Method, device, and computer program for providing medical guidance by using generative artificial intelligence-based medical staff model

Abstract

Provided are a method, device, and computer program for providing medical guidance by using a generative artificial intelligence-based medical staff model. According to various embodiments of the present disclosure, provided is a method, performed by a computing device, for providing medical guidance by using a generative artificial intelligence-based medical staff model. The method comprises the steps of: acquiring clinical data generated in a process in which a medical staff member treats a patient; selecting one or more medical staff models from among a plurality of medical staff models on the basis of the acquired clinical data; deriving result data according to analyzing the acquired clinical data by means of the selected one or more medical staff models; and providing medical guidance information to the medical staff by using the derived result data.

Description

Method, device, and computer program for providing medical guidance using a generative AI-based medical staff model

[0001] Various embodiments of the present disclosure relate to a method, apparatus, and computer program for providing medical guidance using a generative artificial intelligence-based medical staff model.

[0002] Outpatient care refers to the process in which a patient visits a hospital to receive medical treatment from a healthcare professional. Generally, the outpatient process proceeds in the order of a patient visiting the hospital, completing registration procedures, waiting, and then meeting with a doctor. The doctor listens to the patient's medical history, performs necessary physical examinations, and, if necessary, orders additional tests or prescribes medication. During this process, the doctor comprehensively analyzes the patient's symptoms, medical history, and test results to make a diagnosis and establish a treatment plan.

[0003] However, this outpatient consultation process presents several inconveniences. First, doctors must rapidly grasp and analyze a large amount of patient information within a limited timeframe. Particularly when consultation times are short, there is a risk that a diagnosis may be made without sufficiently reviewing the patient's medical history or previous test results. Second, it is difficult to systematically record and manage various symptoms when a patient complains of them. This can act as an obstacle to doctors accurately assessing the patient's condition and determining appropriate treatment. Third, selecting the most suitable option for the patient from among various tests and treatment choices can be time-consuming.

[0004] To address these inconveniences, various medical information systems and Electronic Health Record (EHR) systems are currently being introduced. These systems digitize patients' medical records, allowing doctors to easily access and utilize them. Furthermore, some systems analyze patients' medical records to provide doctors with useful information. For example, they offer features such as presenting diagnostic guidelines for specific symptoms or automatically analyzing test results to notify doctors of abnormal findings.

[0005] However, existing medical information systems and EHR systems also have limitations. First, many systems are limited to simply storing and searching data. Consequently, they often fail to provide information that is actually helpful to doctors during consultations. Second, some systems are overly complex and inconvenient to use, or they may provide an excessive amount of information that could actually confuse doctors' judgments. Third, compatibility issues between various systems can make it difficult to share consistent information across multiple hospitals or clinics.

[0006] The aforementioned background technology is one that the inventor possessed or acquired in the process of deriving the content of the present disclosure, and it cannot be considered as prior art disclosed to the general public prior to the filing of this application.

[0007] The problem that the present disclosure aims to solve is to provide a method, apparatus, and computer program for providing medical guidance using a generative AI-based medical staff model, which derives result data by analyzing medical data generated during the process of medical staff conducting outpatient treatment for a patient through a generative AI-based medical staff model, and provides medical guidance information to the medical staff using this data to assist in the decision-making of the medical staff and thereby enable the medical staff to perform outpatient treatment more easily.

[0008] The problem that the present disclosure aims to solve is to provide a method, apparatus, and computer program for providing medical guidance using a generative AI-based medical staff model, which enables securing highly reliable result data by constructing multiple medical staff models for each medical staff in advance, selecting the optimal medical staff model among the multiple medical staff models, and deriving result data through this.

[0009] The problems that this disclosure aims to solve are not limited to those mentioned above, and other unmentioned problems will be clearly understood by a person skilled in the art from the description below.

[0010] A method for providing medical guidance using a generative artificial intelligence-based medical staff model according to one embodiment of the present disclosure for solving the above-mentioned problem may include, in a method performed by a computing device, the step of acquiring medical data generated during the process of a medical staff member providing medical care to a patient; the step of selecting one or more medical staff models from a plurality of medical staff models based on the acquired medical data; the step of deriving result data by analyzing the acquired medical data through the selected one or more medical staff models; and the step of providing medical guidance information to the medical staff member using the derived result data.

[0011] In various embodiments, the plurality of medical staff models may be a Generative Artificial Intelligence Model trained using multiple medical data generated during the process of each of the multiple medical staff treating a patient and result data derived from each of the multiple medical data as training data.

[0012] In various embodiments, the step of selecting one or more medical staff models may include the step of extracting disease information about the patient from the acquired medical data and the step of selecting one or more medical staff models among the plurality of medical staff models based on the extracted disease information.

[0013] In various embodiments, the step of selecting one or more medical staff models may include the step of extracting one or more features related to the patient from the acquired medical data, the step of selecting one or more past patients having characteristics similar to the patient among a plurality of past patients based on the extracted one or more features, and the step of selecting a medical staff model trained based on the medical data of the selected one or more past patients among the plurality of medical staff models.

[0014] In various embodiments, the step of selecting one or more medical staff models may include the step of calculating a score for each of the plurality of medical staff models based on feedback obtained corresponding to each of the plurality of medical staff models during a predetermined period, and the step of selecting one or more of the plurality of medical staff models based on the calculated score.

[0015] In various embodiments, the feedback may be a response obtained from a plurality of medical staff corresponding to each of the plurality of medical staff models by providing a plurality of result data derived by inputting a plurality of medical staff models obtained during a predetermined period into the plurality of medical staff models, and may include evaluation information regarding the plurality of result data and information regarding whether to modify the plurality of result data.

[0016] In various embodiments, each of the plurality of medical staff models includes a medical chart generation model, a diagnosis model, a test recommendation model, and a treatment recommendation model, and the step of selecting one or more medical staff models may include the step of individually selecting one or more medical chart generation models, one or more diagnosis models, one or more test recommendation models, and one or more treatment recommendation models from the plurality of medical staff models based on the acquired medical data.

[0017] In various embodiments, the step of selecting one or more medical staff models may include the step of determining the type of service to be provided to the medical staff based on the type of acquired medical data, and the step of selecting one or more medical staff models among the plurality of medical staff models based on the determined type.

[0018] In various embodiments, the step of selecting one or more medical staff models includes the step of selecting one main medical staff model and one or more sub-medical staff models based on the acquired medical data, and the step of deriving result data may include the step of deriving first result data by analyzing the acquired medical data through the selected one main medical staff model, the step of deriving one or more second result data by analyzing the acquired medical data through the one or more sub-medical staff models, and the step of verifying the derived first result data using the derived one or more second result data.

[0019] A computing device for performing a method of providing medical guidance using a generative artificial intelligence-based medical staff model according to another embodiment of the present disclosure for solving the above-described problem comprises a processor, a network interface, a memory, and a computer program loaded into the memory and executed by the processor, wherein the computer program may include an instruction for acquiring medical data generated during the process of a medical staff member performing medical treatment on a patient, an instruction for selecting one or more medical staff models among a plurality of medical staff models based on the acquired medical data, an instruction for deriving result data by analyzing the acquired medical data through the selected one or more medical staff models, and an instruction for providing medical guidance information to the medical staff member using the derived result data.

[0020] A computer program according to another embodiment of the present disclosure for solving the above-described problem may be combined with a computing device and stored on a recording medium readable by the computing device to execute a method for providing medical guidance using a generative artificial intelligence-based medical staff model, comprising the steps of: acquiring medical data generated during the process of a medical staff member performing medical treatment on a patient; selecting one or more medical staff models from a plurality of medical staff models based on the acquired medical data; deriving result data by analyzing the acquired medical data through the selected one or more medical staff models; and providing medical guidance information to the medical staff member using the derived result data.

[0021] Other specific details of the present disclosure are included in the detailed description and drawings.

[0022] In addition, a method for providing an outpatient medical assistance service based on generative artificial intelligence according to one embodiment of the present disclosure may include, in a method performed by a computing device, a step of acquiring medical data generated during the process of medical staff conducting outpatient treatment for a patient, and a step of providing an outpatient medical assistance service that provides result data derived by analyzing the acquired medical data through a pre-trained generative artificial intelligence model.

[0023] In various embodiments, the step of providing the outpatient medical assistance service may include the step of generating a prompt including a command instructing to generate a result corresponding to the service to be provided based on the acquired medical data, and the step of providing result data derived by inputting the acquired medical data and the generated prompt into the generative artificial intelligence model as input data.

[0024] In various embodiments, the acquired medical data includes medical data in the form of voice generated by recording a voice conversation between the medical staff and the patient during the process of the medical staff providing medical treatment to the patient, and the step of providing the outpatient medical assistance service may include the step of converting the medical data in the form of voice into medical data in the form of text, the step of generating a medical chart using the converted medical data in the form of text, and the step of providing a medical chart generation service that provides the generated medical chart.

[0025] In various embodiments, the acquired medical data further includes medical image data generated as the medical staff filmed the process of conducting outpatient treatment on the patient, and the step of generating the medical chart may include the step of extracting information regarding medical acts performed by the medical staff on the patient from the medical image data and the step of updating the generated medical chart using the extracted information regarding medical acts.

[0026] In various embodiments, the step of generating the medical chart may include, if additional medical data corresponding to the patient—the additional medical data includes at least one of a medical record previously written for the patient and medical image data previously taken—extracting one or more pieces of information from the additional medical data and updating the generated medical chart using the extracted one or more pieces of information.

[0027] In various embodiments, the acquired medical data includes voice-form medical data generated by recording a voice conversation between the medical staff and the patient during the process of the medical staff conducting an initial medical examination of the patient, and the step of providing the outpatient medical assistance service may include the step of converting the voice-form medical data into text-form medical data, the step of generating an estimated diagnosis list for the patient using the converted text-form medical data, and the step of providing an estimated diagnosis recommendation service that provides the generated estimated diagnosis list.

[0028] In various embodiments, the acquired medical data further includes medical image data generated as the medical staff filmed the process of conducting an initial medical examination of the patient, and the step of generating the estimated diagnosis list may include the step of extracting information regarding medical acts performed by the medical staff on the patient from the medical image data and the step of updating the generated estimated diagnosis list using the extracted information regarding medical acts.

[0029] In various embodiments, the step of generating the estimated diagnosis list may include, if additional medical data corresponding to the patient—the additional medical data includes at least one of a medical record previously written for the patient and medical image data previously taken—extracting one or more pieces of information from the additional medical data and updating the generated estimated diagnosis list using the extracted one or more pieces of information.

[0030] In various embodiments, the acquired medical data is prepared by the medical staff and includes initial consultation chart record data containing initial medical details for the patient, and the step of providing the outpatient medical assistance service may include the step of generating an estimated diagnosis list for the patient using the initial consultation chart record data and the step of providing an estimated diagnosis recommendation service that provides the generated estimated diagnosis list.

[0031] In various embodiments, the acquired medical data includes test result data derived by performing one or more tests determined according to the initial medical results of the patient, and the step of providing the outpatient medical assistance service may include the step of generating a confirmed diagnosis list for the patient using the test result data and the step of providing a confirmed diagnosis recommendation service that provides the generated confirmed diagnosis list.

[0032] In various embodiments, the step of generating the confirmed diagnosis list may include, if additional medical data corresponding to the patient exists—the additional medical data includes at least one of voice-form medical data generated by recording a voice conversation between the medical staff and the patient during the process of the medical staff conducting a follow-up examination of the patient, medical image data generated by filming the process of the medical staff conducting a follow-up examination of the patient, and follow-up chart record data created by the medical staff and containing the details of the follow-up examination of the patient—a step of extracting one or more pieces of information from the additional medical data and a step of updating the generated confirmed diagnosis list using the extracted one or more pieces of information.

[0033] In various embodiments, the acquired medical data includes medical data in the form of voice generated by recording a voice conversation between the medical staff and the patient during the process of the medical staff providing medical treatment to the patient, and the step of providing the outpatient medical assistance service may include the step of converting the medical data in the form of voice into medical data in the form of text, the step of generating a test list using the converted medical data in the form of text, and the step of providing a test recommendation service that provides the generated test list.

[0034] In various embodiments, the step of generating the test list may include, if additional medical data corresponding to the patient—the additional medical data includes at least one of a diagnosis result and a previously written medical record for the patient—extracting one or more pieces of information from said additional medical data and updating the generated test list using said one or more pieces of information extracted.

[0035] In various embodiments, the acquired medical data includes voice-form medical data generated by recording a voice conversation between the medical staff and the patient during the process of the medical staff providing medical treatment to the patient, and the step of providing the outpatient medical assistance service may include the step of converting the voice-form medical data into text-form medical data, the step of generating a treatment list using the converted text-form medical data, and the step of providing a treatment recommendation service that provides the generated treatment list.

[0036] In various embodiments, the step of generating the treatment list may include updating the generated treatment list based on at least one of basic information including the age and gender of the patient, test result data derived from performing one or more tests on the patient, and previously stored medical data.

[0037] In various embodiments, the generative artificial intelligence model is retrained using the acquired medical data and the provided result data as training data, and if the provided result data is modified by the medical staff, the method may further include the step of retraining the generative artificial intelligence model using the acquired medical data and the modified result data as training data.

[0038] A computing device for performing a method for providing an outpatient medical assistance service based on generative artificial intelligence according to another embodiment of the present disclosure for solving the above-described problem comprises a processor, a network interface, a memory, and a computer program loaded into the memory and executed by the processor, wherein the computer program may include an instruction for acquiring medical data generated during the process of medical staff conducting outpatient treatment on a patient, and an instruction for providing an outpatient medical assistance service that provides result data derived by analyzing the acquired medical data through a pre-trained generative artificial intelligence model.

[0039] A computer program according to another embodiment of the present disclosure for solving the above-mentioned problem may be stored on a recording medium readable by a computing device to execute a method for providing an outpatient care assistance service based on generative artificial intelligence, comprising the steps of acquiring medical data generated during the process of medical staff conducting outpatient care for a patient, and providing result data derived by analyzing the acquired medical data through a pre-trained generative artificial intelligence model.

[0040] Other specific details of the present disclosure are included in the detailed description and drawings.

[0041] In addition, a method for providing a generative artificial intelligence-based medical advisory service according to one embodiment of the present disclosure may include, in a method performed by a computing device, a step of acquiring medical data including information regarding medical treatment, examination, diagnosis, prescription, and treatment performed on a patient, and a step of providing a medical advisory service based on result data derived by analyzing the acquired medical data through a pre-trained generative artificial intelligence model.

[0042] In various embodiments, the step of providing the medical advisory service may include: generating patient summary information based on the acquired medical data; acquiring one or more queries from a user; generating a prompt instructing to derive an answer to the acquired one or more queries based on the generated patient summary information; deriving an answer by inputting the generated prompt into the generative artificial intelligence model; and providing the derived answer to the user.

[0043] In various embodiments, the step of providing the derived answer may include the step of generating one or more recommended queries based on at least one of the obtained one or more queries and the derived answer, and the step of guiding the user to the generated one or more recommended queries or providing the generated one or more recommended queries and the answer corresponding to the generated one or more recommended queries.

[0044] In various embodiments, the step of deriving the answer includes the step of calculating a score corresponding to the reliability of the derived answer, the step of selecting one of a plurality of medical professionals in different fields based on the generated patient summary information when the calculated score is less than a reference score, and the step of re-deriving the answer by inputting the generated prompt into a medical professional model corresponding to one of the selected medical professional among the plurality of medical professional models, wherein the plurality of medical professional models may be generative artificial intelligence models trained using multiple medical data generated during the process in which each of the plurality of different medical professionals performs medical treatment, examination, diagnosis, prescription, and treatment on a patient, result data derived from each of the plurality of medical data, and medical data corresponding to each of the fields of the plurality of medical professionals as training data.

[0045] In various embodiments, the step of deriving the answer may include: calculating a score corresponding to the reliability of the derived answer; if the calculated score is less than a reference score, selecting one of a plurality of medical professionals in different fields based on the generated patient summary information; providing one or more of the obtained questions to the terminal of the selected medical professional; and obtaining an answer from the terminal of the selected medical professional as a response to the one or more of the provided questions.

[0046] In various embodiments, the step of providing the medical advisory service comprises: generating patient summary information based on the acquired medical data; generating a prompt instructing a judgment of appropriateness for each of a plurality of pre-set items based on the generated patient summary information; deriving a judgment of appropriateness result by inputting the generated prompt into the generative artificial intelligence model and providing the derived judgment of appropriateness result to a user, wherein the plurality of pre-set items may include at least one of the necessity of medical treatment, treatment method, medical treatment method, cost, insurance terms and conditions, compliance with regulations, and the accuracy and consistency of records.

[0047] In various embodiments, the step of generating the prompt may include setting appropriateness judgment criteria for each of the plurality of items and generating a prompt that instructs to judge the appropriateness of each of the plurality of items according to the set appropriateness judgment criteria, wherein the set appropriateness judgment criteria are determined based on the patient's age, gender, and disease history.

[0048] In various embodiments, the step of providing the derived appropriateness judgment result may include, based on the derived appropriateness judgment result, if at least one of the plurality of items is determined to be inappropriate, the step of generating a recommendation query corresponding to the at least one item, and the step of guiding the user to the generated one or more recommendation queries, or providing the generated one or more recommendation queries and an answer corresponding to the generated one or more recommendation queries.

[0049] A computing device for performing a method for providing a generative artificial intelligence-based medical advisory service according to another embodiment of the present disclosure for solving the above-described problem comprises a processor, a network interface, a memory, and a computer program loaded into the memory and executed by the processor, wherein the computer program may include an instruction for acquiring medical data including information regarding medical treatment, examination, diagnosis, prescription, and treatment performed on a patient, and an instruction for providing a medical advisory service that provides medical advice based on result data derived by analyzing the acquired medical data through a pre-trained generative artificial intelligence model.

[0050] A computer program according to another embodiment of the present disclosure for solving the above-described problem may be stored on a recording medium readable by a computing device to execute a method for providing a generative artificial intelligence-based medical advisory service, which includes the steps of acquiring medical data including information regarding medical treatment, examination, diagnosis, prescription, and treatment performed on a patient, and providing a medical advisory service that provides medical advice based on result data derived by analyzing the acquired medical data through a pre-trained generative artificial intelligence model.

[0051] Other specific details of the present disclosure are included in the detailed description and drawings.

[0052] According to various embodiments of the present disclosure, there is an advantage in that, through a generative AI-based medical staff model, result data is derived by analyzing medical data generated during the process of medical staff conducting outpatient treatment for patients, and medical guide information is provided to the medical staff using this data, thereby assisting in the decision-making of the medical staff and enabling the medical staff to perform outpatient treatment more easily.

[0053] In addition, there is an advantage in that highly reliable result data can be secured by building multiple medical team models for each medical team in advance, selecting the optimal model among them, and deriving result data through it.

[0054] In addition, by utilizing generative AI models to analyze medical data generated during outpatient consultations by medical staff and deriving result data, and by providing an outpatient assistance service that delivers this information, there is an advantage in significantly alleviating inconveniences during the outpatient process and offering a better medical experience for both patients and doctors.

[0055] Furthermore, by utilizing generative AI models to provide medical advice that analyzes patient medical data (e.g., information regarding medical consultations, tests, diagnoses, prescriptions, and treatments performed on the patient) or responds to inquiries based on such analysis, it is possible to resolve issues arising from existing medical advice methods and provide more efficient and accurate advice. Additionally, by rapidly analyzing large-scale data and providing objective advice based on consistent standards, there is an advantage that can bring benefits to both insurers and policyholders.

[0056] The effects of the present disclosure are not limited to those mentioned above, and other unmentioned effects will be clearly understood by a person skilled in the art from the description below.

[0057] The following drawings attached to this specification illustrate preferred embodiments of the present disclosure and serve to further enhance understanding of the technical concept of the present disclosure together with the detailed description of the invention; therefore, the present disclosure should not be interpreted as being limited only to the matters described in such drawings.

[0058] FIG. 1 is a diagram illustrating a generative artificial intelligence-based medical service provision system according to one embodiment of the present disclosure. More specifically, the generative artificial intelligence-based medical service provision system of FIG. 1 may include a medical guide provision system using a generative artificial intelligence-based medical staff model according to various embodiments of the present disclosure, a generative artificial intelligence-based outpatient treatment assistance service provision system, and a generative artificial intelligence-based medical advisory service provision system.

[0059] FIG. 2 is a diagram illustrating the hardware configuration of a computing device that performs a method for providing medical-related services based on generative artificial intelligence according to another embodiment of the present disclosure.

[0060] FIG. 3 is a diagram illustrating an exemplary network function applicable to various embodiments.

[0061] FIG. 4 is a diagram illustrating an exemplary generative artificial intelligence model applicable to various embodiments.

[0062] FIG. 5 is a flowchart of a method for providing medical guidance using a generative artificial intelligence-based medical staff model according to another embodiment of the present disclosure.

[0063] FIG. 6 is a flowchart of a method for providing outpatient medical assistance services based on generative artificial intelligence according to another embodiment of the present disclosure.

[0064] FIG. 7 is a flowchart of a method for providing a medical chart generation service according to various embodiments.

[0065] FIG. 8 is a flowchart of a method for providing an estimated diagnosis recommendation service according to various embodiments.

[0066] FIG. 9 is a diagram illustrating, exemplarily, a method of providing an estimated diagnostic list applicable to various embodiments.

[0067] FIG. 10 is a flowchart of a method for providing a confirmed diagnosis recommendation service according to various embodiments.

[0068] FIG. 11 is a flowchart of a method for providing an inspection recommendation service according to various embodiments.

[0069] FIG. 12 is a diagram illustrating an exemplary method of providing a test list applicable to various embodiments.

[0070] FIG. 13 is a flowchart of a method for providing a treatment recommendation service according to various embodiments.

[0071] FIG. 14 is a diagram illustrating, exemplarily, a method of providing a treatment list applicable to various embodiments.

[0072] FIGS. 15 and 16 are drawings illustrating an exemplary method of providing drug recommendation information applicable to various embodiments.

[0073] FIG. 17 is a flowchart of a method for providing a generative artificial intelligence-based medical advisory service according to another embodiment of the present disclosure.

[0074] FIG. 18 is a flowchart illustrating a method of providing answers to user inquiries as a medical advisory service in various embodiments.

[0075] FIG. 19 is a flowchart illustrating a method for providing results of determining the appropriateness of medical services as a medical advisory service in various embodiments.

[0076] A method, apparatus, and computer program for providing medical guidance using a generative artificial intelligence-based medical staff model are provided. A method for providing medical guidance using a generative artificial intelligence-based medical staff model according to various embodiments of the present disclosure comprises, in a method performed by a computing device, the steps of: acquiring medical data generated during the process of a medical staff member performing medical treatment on a patient; selecting one or more medical staff models from a plurality of medical staff models based on the acquired medical data; deriving result data by analyzing the acquired medical data through the selected one or more medical staff models; and providing medical guidance information to the medical staff member using the derived result data.

[0077] The advantages and features of the present disclosure and the methods for achieving them will become clear by referring to the embodiments described below in detail together with the accompanying drawings. However, the present disclosure is not limited to the embodiments disclosed below but may be implemented in various different forms. These embodiments are provided merely to ensure that the disclosure is complete and to fully inform those skilled in the art of the scope of the present disclosure, and the present disclosure is defined only by the scope of the claims.

[0078] The terms used herein are for describing the embodiments and are not intended to limit the disclosure. In this specification, the singular form includes the plural form unless specifically stated otherwise in the text. As used herein, "comprises" and / or "comprising" do not exclude the presence or addition of one or more other components in addition to the components mentioned.

[0079] Throughout this specification, the same reference numerals refer to the same components, and "and / or" includes each of the mentioned components and all combinations of one or more thereof. Although terms such as "first," "second," etc., are used to describe various components, they are not limited by these terms. These terms are used merely to distinguish one component from another. Accordingly, the first component mentioned below may be the second component within the technical scope of this disclosure.

[0080] As used herein, the terms “part” or “module” refer to hardware components such as software, FPGAs, or ASICs, and the “part” or “module” performs certain roles. However, the “part” or “module” is not limited to software or hardware. The “part” or “module” may be configured to reside in an addressable storage medium or configured to run on one or more processors. Thus, by example, the “part” or “module” includes components such as software components, object-oriented software components, class components, and task components, as well as processes, functions, attributes, procedures, subroutines, segments of program code, drivers, firmware, microcode, circuits, data, databases, data structures, tables, arrays, and variables. The functions provided within the components and “parts” or “modules” may be combined into a smaller number of components and “parts” or “modules,” or further separated into additional components and “parts” or “modules.”

[0081] Spatially relative terms such as "below," "beneath," "lower," "above," and "upper" may be used to facilitate the description of the relationship between one component and other components as illustrated in the drawings. Spatially relative terms should be understood as encompassing different orientations of components during use or operation, in addition to the orientations depicted in the drawings. For example, if a component depicted in a drawing is inverted, a component described as "below" or "beneath" of another component may be placed "above" of that component. Therefore, the exemplary term "below" may encompass both the lower and upper directions. Components may also be oriented in other directions, and accordingly, spatially relative terms may be interpreted according to the orientation.

[0082] Expressions such as "first," "second," or "first," "second" as used in this specification are used to distinguish one object from another when referring to a plurality of objects of the same kind, unless otherwise indicated by the context, and do not limit the order or importance of said objects.

[0083] Expressions used herein such as “A, B, and C,” “A, B, or C,” “A, B, and / or C,” or “at least one of A, B, and C,” “at least one of A, B, or C,” “at least one of A, B, and / or C,” “at least one selected from A, B, and C,” “at least one selected from A, B, or C,” “at least one selected from A, B, and / or C,” etc., may mean each of the listed items or all possible combinations of the listed items. For example, “at least one selected from A and B” may refer to (1) A, (2) at least one of A, (3) B, (4) at least one of B, (5) at least one of A and at least one of B, (6) at least one of A and B, (7) at least one of B and A, and (8) all of A and B.

[0084] As used herein, the expression “based on” is used to describe one or more factors affecting an act or action of a decision or judgment described in the phrase or sentence containing such expression, and such expression does not exclude additional factors affecting said act or action of a decision or judgment.

[0085] As used in this specification, the expression that a certain component (e.g., a first component) is "connected" or "connected" to another component (e.g., a second component) may mean that the said certain component is not only directly connected or connected to the said other component, but is also connected or connected through a new other component (e.g., a third component).

[0086] As used herein, the expression "configured to" may have meanings such as "set to," "capable of," "modified to," "made to," or "capable of." Such expression is not limited to the meaning of "specifically designed in hardware," and, for example, a processor configured to perform a specific operation may mean a generic-purpose processor capable of performing that specific operation by executing software.

[0087] Unless otherwise defined, all terms used herein (including technical and scientific terms) may be used in a meaning commonly understood by those skilled in the art to which this disclosure pertains. Additionally, terms defined in commonly used dictionaries are not to be interpreted ideally or excessively unless explicitly and specifically defined otherwise.

[0088] In this specification, the term "computer" refers to any type of hardware device comprising at least one processor, and may be understood to include software configurations operating on said hardware device according to the embodiments. For example, the term "computer" may be understood to include smartphones, tablet PCs, desktops, laptops, and user clients and applications running on each of these devices, but is not limited thereto.

[0089] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the attached drawings.

[0090] Each step described in this specification is described as being performed by a computer, but the subject of each step is not limited thereto, and depending on the embodiment, at least some of each step may be performed on different devices.

[0091]

[0092] FIG. 1 is a diagram illustrating a system for providing medical-related services based on generative artificial intelligence according to an embodiment of the present disclosure. More specifically, the system illustrated in FIG. 1 may include a system for providing medical guidance using a generative artificial intelligence-based medical staff model, a system for providing outpatient care assistance services based on generative artificial intelligence, and a system for providing medical advisory services based on generative artificial intelligence according to an embodiment of the present disclosure. Accordingly, the generative artificial intelligence-based medical-related services in the present disclosure may be understood to mean at least one of the services provided through the method for providing medical guidance using a generative artificial intelligence-based medical staff model, the method for providing outpatient care assistance services based on generative artificial intelligence, and the method for providing medical advisory services based on generative artificial intelligence, which are described below in the present disclosure.

[0093] Referring to FIG. 1, a generative artificial intelligence-based medical service provision system according to one embodiment of the present disclosure may include a computing device (100), a terminal (200), an external server (300), and a network (400).

[0094] Here, the generative artificial intelligence-based medical service provision system illustrated in FIG. 1 is according to one embodiment, and its components are not limited to the embodiment illustrated in FIG. 1 and may be added, changed, or deleted as needed.

[0095] In the first embodiment, the computing device (100) can provide medical guide information.

[0096] Here, medical guide information is information provided to medical staff treating a patient, and it may be information that guides the medical staff to perform the treatment more easily.

[0097] For example, medical guide information is information generated based on medical data created during the process of medical staff providing treatment to a patient, and may include, for instance, a medical chart, a list of presumed diagnoses containing information on one or more presumed diseases, a list of confirmed diagnoses containing information on one or more confirmed diseases, a list of tests containing one or more recommended test items, and a list of treatments containing one or more recommended treatments.

[0098] In various embodiments, the computing device (100) can provide outpatient assistance services to medical staff performing outpatient care.

[0099] Here, the outpatient care assistance service provided by the computing device (100) may be a service that provides medical guard information to medical staff who intend to perform outpatient care. For example, the outpatient care assistance service may include a medical chart generation service that generates and provides a medical chart of the medical staff, an estimated diagnosis recommendation service that provides an estimated diagnosis list for a first-time patient, a confirmed diagnosis recommendation service that provides a confirmed diagnosis list for a follow-up patient, an examination recommendation service that provides an examination list for the patient, and a treatment recommendation service that provides a treatment list for the patient. However, it is not limited thereto.

[0100] In the second embodiment, the computing device (100) can provide outpatient assistance services.

[0101] Here, the outpatient care assistance service provided by the computing device (100) may be a service provided to medical staff who intend to perform outpatient care. For example, the outpatient care assistance service may include a medical chart generation service to assist medical staff in creating medical charts, an estimated diagnosis recommendation service to assist in the estimated diagnosis of a first-time patient, a confirmed diagnosis recommendation service to assist in the confirmed diagnosis of a follow-up patient, a test recommendation service to assist in the process of determining tests required for a patient, and a treatment recommendation service to assist in prescribing treatment for a patient. However, it is not limited thereto.

[0102] In the third embodiment, the computing device (100) may provide a medical advisory service. Here, the medical advisory service may be a service that provides medical advice based on result data derived from analyzing medical data.

[0103] For example, a medical advisory service may be a service that provides answers to queries obtained from a user (e.g., a medical advisory requester) based on medical data about a patient.

[0104] As another example, a medical advisory service may be a service that determines the appropriateness of medical treatment, examinations, diagnoses, prescriptions, and treatments performed on a patient based on medical data, provides the results of the appropriateness determination, or provides various information based on the results of the appropriateness determination.

[0105] As another example, a medical advisory service may be a service that generates patient summary information based on medical data about a patient and provides it to the user.

[0106] In addition, the generative artificial intelligence-based medical-related service provided by the computing device (100) may be a service that can be accessed and executed through a web-based interface and a mobile application.

[0107] For example, a generative AI-based medical service may be a service that connects to a web server using HTTP / HTTPS protocols and is provided through a user interface configured using HTML, JavaScript, and CSS.

[0108] As another example, generative AI-based medical services may be provided in the form of mobile applications supporting iOS and Android operating systems to enable efficient access through various types of mobile devices. However, they are not limited to this.

[0109] In various embodiments, the computing device (100) may be connected to a terminal (200) via a network (400) and may provide a generative artificial intelligence-based medical-related service to the terminal (200).

[0110] In the first embodiment, the computing device (100) can acquire medical data generated during the process of medical staff conducting outpatient treatment for a patient, and can provide result data derived by analyzing the medical data through a pre-trained medical staff model as medical guide information to the medical staff. That is, the computing device (100) can be connected to the medical staff's terminal (200) through a network (400), and can provide medical guide information to the medical staff's terminal (200) in response to a request obtained from the medical staff's terminal (200).

[0111] In the second embodiment, the computing device (100) can acquire medical data generated during the process of medical staff conducting outpatient treatment for a patient, and can provide an outpatient treatment assistance service that provides result data derived by analyzing the medical data through a pre-trained artificial intelligence model. That is, the computing device (100) can be connected to the medical staff's terminal (200) through a network (400), and can provide an outpatient treatment assistance service to the medical staff's terminal (200) in response to a request obtained from the medical staff's terminal (200).

[0112] In the third embodiment, the computing device (100) can be connected to a terminal (200) via a network (400) and can provide a generative artificial intelligence-based medical advisory service to the terminal (200).

[0113] Accordingly, the computing device (100) in the present disclosure performing a method for providing medical-related services based on generative artificial intelligence may be understood as the computing device (100) performing at least one of a method for providing medical guidance using a generative artificial intelligence-based medical staff model, a method for providing outpatient medical assistance services based on generative artificial intelligence, or a method for providing medical advisory services based on generative artificial intelligence.

[0114] Here, a terminal (200) (e.g., a terminal (200) of a user, patient, and / or medical staff) may refer to any form of entity(s) in a system having a mechanism for communicating with a computing device (100). For example, such a terminal (200) may include a personal computer (PC), a notebook, a mobile terminal, a smartphone, a tablet PC, and a wearable device, and may include any type of terminal capable of connecting to a wired or wireless network. Additionally, the terminal (200) may include any computing device implemented by at least one of an agent, an API (Application Programming Interface), and a plug-in. Additionally, the terminal (200) may include an application source and / or a client application.

[0115] Additionally, the network (400) may refer to a connection structure capable of exchanging information between each node, such as multiple terminals and servers. For example, the network (400) may include a Local Area Network (LAN), a Wide Area Network (WAN), the World Wide Web (WWW), a wired / wireless data network, a telephone network, a wired / wireless television network, a Controller Area Network (CAN), and Ethernet.

[0116] Wireless data communication networks may include, but are not limited to, 3G, 4G, 5G, 3GPP (3rd Generation Partnership Project), 5GPP (5th Generation Partnership Project), LTE (Long Term Evolution), WIMAX (World Interoperability for Microwave Access), WiFi, Internet, LAN (Local Area Network), Wireless LAN (Wireless Local Area Network), WAN (Wide Area Network), PAN (Personal Area Network), RF (Radio Frequency), Bluetooth network, NFC (Near-Field Communication) network, satellite broadcasting network, analog broadcasting network, DMB (Digital Multimedia Broadcasting) network, etc.

[0117] In one embodiment, an external server (300) is connected to a computing device (100) via a network (400) and can store and manage various information / data necessary for the computing device (100) to perform a generative AI-based medical-related service provision method, or receive, store, and manage various information / data derived as the computing device (100) performs a generative AI-based medical-related service provision method. For example, the external server (300) may be a storage server separately provided outside the computing device (100), but is not limited thereto. Hereinafter, with reference to FIG. 2, the hardware configuration of a computing device (100) that performs a generative AI-based medical-related service provision method according to another embodiment of the present disclosure will be described.

[0118]

[0119] FIG. 2 is a diagram illustrating the hardware configuration of a computing device that performs a method for providing medical-related services based on generative artificial intelligence according to another embodiment of the present disclosure.

[0120] Referring to FIG. 2, in various embodiments, a computing device (100) may include one or more processors (110), a memory (120) for loading a computer program (151) executed by the processor (110), a bus (130), a communication interface (140), and a storage (150) for storing the computer program (151). Here, FIG. 2 illustrates only the components relevant to the embodiments of the present disclosure. Accordingly, a person skilled in the art to which the present disclosure pertains will understand that other general-purpose components may be included in addition to the components illustrated in FIG. 2.

[0121] The processor (110) controls the overall operation of each component of the computing device (100). The processor (110) may be configured to include a CPU (Central Processing Unit), an MPU (Micro Processor Unit), an MCU (Micro Controller Unit), a GPU (Graphic Processing Unit), or any form of processor well known in the art of the present disclosure.

[0122] Additionally, the processor (110) may perform operations for at least one application or program for executing the method according to the embodiments of the present disclosure, and the computing device (100) may have one or more processors.

[0123] In various embodiments, the processor (110) may further include Random Access Memory (RAM) (not shown) and Read-Only Memory (ROM) (not shown) for temporarily and / or permanently storing signals (or data) processed within the processor (110). Additionally, the processor (110) may be implemented in the form of a System on Chip (SoC) comprising at least one of a graphics processing unit, RAM, and ROM.

[0124] Memory (120) stores various data, instructions and / or information. Memory (120) may load a computer program (151) from storage (150) to execute a method / operation according to various embodiments of the present disclosure. When the computer program (151) is loaded into memory (120), the processor (110) may perform the method / operation by executing one or more instructions constituting the computer program (151). Memory (120) may be implemented as a volatile memory such as RAM, but the technical scope of the present disclosure is not limited thereto.

[0125] The bus (130) provides communication functions between components of the computing device (100). The bus (130) can be implemented as various types of buses, such as an address bus, a data bus, and a control bus.

[0126] The communication interface (140) supports wired and wireless internet communication of the computing device (100). Additionally, the communication interface (140) may support various communication methods other than internet communication. To this end, the communication interface (140) may be configured to include a communication module well known in the art of the present disclosure. In some embodiments, the communication interface (140) may be omitted.

[0127] Storage (150) can store computer programs (151) non-temporarily. When performing a medical guidance provision process using a generative AI-based medical staff model through a computing device (100), storage (150) can store various information necessary to provide a medical guidance provision process using a generative AI-based medical staff model, various information necessary to provide a generative AI-based outpatient care assistance service provision process, and various information necessary to provide a generative AI-based medical advisory service provision process.

[0128] The storage (150) may be configured to include non-volatile memory such as ROM (Read Only Memory), EPROM (Erasable Programmable ROM), EEPROM (Electrically Erasable Programmable ROM), flash memory, a hard disk, a removable disk, or any form of computer-readable recording medium well known in the art to which the present disclosure belongs.

[0129] A computer program (151) may include one or more instructions that cause a processor (110) to perform a method / operation according to various embodiments of the present disclosure when loaded into memory (120). That is, the processor (110) may perform the method / operation according to various embodiments of the present disclosure by executing the one or more instructions.

[0130] In the first embodiment, the computer program (151) may include one or more instructions for performing a method of providing medical guidance using a generative artificial intelligence-based medical staff model, comprising the steps of: acquiring medical data generated during the process of a medical staff member performing medical treatment on a patient; selecting one or more medical staff models from a plurality of medical staff models based on the acquired medical data; deriving result data by analyzing the acquired medical data through the selected one or more medical staff models; and providing medical guidance information to the medical staff member using the derived result data.

[0131] In the second embodiment, the computer program (151) may include one or more instructions for performing a method of providing an outpatient care assistance service based on generative artificial intelligence, the method comprising the steps of acquiring medical data generated during the process of medical staff conducting outpatient care for a patient, and providing an outpatient care assistance service that provides result data derived by analyzing the acquired medical data through a pre-trained generative artificial intelligence model.

[0132] In the third embodiment, the computer program (151) may include one or more instructions for performing a method of providing a generative artificial intelligence-based medical advisory service, which includes the step of acquiring medical data including information regarding medical treatment, examination, diagnosis, prescription, and treatment performed on a patient, and the step of providing a medical advisory service that provides medical advice based on result data derived by analyzing the acquired medical data through a pre-trained generative artificial intelligence model.

[0133] The steps of the method or algorithm described in connection with the embodiments of the present disclosure may be implemented directly in hardware, implemented as a software module executed by hardware, or implemented by a combination thereof. The software module may reside in RAM (Random Access Memory), ROM (Read Only Memory), EPROM (Erasable Programmable ROM), EEPROM (Electrically Erasable Programmable ROM), Flash Memory, a hard disk, a removable disk, a CD-ROM, or any form of computer-readable recording medium well known in the art to which the present disclosure belongs.

[0134] The components of the present disclosure may be implemented as a program (or application) and stored on a medium to be executed in combination with a computer, which is hardware. The components of the present disclosure may be implemented as software programming or software elements, and similarly, embodiments may be implemented in programming or scripting languages ​​such as C, C++, Java, assembler, etc., including various algorithms implemented as combinations of data structures, processes, routines, or other programming configurations. Functional aspects may be implemented as algorithms executed on one or more processors.

[0135] Meanwhile, the medical staff model and / or artificial intelligence model (e.g., neural network) in the present disclosure is composed of one or more network functions, and one or more network functions may be composed of a set of interconnected computational units that may generally be referred to as 'nodes'. These 'nodes' may also be referred to as 'neurons'. One or more network functions are composed of at least one node. The nodes (or neurons) constituting one or more network functions may be interconnected by one or more 'links'.

[0136] In an artificial intelligence model, one or more nodes connected via links can form a relative relationship between an input node and an output node. The concepts of input and output nodes are relative; any node in an output node relationship with respect to one node may be in an input node relationship with respect to another node, and vice versa. As previously mentioned, the input node versus output node relationship can be generated based on links. One or more output nodes may be connected to a single input node via links, and vice versa.

[0137] In a relationship between input and output nodes connected via a single link, the value of the output node can be determined based on data input into the input node. Here, the nodes interconnecting the input and output nodes may have weights. These weights can be variable and may be varied by a user or an algorithm to enable the artificial intelligence model to perform desired functions. For example, if one or more input nodes are interconnected to a single output node via respective links, the output node value can be determined based on the values ​​input into the input nodes connected to the output node and the weights set on the links corresponding to each input node.

[0138] As described above, an artificial intelligence model consists of one or more nodes interconnected through one or more links, forming input and output node relationships within the model. The characteristics of an artificial intelligence model can be determined by the number of nodes and links within the model, the relationships between the nodes and links, and the weight values ​​assigned to each link. For example, if two artificial intelligence models exist with the same number of nodes and links but different weight values ​​between the links, the two models may be perceived as different from each other.

[0139] Some of the nodes constituting an artificial intelligence model may form a layer based on their distances from the initial input node. For example, a set of nodes with a distance of n from the initial input node may form n layers. The distance from the initial input node can be defined by the minimum number of links that must be traversed to reach that node from the initial input node. However, this definition of a layer is arbitrary for illustrative purposes, and the degree of a layer within the artificial intelligence model may be defined in a way different from that described above. For example, the layer of nodes may be defined by their distance from the final output node.

[0140] The initial input node may refer to one or more nodes within the artificial intelligence model to which data is directly input without passing through links in relation to other nodes. Alternatively, within the artificial intelligence model network, in terms of relationships between nodes based on links, it may refer to nodes that do not have other input nodes connected by links. Similarly, the final output node may refer to one or more nodes within the artificial intelligence model that do not have output nodes in relation to other nodes. Additionally, the hidden node may refer to nodes constituting the artificial intelligence model that are neither the initial input node nor the final output node. An artificial intelligence model according to one embodiment of the present disclosure may have more nodes in the input layer than nodes in the hidden layer that are close to the output layer, and may be an artificial intelligence model in which the number of nodes decreases as it progresses from the input layer to the hidden layer.

[0141] An artificial intelligence model may include one or more hidden layers. The hidden nodes of a hidden layer can take the output of the previous layer and the output of neighboring hidden nodes as input. The number of hidden nodes for each hidden layer may be the same or different. The number of nodes in the input layer may be determined based on the number of data fields in the input data and may be the same or different from the number of hidden nodes. The input data fed into the input layer can be processed by the hidden nodes of the hidden layer and output by the fully connected layer (FCL), which is the output layer.

[0142] In various embodiments, the artificial intelligence model may be a deep learning model (e.g., FIG. 3).

[0143] A deep learning model (e.g., a deep neural network (DNN)) can refer to an artificial intelligence model that includes multiple hidden layers in addition to input and output layers. Using a deep neural network, one can identify the latent structures of data. That is, one can identify the latent structures of photos, text, videos, voice, and music (e.g., what objects are in a photo, what the content and emotions of the text are, what the content and emotions of the voice are, etc.).

[0144] Deep neural networks may include, but are not limited to, convolutional neural networks (CNN), recurrent neural networks (RNN), autoencoders, Generative Adversarial Networks (GAN), restricted Boltzmann machines (RBM), deep belief networks (DBN), Q networks, U networks, and Siamese networks.

[0145] In various embodiments, the network function may include an autoencoder. Here, the autoencoder may be a type of artificial neural network for outputting output data similar to the input data.

[0146] An autoencoder may include at least one hidden layer, and an odd number of hidden layers may be placed between the input and output layers. The number of nodes in each layer may be reduced from the number of nodes in the input layer to an intermediate layer called the bottleneck layer (encoding), and then expanded symmetrically from the bottleneck layer to the output layer (symmetrical to the input layer). The nodes of the dimensionality reduction layer and the dimensionality restoration layer may or may not be symmetrical. Additionally, the autoencoder can perform non-linear dimensionality reduction. The number of input and output layers may correspond to the number of sensors remaining after the preprocessing of the input data. In the autoencoder structure, the number of nodes in the hidden layers included in the encoder may have a structure where it decreases as it moves away from the input layer. Since the number of nodes in the bottleneck layer (the layer with the fewest nodes located between the encoder and the decoder) may not transmit a sufficient amount of information if it is too small, it may be maintained at a certain number or higher (e.g., more than half the number of the input layer).

[0147] In various embodiments, the artificial intelligence model may be a generative artificial intelligence model (Generative AI) (e.g., FIG. 4).

[0148] Generative artificial intelligence models can refer to natural language models that have been pre-trained on large-scale data using Transformers. Transformer models are constructed based on an attention mechanism and are characterized by an encoder-decoder structure that improves upon the shortcomings of existing sequence-to-sequence models and designs an encoder-decoder that does not use recurrent neural networks.

[0149] Attention was previously used to calibrate recurrent neural networks in sequence-to-sequence models, but the Transformer designed an encoder-decoder using only attention without recurrent neural networks. The attention technique used in the Transformer is called Multi-Head Attention or Self-Attention.

[0150] The Transformer model is characterized by being trained to increase the probability value of the correct answer (the next word in the output sequence) given the encoder input (input sequence) and the decoder input (words constituting the output sequence).

[0151] A pre-trained language model can refer to a language model that has learned the process of performing a specific task using large-scale language data (corpus). For example, GPT-family models are trained on the task of predicting the next word given previous words, and undergo a training process to update the model to correctly predict the next word. As another example, BERT-family models are trained on the task of predicting the empty word in the middle from given words on both sides, and undergo a training process to update the model to correctly predict the empty word.

[0152] The characteristic of these tasks is that the computer can perform self-supervised learning by automatically covering and guessing words one by one, even when a large amount of language data, which is relatively easy to collect, is input without any separate preliminary work. Once a model pre-trained on a large amount of language data in this way is completed, it can be utilized by fine-tuning it to perform various detailed tasks.

[0153] Fine-tuning is a type of transfer learning method that retrains a model trained on a specific task to perform a different task; it leverages the experience of a pre-trained model to enable it to learn new tasks relatively easily. For example, a pre-trained language model is trained to fill in blanks, but through fine-tuning, one can obtain a model capable of generating responses to questions or classifying documents.

[0154] By significantly increasing the scale of these pre-trained language models and pre-training them with an overwhelmingly large amount of data, results were obtained that greatly surpassed the performance of existing artificial intelligence models. Consequently, the AI ​​models developed in this way are called Hyperscale AI or Large Language Models (LLM).

[0155]

[0156] Hereinafter, with reference to FIGS. 5 to 19, a method for providing medical-related services based on generative artificial intelligence performed by a computing device (100) will be described. More specifically, with reference to FIGS. 5 to 19, a method for providing medical guidance using a generative artificial intelligence-based medical staff model, a method for providing outpatient treatment assistance services based on generative artificial intelligence, and a method for providing medical consultation services based on generative artificial intelligence performed by a computing device (100) will be described in detail.

[0157] First, with reference to FIG. 5, a method for providing medical guidance using a generative artificial intelligence-based medical staff model performed by a computing device (100) will be described.

[0158] FIG. 5 is a flowchart of a method for providing medical guidance using a generative artificial intelligence-based medical staff model according to another embodiment of the present disclosure.

[0159] Referring to FIG. 5, in step S510, the computing device (100) can acquire medical data.

[0160] For example, the computing device (100) can acquire medical data generated during the process of medical staff performing medical treatment on a patient.

[0161] Here, medical data may refer to data generated and collected during the process of medical staff conducting outpatient treatment for patients.

[0162] For example, medical data may be basic information about patients receiving outpatient care (e.g., age, gender, underlying diseases, etc.).

[0163] As another example, medical data may be audio-based data generated by recording voice conversations between medical staff and a patient during the course of medical treatment (e.g., initial consultation, follow-up consultation, etc.).

[0164] As another example, medical data may be medical imaging data generated as medical staff record the process of conducting outpatient consultations with patients.

[0165] As another example, medical data may be medical records previously created for a patient (e.g., previous medical records created at the same institution or medical records from other institutions).

[0166] As another example, medical data may be test result data derived from performing one or more tests on a patient (e.g., medical imaging data such as MRI, X-RAY, CT, etc.).

[0167] As another example, medical data is created by medical staff and may be chart record data containing outpatient treatment details (e.g., initial consultation chart record data and follow-up consultation chart record data, etc.). However, it is not limited to this.

[0168] In step S520, the computing device (100) can select one or more medical staff models from among multiple medical staff models based on the medical data obtained through step S510.

[0169] Here, the multiple medical staff model is a model generated individually corresponding to each of the multiple medical staff, and may be a model trained using multiple medical data generated during the process of each of the different multiple medical staff treating a patient and result data derived from each of the multiple medical data as training data.

[0170] For example, the computing device (100) can generate training data based on medical data generated as the first medical staff performs medical treatment on a patient, and use this to train a first medical staff model corresponding to the first medical staff.

[0171] Additionally, the computing device (100) can generate training data based on medical data generated as the second medical staff performs medical treatment on the patient, and use this to train a second medical staff model corresponding to the second medical staff. However, it is not limited thereto.

[0172] In various embodiments, the computing device (100) can extract disease information about a patient from medical data (e.g., information about a disease that has been presumed or confirmed diagnosed about the patient) and can select one or more medical staff models among a plurality of medical staff models using the extracted disease information.

[0173] For example, when a specific disease is extracted as disease information about a patient from medical data, the computing device (100) may select a medical team model of a medical team specializing in a field that includes the specific disease.

[0174] As another example, when a specific disease is extracted as disease information about a patient from medical data, the computing device (100) may select a medical professional model trained using medical data of patients with the specific disease as training data.

[0175] In various embodiments, the computing device (100) can extract patient information (e.g., patient's age, gender, occupation, nationality, underlying disease, etc.) from medical data and can select one or more medical staff models from a plurality of medical staff models using the extracted patient information.

[0176] More specifically, the computing device (100) can extract one or more features related to patient information (e.g., patient's age, gender, occupation, nationality, underlying disease, etc.) from medical data, and can select one or more past patients among a plurality of past patients who have characteristics similar to the patient based on the extracted one or more features. For example, the computing device (100) can calculate the similarity between the feature information of past patients stored in a previously stored patient database and one or more features extracted from the medical data, and can select past patients whose calculated similarity is greater than or equal to a threshold value.

[0177] Here, various methods may be applied to calculate similarity, such as the Euclidean distance method, which calculates the distance between two patients after quantifying all features regarding the patients, and the cosine similarity method, which calculates the similarity based on the cosine value between vectors corresponding to the features of the two patients, but is not limited thereto.

[0178] In addition, the previously stored patient database stores information about the patient, medical data corresponding to the patient, and result data derived by inputting the medical data into the medical staff model, and may be a repository of data used for training multiple medical staff models.

[0179] Afterwards, the computing device (100) can select a medical team model that has been learned based on the medical data of past patients selected according to the above method among a plurality of medical team models.

[0180] In various embodiments, the computing device (100) can calculate a score for each of the plurality of medical staff models based on feedback obtained corresponding to each of the plurality of medical staff models for a predetermined period, and can select one or more of the plurality of medical staff models based on the score.

[0181] Here, feedback may refer to responses obtained from multiple medical professionals by providing multiple result data, derived by inputting multiple medical data acquired over a predetermined period into multiple medical professional models, to multiple medical professionals corresponding to each of the multiple medical professional models.

[0182] For example, the feedback may include evaluation information (e.g., evaluation scores, etc.) obtained from multiple medical staff as multiple result data is provided to multiple medical staff corresponding to each of the multiple medical staff models.

[0183] In addition, the feedback may include whether the result data obtained from the multiple medical staff has been modified (e.g., whether it has been modified and the degree of modification, etc.) as the multiple result data is provided to the multiple medical staff corresponding to each of the multiple medical staff models.

[0184] For example, when a computing device (100) obtains multiple feedbacks corresponding to each specific medical staff model during a predetermined period, it can calculate the score for a specific medical staff model by calculating the average value of the evaluation scores included in the multiple feedbacks.

[0185] As another example, when a computing device (100) obtains multiple feedbacks corresponding to each specific medical staff model during a predetermined period, it can calculate the ratio of modified result data to the total result data based on whether result data included in the multiple feedbacks has been modified, and calculate a score for the specific medical staff model according to the calculated ratio (assigning a lower score as the ratio increases).

[0186] In various embodiments, the computing device (100) may select one main medical staff model and one or more sub-medical staff models based on acquired medical data. For example, the computing device (100) may calculate a score for each of the multiple medical staff models, select one medical staff model with the highest calculated score as the main medical staff model, and sequentially select n medical staff models starting from the medical staff model with the second highest score as sub-medical staff models, but is not limited thereto.

[0187] In various embodiments, each of the plurality of medical staff models may include a medical chart generation model, a diagnosis model, a test recommendation model, and a treatment recommendation model, and the computing device (100) may individually select one or more medical chart generation models, one or more diagnosis models, one or more test recommendation models, and one or more treatment recommendation models from the plurality of medical staff models based on medical data.

[0188] In various embodiments, the computing device (100) can determine the type of service to be provided based on medical data, select the result to be provided through the service based on the determined type of service, and select a medical staff model among a plurality of medical staff models to output the result to be provided through the service.

[0189] For example, when the computing device (100) obtains medical data in the form of voice corresponding to the initial medical treatment from the medical staff, it may determine the type of service to be provided as a medical chart generation service and / or an estimated diagnosis recommendation service, and accordingly determine the result to be provided as a medical chart and / or an estimated diagnosis list, and based on this, it may select one or more medical staff models for creating the medical chart and / or estimated diagnosis list among a plurality of medical staff models.

[0190] Additionally, when the computing device (100) obtains test result data from medical staff, it can determine the type of service to be provided as a confirmed diagnosis recommendation service and, accordingly, determine the result to be provided as a confirmed diagnosis list, and based on this, select one or more medical staff models among multiple medical staff models for creating the confirmed diagnosis list.

[0191] Additionally, when the computing device (100) obtains medical data in the form of voice corresponding to medical treatment from the medical staff, it can determine the type of service to be provided as a test recommendation service and / or a treatment recommendation service, and accordingly determine the result to be provided as a recommended test list and / or a recommended treatment list, and based on this, it can select one or more medical staff models among multiple medical staff models for creating the recommended test list and / or recommended treatment list.

[0192] Selecting one or more medical staff models from among multiple medical staff models based on a feature service means selecting a medical staff model with high performance regarding the outcome to be provided through the feature service; such performance may be determined by a score calculated based on feedback regarding the outcome to be provided through the feature service, but is not limited thereto.

[0193] In step S530, the computing device (100) can derive result data by analyzing medical data using one or more medical staff models selected through step S520.

[0194] In various embodiments, the computing device (100) can generate input data based on medical data and output result data as the input data is input into a medical staff model.

[0195] More specifically, first, the computing device (100) can generate a prompt containing a command that instructs to generate a result to be derived through the medical staff model.

[0196] In various embodiments, when the computing device (100) obtains a request from a user to provide a specific service, it may generate a prompt including a command instructing to generate a result to be provided through the specific service. For example, when the computing device (100) obtains a request from a user to provide a medical chart generation service, it may generate a prompt including a command instructing to create a medical chart based on a specific template.

[0197] In various embodiments, the computing device (100) can determine the type of service to be provided based on medical data and can generate a prompt including a command that instructs to generate a result to be provided through the service based on the determined type of service.

[0198] For example, the computing device (100) can generate a prompt that instructs the generation of a medical chart or an estimated diagnosis list when the data obtained from the medical staff is voice-type medical data corresponding to the initial medical treatment.

[0199] Additionally, the computing device (100) can generate a prompt instructing the creation of a confirmed diagnosis list if the data obtained from the medical staff is test result data.

[0200] Additionally, the computing device (1000) may generate a prompt instructing the creation of an examination list and a treatment list when the data obtained from the medical staff is medical data in the form of voice corresponding to a subsequent medical examination. However, it is not limited thereto.

[0201] Afterwards, the computing device (100) can generate input data using medical data and a prompt.

[0202] Afterwards, the computing device (100) can extract result data as input data is input into the medical staff model.

[0203] Here, the result data derived through the medical team model may be, but is not limited to, medical charts, lists of estimated diagnoses, lists of confirmed diagnoses, lists of tests, and lists of treatments.

[0204] In various embodiments, when one main medical team model and one or more sub-medical team models are selected based on medical data, the computing device (100) can derive a first result data by analyzing the medical data through one main medical team model, derive one or more second result data by analyzing the medical data through one or more sub-medical team models, and verify the first result data using one or more second result data.

[0205] For example, the computing device (100) may determine that the first result data is valid if the similarity between the first result data and one or more second result data is greater than or equal to a threshold value.

[0206] Meanwhile, the computing device (100) may determine that the first result data is invalid if the similarity between the first result data and one or more second result data is less than a threshold value.

[0207] In step S540, the computing device (100) can provide the result data generated through step S530 to the medical staff as medical guide information.

[0208] In various embodiments, the computing device (100) is connected to a medical staff terminal (200) via a network (400) and may provide a user interface (UI) that provides outpatient medical assistance services to the terminal (200), and may provide medical guide information including result data through the UI. However, it is not limited thereto.

[0209] The method for providing medical guidance using the aforementioned generative AI-based medical staff model has been described with reference to the flowchart illustrated in the drawings. For the sake of simplicity, the method for providing medical guidance using the generative AI-based medical staff model has been illustrated and described using a series of blocks; however, the present disclosure is not limited to the order of the blocks, and some blocks may be performed in a different order or simultaneously than those illustrated and described in this specification. Furthermore, new blocks not described in this specification and drawings may be added, or some blocks may be deleted or modified.

[0210]

[0211] Hereinafter, with reference to FIGS. 6 to 16, a method for providing an outpatient medical assistance service based on generative artificial intelligence performed by a computing device (100) will be described.

[0212] FIG. 6 is a flowchart of a method for providing outpatient medical assistance services based on generative artificial intelligence according to another embodiment of the present disclosure.

[0213] Referring to FIG. 6, in step S610, the computing device (100) can acquire medical data.

[0214] Here, medical data may refer to data generated and collected during the process of medical staff conducting outpatient treatment for patients.

[0215] For example, medical data may be audio-based data generated by recording voice conversations between medical staff and a patient during the process of medical staff providing medical care to the patient (e.g., initial consultation, follow-up consultation, etc.).

[0216] As another example, medical data may be medical imaging data generated as medical staff record the process of conducting outpatient consultations with patients.

[0217] As another example, medical data may be medical records previously created for a patient (e.g., previous medical records created at the same institution or medical records from other institutions).

[0218] As another example, medical data may be test result data derived from performing one or more tests on a patient (e.g., medical imaging data such as MRI, X-RAY, CT, etc.).

[0219] As another example, medical data is created by medical staff and may be chart record data containing outpatient treatment details (e.g., initial consultation chart record data and follow-up consultation chart record data, etc.). However, it is not limited to this.

[0220] In step S620, the computing device (100) can generate input data for an artificial intelligence model based on the medical data obtained through step S610.

[0221] More specifically, first, the computing device (100) can generate a prompt containing a command that instructs to generate a result corresponding to the service to be provided.

[0222] In various embodiments, when the computing device (100) obtains a request from a user to provide a specific service, it may generate a prompt including a command instructing to generate a result to be provided through the specific service. For example, when the computing device (100) obtains a request from a user to provide a medical chart generation service, it may generate a prompt including a command instructing to create a medical chart based on a specific template.

[0223] In various embodiments, the computing device (100) can determine the type of service to be provided based on medical data and can generate a prompt including a command that instructs to generate a result to be provided through the service based on the determined type of service.

[0224] For example, the computing device (100) can generate a prompt that instructs the generation of a medical chart or an estimated diagnosis list when the data obtained from the medical staff is voice-type medical data corresponding to the initial medical treatment.

[0225] Additionally, the computing device (100) can generate a prompt instructing the creation of a confirmed diagnosis list if the data obtained from the medical staff is test result data.

[0226] Additionally, the computing device (1000) may generate a prompt instructing the creation of an examination list and a treatment list when the data obtained from the medical staff is medical data in the form of voice corresponding to a subsequent medical examination. However, it is not limited thereto.

[0227] Afterwards, the computing device (100) can generate input data using medical data and a prompt.

[0228] In step S630, the computing device (100) can extract result data by inputting the input data generated through step S620 into a generative artificial intelligence model.

[0229] Here, the result data derived through the generative artificial intelligence model may be, but is not limited to, medical charts, estimated diagnosis lists, confirmed diagnosis lists, test lists, and treatment lists.

[0230] In step S640, the computing device (100) can provide an outpatient medical assistance service that provides result data derived through step S630. Hereinafter, with reference to FIGS. 7 to 16, a method for providing various detailed services included in the outpatient medical assistance service will be described.

[0231] FIG. 7 is a flowchart of a method for providing a medical chart generation service according to various embodiments.

[0232] Referring to FIG. 7, in step S710, the computing device (100) can obtain voice-form medical data generated by recording a voice conversation between the medical staff and the patient while the medical staff is providing medical care to the patient.

[0233] In step S720, the computing device (100) can generate a medical chart using the voice-form medical data obtained through step S710.

[0234] More specifically, first, the computing device (100) can convert medical data in voice form into medical data in text form. For example, the computing device (100) can convert medical data in voice form into medical data in text form using a speech-to-text model, but is not limited thereto.

[0235] Subsequently, the computing device (100) can generate a medical chart using medical data in text form. For example, the computing device (100) can generate a prompt including a command instructing the creation of a medical chart in a pre-set format based on medical data in text form, and can generate a medical chart as result data by inputting input data generated by combining the medical data in text form and the prompt into a generative artificial intelligence model.

[0236] In step S730, the computing device (100) can determine whether there is additional medical data corresponding to the patient.

[0237] In step S740, if the computing device (100) determines through step S730 that there is no additional medical data corresponding to the patient, it may provide a medical chart generation service to the medical staff's terminal (200) that provides a medical chart generated through step S720.

[0238] In step S750, if the computing device (100) determines through step S730 that additional medical data corresponding to the patient exists, it can update the medical chart based on information extracted from the additional medical data.

[0239] Here, updating a medical chart may mean, but is not limited to, additionally recording information extracted from additional medical data in the medical chart, and depending on the case, may involve replacing information recorded in the medical chart with information extracted from medical data, or modifying and / or deleting information recorded in the medical chart based on information extracted from medical data.

[0240] For example, if there is medical image data generated as additional medical data by medical staff filming the process of outpatient treatment of a patient, the computing device (100) can extract information regarding medical acts performed by medical staff on the patient (e.g., examination using a stethoscope, examination of a specific part of the patient, pressing or tapping a specific part of the patient, etc.) from the medical image data, and can update the medical chart using the extracted information regarding medical acts.

[0241] As another example, if there is a medical record already written for the patient as additional medical data (e.g., a previous medical record written by medical staff and / or a medical record written by another institution, etc.), the computing device (100) can extract one or more pieces of information related to the patient and / or the treatment results for the patient from the medical record and can update the medical chart using the extracted information.

[0242] As another example, the computing device (100) can extract one or more pieces of information (e.g., image analysis results) by analyzing the medical image data, in the case where the additional medical data is medical image data (e.g., X-RAY, CT, MRI images, etc.) that has already been taken of the patient, and can update the medical chart using the extracted information.

[0243] At this time, if there is a pre-generated image reading result corresponding to the medical image data (e.g., a result of analyzing the medical image data by another institution or a result of analyzing the medical image data through a separate image analysis program), the computing device (100) can update the medical chart by comparing the pre-generated image reading result corresponding to the medical image data (hereinafter referred to as the 'first image reading result') with the image reading result derived by the computing device (100) independently analyzing the medical image data (hereinafter referred to as the 'second image reading result').

[0244] For example, the computing device (100) can update a medical chart using the first image reading result or the second image reading result when the first image reading result and the second image reading result match.

[0245] Meanwhile, if the computing device (100) determines that the first image reading result and the second image reading result do not match, it can update the medical chart using only the second image reading result, that is, the image reading result generated in correspondence with the medical image data.

[0246] In various embodiments, the computing device (100) updates a medical chart using information extracted from additional medical data, but may update the medical chart using only the information confirmed by the medical staff among the information extracted from the additional medical data. For example, when multiple pieces of information are extracted from additional medical data, the computing device (100) may provide multiple pieces of information to the medical staff, and when at least one piece of information among the multiple pieces of information is selected by the medical staff, it may update the medical chart using only at least one piece of information.

[0247] In various embodiments, the computing device (100) can retrain a generative artificial intelligence model using medical data and medical charts generated in correspondence with the medical data as training data.

[0248] At this time, the computing device (100) can retrain a generative artificial intelligence model using the medical data and the medical chart modified by the medical staff as training data in response to providing the medical chart to the medical staff, and when at least a part of the medical chart is modified by the medical staff.

[0249] FIG. 8 is a flowchart of a method for providing an estimated diagnosis recommendation service according to various embodiments.

[0250] Referring to FIG. 8, in step S810, the computing device (100) can obtain voice-form medical data generated by recording a voice conversation between the medical staff and the patient while the medical staff is providing medical care to the patient.

[0251] In step S820, the computing device (100) can generate an estimated diagnosis list using the voice-form medical data obtained through step S810.

[0252] Here, the estimated diagnosis list may refer to a list containing information on one or more diseases presumed to be present in the patient based on the content of conversations between medical staff and the patient.

[0253] More specifically, first, the computing device (100) can convert medical data in voice form into medical data in text form. For example, the computing device (100) can convert medical data in voice form into medical data in text form using a speech-to-text model, but is not limited thereto.

[0254] Subsequently, the computing device (100) can generate an estimated diagnosis list using medical data in text form. For example, the computing device (100) can generate a prompt including a command instructing to create a list of diseases presumed to be possessed by the patient from the medical data in text form, and can generate an estimated diagnosis list as result data by inputting the input data generated by combining the medical data in text form and the prompt into a generative artificial intelligence model.

[0255] In various embodiments, the computing device (100) can identify a voice corresponding to a medical professional from voice-type medical data, extract only the text corresponding to the identified voice, and generate an estimated diagnosis list using only the extracted text.

[0256] In step S830, the computing device (100) can determine whether there is additional medical data corresponding to the patient.

[0257] In step S840, if the computing device (100) determines through step S830 that there is no additional medical data corresponding to the patient, it may provide an estimated diagnosis recommendation service to the medical staff's terminal (200) that provides an estimated diagnosis list generated through step S820.

[0258] In various embodiments, the computing device (100) may provide a user interface (UI) for creating a medical chart to a medical staff terminal (200) and may provide an estimated diagnosis list through the UI. For example, as illustrated in FIG. 9, the computing device (100) may provide a popup window that provides an estimated diagnosis list and may provide disease information included in the estimated diagnosis list in a dropdown form through the popup window, but is not limited thereto.

[0259] In step S850, if the computing device (100) determines through step S830 that there is additional medical data corresponding to the patient, it can update the estimated diagnosis list based on information extracted from the additional medical data.

[0260] Here, updating the estimated diagnosis list may mean adding diseases estimated from additional medical data to the estimated diagnosis list, but is not limited thereto; depending on the case, it may mean replacing specific diseases included in the estimated diagnosis list with diseases estimated from additional medical data, or modifying and / or deleting specific diseases included in the estimated diagnosis list based on diseases estimated from additional medical data.

[0261] For example, if there is medical image data generated as additional medical data by medical staff filming the process of outpatient treatment of a patient, the computing device (100) can extract information regarding medical acts performed by medical staff on the patient from the medical image data (e.g., examination using a stethoscope, examination of a specific part of the patient, pressing or tapping a specific part of the patient, etc.) and can update the estimated diagnosis list using the extracted information regarding medical acts.

[0262] As another example, if there is a medical record created for a patient as additional medical data (e.g., a medical record created by another institution, etc.), the computing device (100) can extract one or more pieces of information related to the patient and / or the treatment results of the patient from the medical record and update the estimated diagnosis list using the extracted information.

[0263] At this time, if the computing device (100) has an analysis result of medical image data taken at another institution in a medical record created at another institution, it can update the estimated diagnosis list using the analysis result of medical image data taken at another institution.

[0264] As another example, the computing device (100) can extract one or more pieces of information (e.g., image analysis results) by analyzing the medical image data, in the case where the additional medical data is medical image data (e.g., X-RAY, CT, MRI images, etc.) taken of the patient at another institution, and can update the estimated diagnosis list using the extracted information.

[0265] In various embodiments, the computing device (100) updates the estimated diagnosis list using information extracted from additional medical data, but may update the estimated diagnosis list using only the information confirmed by the medical staff among the information extracted from the additional medical data. For example, when multiple pieces of information are extracted from the additional medical data, the computing device (100) may provide multiple pieces of information to the medical staff, and when at least one piece of information among the multiple pieces of information is selected by the medical staff, it may update the estimated diagnosis list using only at least one piece of information.

[0266] In various embodiments, the computing device (100) can retrain a generative artificial intelligence model using medical data and an estimated diagnosis list generated in response to the medical data as training data. For example, when an estimated diagnosis list containing multiple disease information is generated in response to the medical data, the computing device (100) can retrain a generative artificial intelligence model using one or more disease information selected by a medical professional among the multiple disease information and the medical data as training data.

[0267] At this time, the computing device (100) can retrain a generative artificial intelligence model using the medical data and the new disease information entered by the medical staff as training data in response to the provision of an estimated diagnosis list to the medical staff, when new disease information not included in the estimated diagnosis list is entered by the medical staff.

[0268] FIG. 10 is a flowchart of a method for providing a confirmed diagnosis recommendation service according to various embodiments.

[0269] Referring to FIG. 10, in step S1010, the computing device (100) can obtain test result data for the patient.

[0270] Here, test result data is data containing test results derived from performing tests on patients, for example, test result data may be medical imaging data such as MRI, X-ray, CT, etc., but is not limited thereto, and test result data may be blood test data or urine test data.

[0271] In addition, the tests performed on the patient herein may be determined based on the initial treatment results of the patient. However, they are not limited thereto.

[0272] In step S1020, the computing device (100) can generate a confirmed diagnosis list using the inspection result data obtained through step S1010.

[0273] Here, the confirmed diagnosis list may refer to a list containing information on one or more diseases determined to be possessed by the patient based on the patient's test result data.

[0274] In various embodiments, the computing device (100) can generate a prompt instructing to determine the patient's disease from the test result data, and can generate a confirmed diagnosis list as result data by inputting input data including the test result data and the prompt into a generative artificial intelligence model.

[0275] In various embodiments, the computing device (100) can analyze test result data to extract one or more features and generate a prompt instructing to determine a patient's disease based on one or more features, and can generate a confirmed diagnosis list as result data by inputting input data including test result data, one or more features and a prompt into a generative artificial intelligence model.

[0276] In step S1030, the computing device (100) can determine whether there is additional medical data corresponding to the patient.

[0277] In step S1040, if the computing device (100) determines through step S1030 that there is no additional medical data corresponding to the patient, it may provide a confirmed diagnosis recommendation service to the medical staff's terminal (200) that provides a confirmed diagnosis list generated through step S1020.

[0278] Here, the method of providing a confirmed diagnosis list may be implemented in the same or similar form as the method of providing an estimated diagnosis list (e.g., FIG. 9), but is not limited thereto.

[0279] In step S1050, if the computing device (100) determines through step S1030 that there is additional medical data corresponding to the patient, it can update the confirmed diagnosis list based on information extracted from the additional medical data.

[0280] Here, updating the confirmed diagnosis list may mean adding diseases estimated from additional medical data to the confirmed diagnosis list, but is not limited thereto; depending on the case, it may mean replacing specific diseases included in the confirmed diagnosis list with diseases determined from additional medical data, or modifying and / or deleting specific diseases included in the confirmed diagnosis list based on diseases determined from additional medical data.

[0281] For example, if there is additional medical data, such as voice-form medical data generated by recording a voice conversation between a medical staff member and a patient during the process of a medical staff member providing follow-up medical treatment to a patient, the computing device (100) can convert the voice-form medical data into text-form medical data, extract information related to disease diagnosis results from the text-form medical data, and update the confirmed diagnosis list based on the extracted information related to disease diagnosis results.

[0282] As another example, if there is medical image data generated as additional medical data by medical staff filming the process of outpatient treatment of a patient, the computing device (100) can extract information regarding medical acts performed by medical staff on the patient from the medical image data (e.g., examination using a stethoscope, examination of a specific part of the patient, pressing or tapping a specific part of the patient, etc.) and can update the confirmed diagnosis list using the extracted information regarding medical acts.

[0283] As another example, if there is re-examination chart record data containing details of a patient's re-examination as additional medical data, the computing device (100) can extract information related to the disease diagnosis results of the re-examination chart record data and update the confirmed diagnosis list based on the extracted information related to the disease diagnosis results.

[0284] In various embodiments, the computing device (100) can retrain a generative artificial intelligence model using medical data and a confirmed diagnosis list generated in response to the medical data as training data. For example, when a confirmed diagnosis list containing multiple disease information is generated in response to the medical data, the computing device (100) can retrain a generative artificial intelligence model using one or more disease information selected by a medical professional among the multiple disease information and the medical data as training data.

[0285] At this time, the computing device (100) can retrain a generative artificial intelligence model using the medical data and the new disease information entered by the medical staff as training data in response to the provision of a confirmed diagnosis list to the medical staff, when new disease information not included in the confirmed diagnosis list is entered by the medical staff.

[0286] FIG. 11 is a flowchart of a method for providing an inspection recommendation service according to various embodiments.

[0287] Referring to FIG. 11, in step S1110, the computing device (100) can obtain voice-form medical data generated by recording a voice conversation between the medical staff and the patient while the medical staff is providing medical care to the patient.

[0288] In step S1120, the computing device (100) can generate a test list using the voice-form medical data obtained through step S1110.

[0289] Here, the test list refers to a list generated based on the conversation between medical staff and the patient, and may mean a list containing information on one or more tests required to confirm one or more diseases presumed to be present in the patient.

[0290] More specifically, first, the computing device (100) can convert medical data in voice form into medical data in text form. For example, the computing device (100) can convert medical data in voice form into medical data in text form using a speech-to-text model, but is not limited thereto.

[0291] Subsequently, the computing device (100) can generate a list of tests using medical data in text form. For example, the computing device (100) can generate a prompt including a command instructing to create a list of tests required of a patient from the medical data in text form, and can generate a list of tests as result data by inputting the input data generated by combining the medical data in text form and the prompt into a generative artificial intelligence model.

[0292] In various embodiments, the computing device (100) can identify a voice corresponding to a medical professional from voice-type medical data, extract only the text corresponding to the identified voice, and generate a test list using only the extracted text.

[0293] In step S1130, the computing device (100) can determine whether there is additional medical data corresponding to the patient.

[0294] In step S1140, if the computing device (100) determines through step S1130 that there is no additional medical data corresponding to the patient, it may provide a test recommendation service to the medical staff's terminal (200) that provides a test list generated through step S1120.

[0295] In various embodiments, the computing device (100) may provide a user interface (UI) for creating a medical chart to a medical staff terminal (200) and may provide a list of tests through the UI. For example, as shown in FIG. 12, the computing device (100) may provide a pop-up window that provides a list of tests and may provide test items included in the list of tests in a drop-down form through the pop-up window, but is not limited thereto.

[0296] In step S1150, if the computing device (100) determines through step S1130 that there is additional medical data corresponding to the patient, it can update the test list based on information extracted from the additional medical data.

[0297] Here, updating the test list may mean, but is not limited to, adding new test items to the test list based on diseases estimated from additional medical data, and may, depending on the case, replace specific test items included in the test list with other test items, or modify and / or delete specific test items included in the test list.

[0298] For example, if a diagnosis result for a patient (e.g., an estimated diagnosis result and / or a confirmed diagnosis result) exists as additional medical data, the computing device (100) can extract information about the disease the patient has from the diagnosis result, determine a test item based on the extracted information about the disease the patient has, and update a test list using the determined test item.

[0299] As another example, if there is a medical record already written for a patient as additional medical data, the computing device (100) can extract one or more pieces of information related to the patient and / or the treatment results for the patient from the medical record, determine an examination item using the extracted information, and update the examination list using the determined examination item.

[0300] In various embodiments, the computing device (100) updates the test list using information extracted from additional medical data, but may update the test list using only the information confirmed by the medical staff among the information extracted from the additional medical data. For example, when multiple pieces of information are extracted from the additional medical data, the computing device (100) may provide multiple pieces of information to the medical staff, and when at least one piece of information among the multiple pieces of information is selected by the medical staff, it may update the test list using only at least one piece of information.

[0301] In various embodiments, the computing device (100) can retrain a generative artificial intelligence model using medical data and a list of tests generated in correspondence with the medical data as training data. For example, when a list of tests including multiple test items is generated in correspondence with the medical data, the computing device (100) can retrain a generative artificial intelligence model using one or more test items selected by medical staff among the multiple test items and the medical data as training data.

[0302] At this time, the computing device (100) can retrain a generative artificial intelligence model using medical data and the new test items entered by the medical staff as training data in response to the medical staff providing a test list to the medical staff, when new test items not included in the test list are entered by the medical staff.

[0303] FIG. 13 is a flowchart of a method for providing a treatment recommendation service according to various embodiments.

[0304] Referring to FIG. 13, in step S1310, the computing device (100) can obtain voice-form medical data generated by recording a voice conversation between the medical staff and the patient while the medical staff is providing medical care to the patient.

[0305] In step S1320, the computing device (100) can generate a treatment list using the voice-form medical data obtained through step S1310.

[0306] Here, the treatment list is a list generated based on the content of the conversation between the medical staff and the patient, and may refer to a list containing information on one or more treatments required to treat one or more diseases determined to be present in the patient.

[0307] More specifically, first, the computing device (100) can convert medical data in voice form into medical data in text form. For example, the computing device (100) can convert medical data in voice form into medical data in text form using a speech-to-text model, but is not limited thereto.

[0308] Subsequently, the computing device (100) can generate a list of tests using medical data in text form. For example, the computing device (100) can generate a prompt including a command instructing the creation of a list of treatments required for a disease determined to be possessed by the patient from the medical data in text form, and can generate a list of treatments as result data by inputting the input data generated by combining the medical data in text form and the prompt into a generative artificial intelligence model.

[0309] In various embodiments, the computing device (100) can identify a voice corresponding to a medical professional from voice-type medical data, extract only the text corresponding to the identified voice, and generate a treatment list using only the extracted text.

[0310] In step S1330, the computing device (100) can determine whether there is additional medical data corresponding to the patient.

[0311] In step S1340, if the computing device (100) determines through step S1330 that there is no additional medical data corresponding to the patient, it may provide a treatment recommendation service to the medical staff's terminal (200) that provides a treatment list generated through step S1320.

[0312] In various embodiments, the computing device (100) may provide a user interface (UI) for creating a medical chart to a medical staff terminal (200) and may provide a treatment list through the UI. For example, as illustrated in FIG. 14, the computing device (100) may provide a pop-up window that provides a treatment list and may provide treatment items included in the treatment list in a drop-down form through the pop-up window, but is not limited thereto.

[0313] In various embodiments, the computing device (100) may generate drug recommendation information corresponding to the patient and provide the drug recommendation information to the medical staff when a drug prescription is selected by the medical staff in response to the provision of a treatment list to the medical staff. For example, as illustrated in FIGS. 15 and 16, when a specific drug prescription is selected by the medical staff as a treatment method for the patient, the computing device (100) may generate and provide drug recommendation information to the medical staff, such as recommending a product for the corresponding drug ingredient or recommending another drug suitable for taking together with the specific drug.

[0314] In step S1350, if the computing device (100) determines through step S1330 that there is additional medical data corresponding to the patient, it can update the treatment list based on information extracted from the additional medical data.

[0315] Here, updating the treatment list may mean, but is not limited to, adding new treatment items to the treatment list based on diseases determined from additional medical data, and may, depending on the case, replace specific treatment items included in the treatment list with other treatment items, or modify and / or delete specific treatment items included in the treatment list.

[0316] For example, if medical data (e.g., medical papers, etc.) exists as additional medical data, the computing device (100) can extract information regarding treatment methods corresponding to the disease from the medical data and update the treatment list using the extracted information.

[0317] As another example, if test result data exists as additional medical data, the computing device (100) can extract information regarding a disease determined to be possessed by the patient from the test result data and update the treatment list based on the extracted information regarding the disease.

[0318] In various embodiments, the computing device (100) can update the treatment list based on the patient's basic information (e.g., age and gender, etc.). For example, the computing device (100) may exclude surgical treatment or replace surgical treatment with non-invasive treatments such as drug treatment, physical therapy, or radiation therapy if the patient is elderly.

[0319] In various embodiments, the computing device (100) updates the treatment list using information extracted from additional medical data, but may update the treatment list using only the information confirmed by the medical staff among the information extracted from the additional medical data. For example, when multiple pieces of information are extracted from the additional medical data, the computing device (100) may provide multiple pieces of information to the medical staff, and when at least one piece of information among the multiple pieces of information is selected by the medical staff, it may update the treatment list using only at least one piece of information.

[0320] In various embodiments, the computing device (100) can retrain a generative artificial intelligence model using medical data and a treatment list generated in response to the medical data as training data. For example, when a treatment list including multiple treatment items is generated in response to the medical data, the computing device (100) can retrain a generative artificial intelligence model using one or more treatment items selected by medical staff among multiple examination treatments and the medical data as training data.

[0321] At this time, the computing device (100) can retrain a generative artificial intelligence model using the medical data and the new treatment item entered by the medical staff as training data in response to the medical staff providing a treatment list to the medical staff, when a new treatment item not included in the treatment list is entered by the medical staff.

[0322] The aforementioned method for providing outpatient medical assistance services based on generative AI has been described with reference to the flowchart illustrated in the drawings. For the sake of simplicity, the method for providing outpatient medical assistance services based on generative AI has been illustrated and described using a series of blocks; however, the present disclosure is not limited to the order of the blocks, and some blocks may be performed in a different order or simultaneously than those illustrated and described in this specification. Furthermore, new blocks not described in this specification and drawings may be added, or some blocks may be deleted or modified.

[0323]

[0324] Hereinafter, with reference to FIGS. 17 to 19, a method for providing a generative artificial intelligence-based medical advisory service performed by a computing device (100) will be described.

[0325] FIG. 17 is a flowchart of a method for providing a generative artificial intelligence-based medical advisory service according to another embodiment of the present disclosure.

[0326] Referring to FIG. 17, in step S1710, the computing device (100) can acquire medical data about the patient.

[0327] Here, medical data may be data containing information regarding medical treatment, examinations, diagnoses, prescriptions, and treatments performed on a patient. For example, medical data may include medical records, examination records, diagnostic records, prescription records, and treatment records, and may be computerized and stored medical records, but is not limited thereto, and may include medical images of the patient (e.g., CT, X-ray, MRI, etc.).

[0328] In various embodiments, the computing device (100) may be connected to a hospital server via a network (400) to obtain medical data about a patient from the hospital server, but is not limited thereto, and may receive medical data directly from a patient and / or a user requesting medical advice about the patient.

[0329] Here, the method of automatically acquiring medical data from the hospital server can be performed only on patients who have consented to the provision of medical information, and to this end, a procedure may be carried out to obtain prior consent from the patient for the provision of medical information or to request the completion of a consent form for the provision of medical information.

[0330] In step S1720, the computing device (100) can provide medical advisory services using the medical data obtained through step S1710.

[0331] In various embodiments, the computing device (100) can derive result data by analyzing medical data through an artificial intelligence model and can provide medical advisory services using the result data.

[0332] Here, the artificial intelligence model may be a model trained using a large amount of medical data and medical materials (e.g., medical textbooks, papers, etc.) as training data, and the specific details are as described above.

[0333] Hereinafter, with reference to FIGS. 18 and FIGS. 19, a method by which a computing device (100) provides medical advisory services based on result data will be described in detail.

[0334] FIG. 18 is a flowchart illustrating a method of providing answers to user inquiries as a medical advisory service in various embodiments.

[0335] Referring to FIG. 18, in step S1810, the computing device (100) can generate patient summary information based on medical data.

[0336] In various embodiments, the computing device (100) can pre-set and store keywords to be extracted from medical data to form a database (e.g., keyword data related to diagnosis names, keyword data related to examination methods and results, keyword data related to treatment methods and effects, etc.), and can generate patient summary information for a specific patient by extracting and collecting keywords from the medical data of a specific patient based on the pre-set and stored data.

[0337] The patient summary information generated through this process may include the following information.

[0338] (1) Diagnosis name extracted from medical records included in medical data

[0339] (2) Test results extracted from test records included in medical data (e.g., results of non-medical imaging tests (blood test, urine test, etc.) and interpretation results of medical images (CT, X-ray, MRI, etc.)

[0340] (3) Total treatment period, patient symptoms by period, treatment history (medication / treatment / procedure / surgery, etc.), number of treatments, treatment duration, hospitalization status, treatment effect (subjective / objective content regarding treatment effect such as VAS, clinical score, muscle strength, range of motion, etc.) extracted from treatment records included in medical data

[0341] (4) Cost of treatment extracted from prescription records included in medical data

[0342] Here, even though medical images are included in the medical data, if the interpretation results of the medical images are not extracted from the examination records of the medical data, the interpretation results of the images can be derived by analyzing the medical images through a separate image analysis algorithm (e.g., an AI-based image analysis model), or the interpretation results of the images can be obtained from a consultant by requesting an interpretation of the medical images from the consultant, and these can be included in the patient summary information.

[0343] This patient summary information may be generated in the form of a report with a separate template for the purpose of providing it to users, but is not limited thereto.

[0344] In step S1820, the computing device (100) may obtain one or more queries from a user. Here, the queries obtained from the user may be queries regarding a patient, for example, related to the appropriateness of medical treatment, examination, treatment, and prescription performed on the patient, but are not limited thereto.

[0345] In step S1830, the computing device (100) can derive an answer to one or more queries obtained through step S1820.

[0346] In various embodiments, the computing device (100) can derive an answer to one or more queries using a generative artificial intelligence model.

[0347] More specifically, first, the computing device (100) can generate a first prompt instructing to derive an answer to one or more questions based on patient summary information.

[0348] Afterwards, the computing device (100) can generate first input data using patient summary information and a first prompt.

[0349] Afterwards, the computing device (100) can derive first result data by inputting first input data into a generative artificial intelligence model.

[0350] Here, the first result data derived through the generative artificial intelligence model may include, but is not limited to, answers to one or more queries derived based on patient summary information.

[0351] In various embodiments, the computing device (100) determines the reliability of an answer derived through a generative artificial intelligence model, and if the reliability is below a standard, it may derive a highly reliable answer or obtain a highly reliable answer from an external source.

[0352] More specifically, first, the computing device (100) can determine the reliability of the response of the generative artificial intelligence model.

[0353] In various embodiments, the computing device (100) can calculate a score corresponding to the reliability of the answer of the generative artificial intelligence model.

[0354] For example, the computing device (100) can generate a prompt that queries a score corresponding to the reliability of the answer derived through the generative artificial intelligence model, and input this into the generative artificial intelligence model to derive a score corresponding to the reliability of the answer of the generative artificial intelligence model.

[0355] As another example, the computing device (100) can derive one or more standard answers by inputting input data into one or more reliability evaluation models, and can calculate the similarity between the answer derived through the generative artificial intelligence model and one or more standard answers derived through one or more reliability evaluation models as a score corresponding to the reliability of the answer of the generative artificial intelligence model.

[0356] As another example, the computing device (100) may provide the answer of a generative artificial intelligence model to one or more pre-designated medical professional terminals (200) for evaluating the reliability of the answer, and as feedback thereon, may obtain a score corresponding to the answer of the generative artificial intelligence model from one or more medical professional terminals (200).

[0357] Afterwards, if the score corresponding to the reliability is less than the reference score, the computing device (100) may derive a highly reliable answer to a query obtained from a user or obtain one from an external source.

[0358] For example, when the score corresponding to the reliability is less than the reference score, the computing device (100) can select one of a plurality of medical staff in different fields based on patient summary information, and can derive an answer by inputting a prompt to a medical staff model corresponding to one of the plurality of medical staff models.

[0359] Here, the multiple medical staff model is a model generated individually corresponding to each of the multiple medical staff, and may be a generative artificial intelligence model trained using multiple medical data generated during the process in which each of the multiple medical staff performs medical examination, diagnosis, prescription, and treatment on a patient, result data derived from each of the multiple medical data, and medical data corresponding to the field of each of the multiple medical staff as training data.

[0360] As another example, when the score corresponding to the reliability is less than the reference score, the computing device (100) may select one of a plurality of medical staff in different fields based on patient summary information, provide one or more queries to the terminal (200) of one of the medical staff, and obtain an answer from the terminal (200) of one of the medical staff as a response to the one or more queries provided to the terminal (200) of one of the medical staff.

[0361] Afterwards, the computing device (100) can continuously improve the performance of the generative artificial intelligence model by retraining the generative artificial intelligence model using the answers derived as above or the answers obtained from medical staff and patient summary information as training data.

[0362] In step S1840, the computing device (100) can provide the first result data derived through step S1830 to the user.

[0363] In various embodiments, the computing device (100) may generate one or more recommended queries based on at least one of a query obtained from a user and an answer corresponding to the query obtained from the user, and may provide the user with one or more recommended queries and / or one or more recommended queries and an answer to the query obtained from the user together.

[0364] For example, the computing device (100) can generate one or more recommended queries based on a query obtained from a user and / or an answer corresponding to a query obtained from a user, or generate a prompt instructing to generate one or more recommended queries and derive an answer thereto, and input this into a generative artificial intelligence model, thereby generating one or more recommended queries and / or one or more recommended queries and an answer thereto through the generative artificial intelligence model.

[0365] As another example, the computing device (100) may define correlations between multiple queries in advance, and when a specific query is obtained from a specific user, it may select one or more queries related to the specific query as recommended queries based on the predefined correlations and provide them to the user. Additionally, the computing device (100) may derive answers to the one or more queries selected as recommended queries through a generative artificial intelligence model and provide them to the user together.

[0366] Here, the correlation between multiple queries may be determined based on the history of providing medical advisory services to multiple users. For example, the computing device (100) may define that query A and query B have an association if a pattern is identified in which query B is entered as a subsequent query after query A is entered, based on the history of providing medical advisory services to multiple users. When query A is obtained from a specific user, it may provide query B as a recommended query in addition to an answer to query A, or provide query B and an answer to query B.

[0367] FIG. 19 is a flowchart illustrating a method for providing results of determining the appropriateness of medical services as a medical advisory service in various embodiments.

[0368] Referring to FIG. 19, in step S1910, the computing device (100) can generate patient summary information based on medical data.

[0369] Here, the method for generating patient summary information may be implemented in the same or similar form as the method performed in step S1810 of FIG. 18, but is not limited thereto.

[0370] In step S1920, the computing device (100) can determine the appropriateness of each of the pre-set multiple items based on the patient summary information generated through step S1910.

[0371] Here, a plurality of pre-set items may include at least one of the necessity of medical treatment, treatment method, treatment method, cost, insurance terms and conditions and compliance with regulations, and the accuracy and consistency of records.

[0372] In various embodiments, the computing device (100) can determine the appropriateness of each of a plurality of pre-set items using a generative artificial intelligence model.

[0373] More specifically, first, the computing device (100) may generate a second prompt instructing a judgment on the appropriateness of each of the multiple items based on patient summary information. For example, the computing device (100) may generate a prompt instructing a judgment on whether there has been overtreatment of the patient based on patient summary information, but is not limited thereto.

[0374] In various embodiments, the computing device (100) may set a criterion for determining appropriateness for each of a plurality of items and generate a second prompt instructing to determine appropriateness for each of a plurality of items according to the criterion for determining appropriateness.

[0375] Here, the criteria for determining the appropriateness of each of the multiple items may be determined based on the patient's age, gender, and disease history (e.g., underlying diseases, etc.), but are not limited thereto.

[0376] Afterwards, the computing device (100) can generate second input data using patient summary information and a second prompt.

[0377] Afterwards, the computing device (100) can derive second result data by inputting second input data into a generative artificial intelligence model.

[0378] Here, the second result data derived through the generative artificial intelligence model may include the result of determining the appropriateness of at least one of the necessity of treatment, treatment method, treatment method, cost, compliance with insurance terms and regulations, and the accuracy and consistency of records based on patient summary information. For example, the result data may include whether each of the multiple items is appropriate, but is not limited thereto, and may include information on items among the multiple items that are judged to be inappropriate depending on the case.

[0379] In step S1930, the computing device (100) can provide the user with the appropriateness judgment result derived through step S1920.

[0380] In various embodiments, the computing device (100) may generate a recommendation query corresponding to at least one item when, based on the result of determining appropriateness, at least one of a plurality of items is determined to be inappropriate, and may guide the user to one or more recommendation queries or provide one or more recommendation queries and answers corresponding to one or more recommendation queries.

[0381] For example, the computing device (100) can generate one or more recommendation queries corresponding to at least one item, or generate a prompt instructing to generate one or more recommendation queries and derive an answer thereto, and input this into a generative artificial intelligence model, thereby generating one or more recommendation queries and / or one or more recommendation queries and an answer thereto through the generative artificial intelligence model.

[0382] As another example, the computing device (100) can store queries related to a plurality of items in advance by matching them to each of a plurality of items, and if a specific item is determined to be inappropriate, it can select the queries matched to the specific item as recommended queries, and can provide guidance on the selected recommended queries or provide the selected recommended queries and answers thereto.

[0383] Herein, the method for providing a generative artificial intelligence-based medical advisory service according to various embodiments of the present disclosure provides a service that provides medical advice on whether medical actions (diagnosis, medical treatment, examination, prescription, treatment, etc.) regarding a patient are appropriate based on the patient's medical data. In order to derive more accurate results, the patient's medical data is essential, and to this end, a procedure to obtain consent for the provision of medical information from the patient in advance may be performed.

[0384] However, in cases where medical advice is to be provided to a specific patient who has not consented to the provision of medical information, medical advice may be provided to the specific patient by selecting patients with characteristics similar to the specific patient from among multiple patients who have received medical advice, based on limited medical data obtained from the specific patient and the specific patient's personal information (e.g., information entered when signing up for insurance), and comparing the medical data of the selected patients with the medical data of the specific patient.

[0385] For example, based on a specific patient's personal information, patients with the same attributes as that specific patient are selected; based on the medical data of patients among the selected patients who have diseases similar to those of the specific patient (e.g., accidents, etc.), the average treatment cost for that disease is calculated; and it is possible to determine whether the treatment cost charged to the specific patient is appropriate based on whether the difference between the calculated treatment cost and the treatment cost charged to the specific patient is within the margin of error.

[0386] In this case, if the treatment costs charged to a specific patient are determined to be inappropriate, the patient may be notified of this and encouraged to consent to the provision of medical information to obtain more accurate results.

[0387] The aforementioned method for providing a generative AI-based medical advisory service has been described with reference to the flowchart illustrated in the drawings. For the sake of simplicity, the method for providing a generative AI-based medical advisory service has been illustrated and described using a series of blocks; however, the present disclosure is not limited to the order of the blocks, and some blocks may be performed in a different order or simultaneously than those illustrated and described in this specification. Furthermore, new blocks not described in this specification and drawings may be added, or some blocks may be deleted or modified.

[0388] Although embodiments of the present disclosure have been described above with reference to the attached drawings, those skilled in the art will understand that the present disclosure may be implemented in other specific forms without altering its technical concept or essential features. Therefore, the embodiments described above should be understood as illustrative in all respects and not restrictive.

Claims

1. In a method performed by a computing device, A step of acquiring medical data generated during the process of medical staff performing medical treatment on a patient; A step of selecting one or more medical staff models from among a plurality of medical staff models based on the medical data obtained above; A step of deriving result data by analyzing the acquired medical data through one or more selected medical staff models; and A step comprising providing medical guide information to the medical staff using the above-derived result data, Method for providing medical guidance using a generative AI-based medical professional model.

2. In Paragraph 1, The above multiple medical staff models are, A Generative Artificial Intelligence Model trained using multiple medical data generated during the process of treating patients by each of multiple different medical professionals, and result data derived from each of the multiple medical data, as training data, Method for providing medical guidance using a generative AI-based medical professional model.

3. In Paragraph 1, The step of selecting one or more of the above medical staff models is, A step of extracting disease information for the patient from the medical data obtained above; and A step comprising selecting one or more medical staff models among the plurality of medical staff models based on the extracted disease information, Method for providing medical guidance using a generative AI-based medical professional model.

4. In Paragraph 1, The step of selecting one or more of the above medical staff models is, A step of extracting one or more features related to the patient from the acquired medical data; A step of selecting one or more past patients having characteristics similar to the patient among a plurality of past patients based on one or more extracted features; and A step comprising selecting a medical team model trained based on the medical data of one or more selected past patients among the plurality of medical team models. Method for providing medical guidance using a generative AI-based medical professional model.

5. In Paragraph 1, The step of selecting one or more of the above medical staff models is, A step of calculating a score for each of the plurality of medical staff models based on feedback obtained corresponding to each of the plurality of medical staff models during a predetermined period; and Based on the above-determined score, the method includes the step of selecting one or more medical staff models among the plurality of medical staff models. Method for providing medical guidance using a generative AI-based medical professional model.

6. In Paragraph 5, The above feedback is, As a response obtained from the plurality of medical staff by providing the plurality of result data derived by inputting the plurality of medical staff models obtained during the aforementioned predetermined period into the plurality of medical staff models to the plurality of medical staff models, The having a plurality of result data including evaluation information regarding the plurality of result data and information regarding whether to modify the plurality of result data. Method for providing medical guidance using a generative AI-based medical professional model.

7. In Paragraph 1, Each of the above multiple medical staff models is, It includes a medical chart generation model, a diagnostic model, a test recommendation model, and a treatment recommendation model, The step of selecting one or more of the above medical staff models is, Based on the medical data obtained above, the method comprises the step of individually selecting one or more medical chart generation models, one or more diagnostic models, one or more test recommendation models, and one or more treatment recommendation models from the plurality of medical staff models. Method for providing medical guidance using a generative AI-based medical professional model.

8. In Paragraph 1, The step of selecting one or more of the above medical staff models is, A step of determining the type of service to be provided to the medical staff based on the type of medical data obtained above; and Based on the above-determined type, the method includes the step of selecting one or more medical staff models among the plurality of medical staff models. Method for providing medical guidance using a generative AI-based medical professional model.

9. In Paragraph 1, The step of selecting one or more of the above medical staff models is, The method includes the step of selecting one main medical team model and one or more sub-medical team models based on the acquired medical data. The step of deriving the above result data is, A step of deriving first result data by analyzing the acquired medical data through the selected main medical staff model; A step of deriving one or more second result data by analyzing the acquired medical data through one or more sub-medical staff models; and A step comprising verifying the derived first result data using one or more derived second result data. Method for providing medical guidance using a generative AI-based medical professional model.

10. Processor; Network interface; Memory; and It includes a computer program that is loaded into the memory and executed by the processor, The above computer program is, Instructions for acquiring medical data generated during the process of medical staff performing treatment on a patient; An instruction to select one or more medical staff models among a plurality of medical staff models based on the above-mentioned acquired medical data; Instructions for deriving result data by analyzing the acquired medical data through one or more selected medical staff models; and Including instructions for providing medical guide information to the medical staff using the above-derived result data, A computing device that performs a method of providing medical guidance using a generative artificial intelligence-based medical staff model.

11. Combined with a computing device, A step of acquiring medical data generated during the process of medical staff performing medical treatment on a patient; A step of selecting one or more medical staff models from among a plurality of medical staff models based on the medical data obtained above; A step of deriving result data by analyzing the acquired medical data through one or more selected medical staff models; and A computer program stored on a recording medium readable by a computing device for executing a method of providing medical guidance using a generative artificial intelligence-based medical staff model, comprising the step of providing medical guidance information to the medical staff using the above-derived result data.

Images