the airway). This contributes to securing the golden time by allowing paramedics or emergency room medical staff to identify clear information in advance before the patient arrives (pre-arrival).

[0076] In addition, the external linkage unit (140) can provide a secure API linkage to a remote medical service platform contracted with the school with the guardian's consent when the treatment guidance unit (120) determines that the patient is mild / borderline (e.g., mild skin rash, temporary abdominal pain) and recommends remote medical treatment. When a prescription is issued after remote medical treatment, the prescription data is immediately transmitted to a linked pharmacy system to manage follow-up measures leading up to the delivery of the prescribed medicine. Through this, the continuity of the student's learning can be ensured by reducing unnecessary hospital visits and early dismissals for patients with mild conditions, and the guardian's peace of mind can be simultaneously secured by transparently sharing all details of the measures with the guardian.

[0077] The situation response training department (150) provides medical situation response training programs to staff and optimizes the entire system by managing all response records of actual and training situations. Here, the situation response training department (150) can support repetitive learning so that staff can use the system to respond quickly and accurately in real-life situations. In addition, the situation response training program provided to staff by the situation response training department (150) can provide personalized simulations to help individual staff members familiarize themselves with non-emergency / chronic disease management procedures or to master specific auxiliary roles (e.g., guardian contact procedures, 119 reporting procedures) that may be assigned to them by the role management department (130) in the future. For example, a homeroom teacher may be provided with a customized simulation for a student with 'epilepsy' in their class. Such simulations present a hypothetical situation, such as "the student exhibits symptoms of epilepsy (seizures, loss of consciousness, biting the tongue, etc.)," ​​and can first enable the homeroom teacher to recognize contraindications, such as "not putting foreign objects in the mouth (even to prevent biting the tongue) or forcibly holding the patient." Subsequently, specific step-by-step procedures can be presented to guide the teacher to sequentially select and learn the correct response steps, such as "removing dangerous objects from the surroundings," "turning the patient to the side to secure the airway," and "measuring the duration of the seizure," and to enable them to execute these actions.

[0078] In this case, the aforementioned situation response training program may also provide virtual scenarios for training role-based cooperative response. Here, these virtual scenarios may set the location of patient occurrence (e.g., classroom, infirmary, playground, cafeteria, etc.) and the severity of the situation as variables, thereby encouraging faculty and staff closest to the relevant location to participate in the training.

[0079] In addition, when group training begins, the above-mentioned situation response training unit (150) can assign virtual primary users, secondary responders, and auxiliary roles (AED request, 119 call) to training participants in real time by using the assignment algorithm of the above-mentioned role management unit (130) according to a scenario (e.g., student cardiac arrest occurring on a sports field). Accordingly, participants can simulate the process of jointly solving a problem by inputting actions on their respective terminals.

[0080] The control unit (160) controls the status check and operation of the information input unit (110), treatment guidance unit (120), role management unit (130), external linkage unit (140), and situation response training unit (150). It feeds back all training results and response logs of all actual cases (e.g., role reception time by staff, action completion time, execution details) stored in the database (DB) (160b) to the treatment guidance unit (120) to be used as re-training data to improve the accuracy of the machine learning model. It also feeds back the response log data to the role management unit (130) to be used as core data for automatically updating proficiency scores by staff based on objective response records. This enables the system of the present invention to implement a virtuous learning cycle that improves its own AI judgment (treatment) and role assignment (management) capabilities as data accumulates through training and actual responses.

[0081] Here, the information input unit (110), treatment guidance unit (120), role management unit (130), external linkage unit (140), situation response training unit (150), control unit (160), and database (DB) (160b) described above may be integrated into one and configured as a computer system.

[0082] Then, below, we will explain the medical guideline applying the virtuous cycle learning structure and the AI-based school health response method based on the medical guideline applying the virtuous cycle learning structure and the AI-based school health response system according to the present invention having the configuration as described above.

[0083] FIG. 2 is a flowchart illustrating the implementation process of a medical guideline and an AI-based school health response method applying a virtuous cycle learning structure according to an embodiment of the present invention.

[0084] Referring to FIGS. 1 and 2, the medical guideline and AI-based school health response method applying a virtuous cycle learning structure according to the present invention is a medical guideline and AI-based school health response method applying a virtuous cycle learning structure based on the medical guideline and AI-based school health response system (100) applying the virtuous cycle learning structure as described above. First, an information input unit (110) receives basic health information of students and staff, regular examination data, the job and work location of staff, guardian contact information, and 119 / external medical institution / pharmacy information (step S201). Here, the information input unit (110) can store the received information in a database (DB) (160b) by classifying it into static input data (Pre-registered Data), which is basic data essential for system operation, and dynamic input data (Real-time Incident Data), which is data entered in real-time by the primary user, the staff, when an actual medical situation occurs within the school. At this time, the static input data (Pre-registered Data) may include basic health information and periodic examination data (EHR) of students and faculty, the job duties, work location (default), and role proficiency (initial value) of faculty, guardian contact information and emergency messenger information, and external contact networks of 119, affiliated medical institutions, and pharmacies. At this time, the dynamic input data (Real-time Incident Data) may also be data entered in real-time by faculty members, who are the primary users, when an actual medical situation occurs within the school. At this time, the information input unit (110) may also automatically retrieve static input data (e.g., "Student A, difficulty breathing") stored in the database (DB) (160b) and merge it in real-time to create an Integrated Situation Data Package when the dynamic input data (e.g., "Student A, difficulty breathing") occurs.

[0085] When information is input through the information input unit (110) as described above, the treatment guidance unit (120) provides basic medical treatment and AI-based personalized treatment based on the information input through the information input unit (110) (step S202). Here, the treatment guidance unit (120) analyzes the basic input lifestyle health records, symptoms, and regular examination data to suggest self-treatment or simple management procedures within the school for non-emergency patients, and provides personalized management guidelines (e.g., medication intake, lifestyle guidance) for patients with chronic diseases. For patients on the border between emergency and non-emergency patients, it re-evaluates the risk level through additional observation or a simple survey to reduce unnecessary external transfer, and if necessary, can quickly switch to external linkage through the external linkage unit (140).

[0086] Afterward, the role management unit (130) combines the information entered through the information input unit (110) and the judgment of the treatment guidance unit (120) to assign roles to the staff (step S203). Here, the role management unit (130) can provide a solution suitable for the primary user based on the judgment result of the treatment guidance unit (120). At this time, if the primary user cannot solve the problem alone or requires additional support, the role management unit (130) can automatically assign secondary response tasks and auxiliary roles to the most suitable staff(s) by considering the staff's job characteristics, current location, and proficiency in health management roles in real time.

[0087] The external linkage unit (140) links in real-time with external personnel and institutions, including guardians, 119, medical institutions, and pharmacies, based on the judgment of the treatment guidance unit (120) or the assignment result of the role management unit (130) (step S204). Here, when a medical situation occurs, the external linkage unit (140) can immediately send an automated first notification through various channels, including the application (App), text message (SMS), and email of a pre-registered guardian, if the situation is severe enough for the role management unit (130) to assign a guardian contact task to a specific staff member. Additionally, as soon as the treatment guidance unit (120) determines that there is an emergency situation (e.g., need to call 119), the external linkage unit (140) can provide a one-touch reporting function to the staff member in charge of reporting to 119, and at the same time transmit an Emergency Data Package through an API linked to the 119 General Situation Room or the emergency room of the hospital to which the patient will be transferred.

[0088] Meanwhile, the situation response training department (150) provides medical situation response training programs to staff members and manages all response records of actual and training to optimize the entire system (step S205). Here, the situation response training program provided to staff members by the situation response training department (150) may provide personalized simulations to help individual staff members familiarize themselves with non-emergency / chronic disease management procedures or to master specific auxiliary roles (e.g., guardian contact procedures, 119 reporting procedures) that may be assigned to them by the role management department (130) in the future.

[0089] Finally, the control unit (160) feeds back data of all training results and response logs of all actual cases (e.g., time of role reception by staff, time of action completion, details of execution) stored in the database (DB) (160b) to the treatment guidance unit (120) and the role management unit (130) so that they can be utilized as useful data (step S206). Here, the control unit (160) feeds back the response log data to the treatment guidance unit (120) so that it is used as re-training data to improve the accuracy of the machine learning model, and feeds back the response log data to the role management unit (130) so that it is utilized as core data to automatically update the proficiency score of each staff member based on objective response records.

[0090] As described above, the medical guideline and AI-based school health response system and method applying the virtuous cycle learning structure according to the present invention have the advantage of fundamentally preventing human error, where a staff member may panic and miss a patient's key medical history in an emergency situation, by automatically merging and packaging dynamic data (patient symptoms) entered in real-time by a staff member and static data (student's underlying diseases, allergies, etc.) stored in advance and transmitting them to an AI engine, and enabling the system to make the most rapid and accurate initial judgment by comprehensively considering the patient's underlying diseases.

[0091] Furthermore, the treatment guidance unit utilizes a hybrid AI structure where a rule-based engine ensures a reliable, standardized response in clear emergency situations, while a personalization engine analyzes complex scenarios, such as underlying conditions, to provide accurate, customized judgments that reduce unnecessary external transfers of borderline patients. Additionally, by converting generated guidelines into easy-to-understand language tailored to the staff's proficiency, it offers the advantage of enabling even staff without medical knowledge to confidently perform accurate initial responses.

[0092] In addition, the role management department collects multidimensional data including real-time location information of school staff and proficiency scores based on training logs, and analyzes this using an optimization algorithm. By automatically and in parallel assigning secondary response tasks (e.g., AED) to the nearest and most skilled staff member, and auxiliary roles such as communicating with guardians or coordinating with 119 to other staff members, the quality of response can be improved so that the primary responder can focus solely on patient treatment. As a result, the workload of school nurses is reduced, and rapid and efficient cooperative response is possible without any gaps in response even when school nurses are absent.

[0093] In addition, the external linkage unit can transmit an emergency data package containing the patient's underlying conditions and on-site measures to 119 or medical institutions before arrival at the hospital, thereby supporting medical staff in preparing customized treatment in advance to secure the golden time. It also provides automated first-line notifications and real-time situation sharing channels to guardians, which can alleviate their anxiety through immediate information sharing and increase the reliability of the school's response. Furthermore, for patients with mild conditions, it offers an integrated linkage of remote medical care and pharmacy delivery, which can reduce unnecessary hospital visits and ensure the continuity of students' learning.

[0094] Furthermore, the Situation Response Training Department can enhance the practical cooperative response capabilities of staff through group training utilizing algorithms from the Actual Role Management Department. Additionally, all training and actual response log data can be fed back and used as retraining data to improve the accuracy of the Treatment Guidance Department, and can be utilized as material to automatically update the Role Management Department's proficiency scores based on objective records. Consequently, this system offers the advantage of implementing a virtuous cycle learning structure where the AI's judgment (treatment) and role assignment (management) evolve and optimize themselves as the system is used.

[0095] In addition, it has the advantage of providing an intelligent response network equipped with a Virtuous Learning Cycle, where the AI's judgment (treatment) and role assignment (management) capabilities evolve and optimize as actual response and training logs accumulate within the school.

Claims

1. An information input section for receiving basic health information of students and faculty, periodic examination data, job duties and work locations of faculty, guardian contact information, and 119 / external medical institution / pharmacy information; A treatment guidance unit that provides guidance on basic medical treatment and AI-based personalized treatment based on information entered through the above information input unit; A role management unit that integrates information entered through the information input unit and the judgment of the treatment guidance unit to assign and manage role assignments to faculty and staff; An external linkage unit that links in real-time with external personnel and institutions, including guardians, 119, medical institutions, and pharmacies, based on the judgment of the treatment guidance unit or the assignment result of the role management unit; A situation response training department that provides medical situation response training programs to faculty and staff and optimizes the entire system by managing all response records from actual and training; and A medical guideline and AI-based school health response system applying a virtuous cycle learning structure, comprising a control unit that controls the status check and operation of the information input unit, treatment guidance unit, role management unit, external linkage unit, and situation response training unit, feeds back all training results and response log data of all actual cases stored in the database (DB) to the treatment guidance unit to be used as re-training data to improve the accuracy of the machine learning model, and feeds back the response log data to the role management unit to be used as core data for automatically updating proficiency scores for each staff member based on objective response records.

2. In Paragraph 1, A medical guideline and AI-based school health response system applying a virtuous cycle learning structure, characterized in that the information input unit above performs the role of a gateway that collects, verifies, standardizes, and manages all data necessary for the treatment guidance unit and role management unit to operate accurately.

3. In Paragraph 1, The above-described information input unit is characterized by classifying the received information into static input data (Pre-registered Data), which is base data essential for system operation, and dynamic input data (Real-time Incident Data), which is data entered in real-time by primary users, such as school staff, when actual medical situations occur within the school, and storing these in a database (DB), thereby applying a virtuous cycle learning structure to the medical guideline and AI-based school health response system.

4. In Paragraph 3, The above static input data (Pre-registered Data) is a medical guideline and AI-based school health response system applying a virtuous cycle learning structure that includes basic health information and periodic examination data (EHR) of students and faculty, the job duties, work locations (default), and role proficiency (initial value) of faculty, guardian contact information and emergency messenger information, and external agency contact networks of 119, affiliated medical institutions, and pharmacies.

5. In Paragraph 3, A medical guideline and AI-based school health response system applying a virtuous cycle learning structure, characterized in that the above dynamic input data (Real-time Incident Data) is data entered in real-time by school staff, who are the primary users, when an actual medical situation occurs within the school.

6. In Paragraph 3, A medical guideline and AI-based school health response system applying a virtuous cycle learning structure, characterized in that the information input unit has the function of automatically querying static input data stored in a database (DB) and merging it in real time to generate an Integrated Situation Data Package when the dynamic input data occurs.

7. In Paragraph 1, A medical guideline and AI-based school health response system applying a virtuous cycle learning structure, characterized in that the above-mentioned information input unit is equipped with functions for data encryption, access control, log recording, and destruction of basic patient information, biometric information, and sensitive information in compliance with the personal information processing policy and relevant laws and regulations provided by the Personal Information Protection Commission.

8. In Paragraph 1, The above-mentioned treatment guidance unit basically analyzes inputted lifestyle health records, symptoms, and regular examination data to suggest self-treatment or simple management procedures within the school for non-emergency patients, provide personalized management guidelines for patients with chronic diseases, and for patients on the boundary between emergency and non-emergency patients, re-evaluate risk levels through additional observation or simple surveys to reduce unnecessary external transfers, while simultaneously rapidly switching to external linkage through the above-mentioned external linkage unit when necessary, thereby applying a virtuous cycle learning structure to the medical guideline and AI-based school health response system.

9. In Paragraph 1, The above-mentioned treatment guidance unit adopts a hybrid AI structure, and the said hybrid AI structure is characterized by comprising a rule-based engine that immediately presents a standard protocol for a clear emergency situation by complying with predefined medical guidelines, and a personalization engine based on said rule-based engine that supplements and adjusts the standard protocol by having an artificial intelligence model analyze the patient's individual health data and the currently input complex situation in real time. This describes a medical guideline and AI-based school health response system applying a virtuous cycle learning structure.

10. In Paragraph 1, The above-mentioned treatment guidance unit is a medical guideline and AI-based school health response system applying a virtuous cycle learning structure, characterized by sequentially (step-by-step) generating considerations and specific tasks to be performed for each stage necessary to ensure that the patient does not deteriorate and is ultimately safely treated or medically successful during the pre-hospital stage.

11. In Paragraph 10, A medical guideline and AI-based school health response system applying a virtuous cycle learning structure, characterized in that the generated task is converted into easy-to-understand medical terms and intuitive behavioral guidelines, taking into account the proficiency level of the primary user receiving the task.

12. In Paragraph 1, A medical guideline and AI-based school health response system applying a virtuous cycle learning structure, characterized in that the above-mentioned role management unit provides a solution suitable for the primary user based on the judgment result of the above-mentioned treatment guidance unit.

13. In Paragraph 12, A medical guideline and AI-based school health response system applying a virtuous cycle learning structure, characterized in that when the above-mentioned primary user cannot resolve the issue alone or requires additional support, the above-mentioned role management unit automatically assigns secondary response tasks and support roles to the most suitable staff(s) by considering the staff(s)' job characteristics, current location, and health management role proficiency in real time.

14. In Paragraph 13, The assignment algorithm (a type of software program) for selecting the most suitable faculty and staff and assigning roles mentioned above is, i) A step in which the Role Management Department collects multidimensional data when a secondary response mission or auxiliary role occurs; ii) a step of applying predefined weights based on the above-mentioned collected multidimensional data or performing an optimization algorithm to select the most suitable faculty member for the situation that occurred; and iii) A medical guideline and AI-based school health response system applying a virtuous cycle learning structure that includes a step in which a personalized secondary response mission or auxiliary role is immediately and automatically transmitted to the optimal staff member selected through the above optimization algorithm.

15. In Paragraph 14, Medical guideline and AI-based school health response system applying a virtuous cycle learning structure, wherein the multidimensional data in step i) includes real-time location information collected from faculty / staff terminals, job characteristics of each faculty / staff member stored in advance through the information input unit, and role-specific proficiency scores updated in real-time from the response logs of the situation response training unit.

16. In Paragraph 1, A medical guideline and AI-based school health response system applying a virtuous cycle learning structure, characterized in that, when a medical situation occurs, the external linkage unit immediately sends an automated first notification through various channels including the application (App), text message (SMS), and email of a pre-registered guardian when the situation is severe enough for the role management unit to assign a guardian contact task to a specific school staff member.

17. In Paragraph 1, A medical guideline and AI-based school health response system applying a virtuous cycle learning structure, characterized in that the above-mentioned external linkage unit provides a one-touch reporting function to a school staff member in charge of 119 reporting immediately upon the above-mentioned treatment guidance unit determining an emergency situation, and simultaneously transmits an Emergency Data Package through an API linked to the 119 General Situation Room or the emergency room of the hospital to which the patient is to be transferred.

18. In Paragraph 17, The above emergency data package is a medical guideline and AI-based school health response system that applies a virtuous cycle learning structure including static data, such as patient basic information, underlying diseases, and allergy information, and dynamic data, such as the time and place of the incident, initial symptoms, and details of measures taken at the school.

19. In Paragraph 1, The above-mentioned external linkage unit is characterized by providing a secure API linkage to a remote medical service platform contracted with the school with the guardian's consent when the above-mentioned treatment guidance unit determines the patient to be mild / borderline and recommends remote medical treatment, and immediately transmitting the prescription data to a linked pharmacy system when a prescription is issued after remote medical treatment, thereby managing follow-up measures to lead to the delivery of the prescribed medicine, and is an AI-based school health response system applying a virtuous cycle learning structure.

20. In Paragraph 1, The above-mentioned situation response training department applies a virtuous cycle learning structure to support iterative learning so that school staff can utilize the system to respond quickly and accurately in real-life situations, and is a medical guideline and AI-based school health response system.

21. In Paragraph 1, Medical guidelines and an AI-based school health response system applying a virtuous cycle learning structure, characterized by the fact that the situation response training program provided to school staff by the above-mentioned situation response training department provides personalized simulations for individual school staff to familiarize themselves with non-emergency / chronic disease management procedures or to master specific auxiliary roles that may be assigned to them by the above-mentioned role management department in the future.

22. In Paragraph 21, The above situation response training program is a medical guideline and AI-based school health response system applying a virtuous cycle learning structure characterized by providing virtual scenarios for training role-based cooperative response.

23. In Paragraph 22, The above-mentioned virtual scenario is a medical guideline and AI-based school health response system applying a virtuous cycle learning structure characterized by setting the location of patient occurrence and the severity of the situation as variables to induce school staff closest to the location to participate in training.

24. In Paragraph 1, A medical guideline and AI-based school health response system applying a virtuous cycle learning structure, characterized in that when group training begins, the above-mentioned situation response training unit assigns virtual primary users, secondary responders, and support roles to training participants in real time according to the scenario, using the assignment algorithm of the above-mentioned role management unit as is.

25. As a medical guideline applying a virtuous cycle learning structure and an AI-based school health response method based on a medical guideline applying a virtuous cycle learning structure and an AI-based school health response system, a) A step in which the information input unit receives basic health information of students and faculty / staff, periodic examination data, job duties and work locations of faculty / staff, guardian contact information, and 119 / external medical institution / pharmacy information; b) A step in which the treatment guidance unit provides guidance on basic medical treatment and AI-based personalized treatment based on information entered through the information input unit; c) A step in which the role management department integrates the information entered through the information input department and the judgment of the treatment guidance department to assign role assignments to faculty and staff; d) A step in which the external linkage unit links in real-time with external personnel and institutions, including guardians, 119, medical institutions, and pharmacies, based on the judgment of the treatment guidance unit or the assignment result of the role management unit; e) A step in which the Situation Response Training Department provides medical situation response training programs to faculty and staff and manages all response records of actual and training to optimize the entire system; and f) A medical guideline and AI-based school health response method applying a virtuous cycle learning structure, comprising the step of a control unit feeding back all training results and response log data of all actual cases stored in a database (DB) to the treatment guidance unit and role management unit so that they can be utilized as useful data.

26. In Paragraph 25, Medical guideline and AI-based school health response method applying a virtuous cycle learning structure, characterized in that in step a) above, the information input unit distinguishes the received information into static input data (Pre-registered Data), which is base data essential for system operation, and dynamic input data (Real-time Incident Data), which is data entered in real-time by primary users, namely school staff, when an actual medical situation occurs within the school, and stores them in a database (DB).

27. In Paragraph 26, The above static input data (Pre-registered Data) includes basic health information and periodic examination data (EHR) of students and faculty, the job duties, work location (default), and role proficiency (initial value) of faculty, guardian contact information and emergency messenger information, and a medical guideline and AI-based school health response method applying a virtuous cycle learning structure including external agency contact networks of 119, affiliated medical institutions, and pharmacies.

28. In Paragraph 26, Medical guidelines and AI-based school health response methods applying a virtuous cycle learning structure, characterized in that the above dynamic input data (Real-time Incident Data) is data entered in real-time by school staff, who are the primary users, when an actual medical situation occurs within the school.

29. In Paragraph 26, A medical guideline and AI-based school health response method applying a virtuous cycle learning structure, characterized in that the above information input unit automatically queries static input data stored in a database (DB) and merges it in real time to generate an Integrated Situation Data Package when the above dynamic input data occurs.

30. In Paragraph 25, A medical guideline and AI-based school health response method applying a virtuous cycle learning structure, characterized in that in step b) above, the treatment guidance unit basically analyzes inputted lifestyle health records, symptoms, and regular examination data to suggest self-treatment or simple management procedures within the school for non-emergency patients, provide personalized management guidelines for patients with chronic diseases, and for patients on the boundary between emergency and non-emergency patients, re-evaluate the risk level through additional observation or a simple survey to reduce unnecessary external transfer, while simultaneously rapidly switching to external linkage through the external linkage unit if necessary.

31. In Paragraph 25, A medical guideline and AI-based school health response method applying a virtuous cycle learning structure, characterized in that in step c) above, the role management unit provides a solution suitable for the primary user according to the judgment result of the treatment guidance unit.

32. In Paragraph 31, Medical guidelines and AI-based school health response methods applying a virtuous cycle learning structure, characterized in that when the above-mentioned primary user cannot resolve the issue alone or requires additional support, the above-mentioned role management unit automatically assigns secondary response tasks and support roles to the most suitable staff(s) by considering the staff(s)'s job characteristics, current location, and health management role proficiency in real time.

33. In Paragraph 25, Medical guideline and AI-based school health response method applying a virtuous cycle learning structure, characterized in that in step d) above, the external linkage unit immediately sends an automated first notification through various channels including the pre-registered guardian's application (App), text message (SMS), and email when a medical situation occurs and the situation is severe enough for the role management unit to assign a guardian contact task to a specific school staff member.

34. In Paragraph 25, Medical guideline and AI-based school health response method applying a virtuous cycle learning structure, characterized in that, in step d) above, the external linkage unit provides a one-touch reporting function to the school staff member in charge of 119 reporting as soon as the treatment guidance unit determines that it is an emergency situation, and at the same time transmits an Emergency Data Package through an API linked to the 119 General Situation Room or the emergency room of the hospital to which the patient is to be transferred.

35. In Paragraph 25, Medical guidelines and AI-based school health response methods applying a virtuous cycle learning structure, characterized in that the situation response training program provided to school staff by the situation response training department in step e) above provides personalized simulations for individual school staff to familiarize themselves with non-emergency / chronic disease management procedures or to master specific auxiliary roles that may be assigned to them by the role management department in the future.

36. In Paragraph 25, A medical guideline and an AI-based school health response method applying a virtuous cycle learning structure, characterized in that, in step f) above, the control unit feeds the response log data to the treatment guidance unit to be used as re-training data to improve the accuracy of the machine learning model, and feeds the response log data to the role management unit to be used as core data for automatically updating proficiency scores for each staff member based on objective response records.

Images