Method and system for addressing derived indications linked to primary indication by using artificial intelligence

Abstract

A method for addressing derived indications linked to primary indications by using artificial intelligence according to the present invention comprises the steps of: collecting patient information including information about patient's primary indications; generating, on the basis of the patient information, a prediction prompt requesting prediction of the possibility of occurrence of derived indications associated with the primary indications; processing the generated prediction prompt as an input to a pre-trained artificial intelligence model to acquire information about the possibility of occurrence of one or more derived indications corresponding to the primary indications; generating prescription information about the patient's primary indications and derived indications on the basis of the information about the possibility of occurrence of the derived indications; and transmitting the prescription information to at least one of a preset server or a terminal.

Description

Method and system for responding to derivative indications linked to primary indications using artificial intelligence

[0001] The present invention relates to a method and system for generating prescription information for corresponding to derivative indications linked to a patient's primary indication using artificial intelligence.

[0002] With the recent rapid advancement of artificial intelligence (AI) technology, various types of AI models capable of independently learning and reasoning from complex data have emerged. In particular, the medical field is actively developing AI-based technologies that comprehensively analyze diverse information—such as patients' biometric data, medical records, and imaging data—to predict disease risk and suggest treatment directions. These AI models are evolving beyond simple pattern recognition to a level where they can learn temporal changes and correlations to support personalized treatment.

[0003] For example, generative AI models based on Large Language Models (LLMs) used in the medical field can understand medical records and prescription information in the form of natural language. Furthermore, these AI models can be extended into a multimodal structure capable of processing medical text, biometric time-series data, and medical imaging data together, enabling them to analyze potential diseases and propose response methods.

[0004] With the advancement of such artificial intelligence technology, there is a growing demand in the medical field for methods to effectively manage patient health by utilizing AI to predict at an early stage the likelihood of a patient's existing indications progressing to other indications, and to suggest appropriate prescriptions and treatments based on the results. Here, "indication" refers to a symptom or clinical situation requiring specific treatment or examination, which can be understood as the user's disease or symptoms.

[0005] However, existing methods for responding to derivative indications uniformly apply predefined fixed rules, so they do not adequately reflect the progression status of the primary indication or changes in biometric information for each patient, and there was a problem of low precision in prophylactic prescriptions because they failed to consider the correlation between the primary indication and derivative indications or the risk trajectory over time.

[0006] Consequently, under existing methods for managing derivative indications, patients were more likely to receive prescriptions that did not align with the risks associated with their derivative indications, which could lead to reduced treatment continuation rates and satisfaction, as well as a failure to prevent preventable complications.

[0007] Accordingly, there is a need for technology that responds to derivative indications linked to the primary indication, which uses artificial intelligence to probabilistically predict the probability of derivative indications occurring for the primary indication from medical data and biometric information, and recommends prescription information and exercise therapy programs based on the predicted probability.

[0008] The present invention is intended to provide a method and system capable of responding to derivative indications linked to a patient's primary indication using artificial intelligence.

[0009] Specifically, the present invention aims to provide a method and system for responding to derivative indications linked to a primary indication using artificial intelligence, which can predict the likelihood of occurrence of derivative indications associated with a primary indication based on patient information using an artificial intelligence model.

[0010] Furthermore, the present invention aims to provide a method and system for responding to derivative indications linked to a primary indication using artificial intelligence, which can generate prescription information regarding a patient's primary indication and derivative indications based on the probability of occurrence of derivative indications associated with the primary indication predicted using an artificial intelligence model.

[0011] More specifically, the present invention aims to provide a method and system for responding to derivative indications linked to a primary indication using artificial intelligence capable of performing adjustments to an exercise therapy program to add exercise items for preventing or alleviating predicted derivative indications by updating existing prescription information for the primary indication.

[0012] To solve the problem described above, the present invention proposes a method for responding to derivative indications linked to a primary indication by analyzing patient information using artificial intelligence. The method for responding to derivative indications linked to a primary indication using artificial intelligence according to the present invention may include the steps of: collecting patient information including information regarding the patient's primary indication; generating a prediction prompt requesting the prediction of the probability of occurrence of a derivative indication associated with the primary indication based on the patient information; processing the generated prediction prompt as input to a pre-trained artificial intelligence model to obtain information on the probability of occurrence of at least one derivative indication corresponding to the primary indication; generating prescription information regarding the patient's primary indication and the derivative indication based on the information on the probability of occurrence of the derivative indication; and transmitting the prescription information to at least one of a pre-configured server and a terminal.

[0013] Furthermore, the step of obtaining information on the probability of occurrence of the derivative indication may include calculating a probability value for occurrence of each derivative indication through the previously trained artificial intelligence model, and the information on the probability of occurrence of the derivative indication may include probability information based on the probability value for occurrence of each derivative indication.

[0014] Furthermore, the step of generating the above prescription information involves generating a prescription information generation prompt that requests the generation of the above prescription information based on the above probability information, such that the intensity or content of the prescription for the above derivative indication varies in the above prescription information, and processing the above prescription information generation prompt as input to the above artificial intelligence model to obtain different prescription information according to the probability of occurrence of the above derivative indication.

[0015] Furthermore, the patient information includes the patient's medical data and the patient's biometric information collected from at least one sensor, and in the step of obtaining information on the probability of occurrence of the derivative indication, after the primary indication occurs, the change pattern of the biometric information is analyzed, and the change pattern of the biometric information and the medical data related to the primary indication are processed as input to the artificial intelligence model to predict the probability of occurrence of the derivative indication, and information on the probability of occurrence of the derivative indication can be generated based on the predicted probability of occurrence of the derivative indication.

[0016] Furthermore, the artificial intelligence model may be configured to analyze the correlation between the change pattern of the bio-information and the primary indication and the derivative indication, and to predict the probability of the derivative indication occurring based on the correlation.

[0017] Furthermore, the artificial intelligence model may be configured to predict the probability of the occurrence of the derivative indication by reflecting the variation characteristics of the bio-information according to the progression state or treatment response of the primary indication and analyzing the correlation between the variation characteristics and the derivative indication.

[0018] Furthermore, the step of generating the above prescription information involves updating existing prescription information for the main indication according to the possibility of the occurrence of the above derivative indication, and the update may include at least one of the type of medication, dosage, treatment cycle, monitoring cycle, and adjustment of the exercise therapy program.

[0019] Furthermore, the adjustment of the above exercise therapy program may be configured to change at least one of the type, intensity, frequency, and duration of the exercise.

[0020] Furthermore, in the step of generating the above prescription information, adjustments to the exercise therapy program are performed to add the above exercise items for preventing or alleviating the predicted above-mentioned derivative indications, and the adjustments to the exercise therapy program may be configured to determine the exercise type, intensity, frequency, and performance area of ​​the above-mentioned exercise items corresponding to the type of above-mentioned derivative indications.

[0021] Furthermore, the artificial intelligence model may be a deep learning-based neural network model trained to predict the likelihood of a derivative indication occurring by using the patient information, biometric information, and medical data as inputs.

[0022] Furthermore, the artificial intelligence model may be a deep learning model composed of at least one of a recurrent neural network (RNN), a long short-term memory network (LSTM), and a gated recurrent unit (GRU) for learning the time-series changes of the patient's biological information.

[0023] Furthermore, the artificial intelligence model includes a Large Language Model (LLM) structure for processing the prediction prompt expressed in natural language, and the LLM may be a model trained with a Transformer-based encoder-decoder structure.

[0024] Furthermore, the artificial intelligence model may be configured as a multimodal hybrid neural network structure that processes text information of the medical data, time-series data of the biometric information, and medical image data together.

[0025] Furthermore, the artificial intelligence model may be a reinforcement learning-based model that learns to continuously improve the performance of predicting derived indications by using feedback on the patient's treatment response or prediction accuracy as a reward signal.

[0026] Furthermore, the artificial intelligence model is composed of a hybrid structure combining a large-scale language model (LLM) and a time-series prediction deep learning model, wherein the LLM interprets the medical data and the prediction prompt, and the deep learning model analyzes the change pattern of the biological information, and can predict the probability of the occurrence of the derivative indication based on the combined result of the LLM and the deep learning model.

[0027] Meanwhile, a system for responding to derivative indications linked to a primary indication using artificial intelligence according to the present invention comprises, in an electronic device, a memory for storing instructions and at least one processor electrically connected to said memory, and when said instructions are executed by said at least one processor, said at least one processor collects patient information including information about the patient's primary indication, generates a prediction prompt requesting to predict the probability of occurrence of a derivative indication associated with said primary indication based on said patient information, processes said prediction prompt as input to a pre-trained artificial intelligence model to obtain information on the probability of occurrence of at least one derivative indication corresponding to said primary indication, generates prescription information for said patient's primary indication and said derivative indication based on the information on the probability of occurrence of said derivative indication, and transmits said prescription information to at least one of a pre-configured server and a terminal.

[0028] Meanwhile, the program is executed by one or more processes in an electronic device and is stored on a computer-readable recording medium, and the program may include instructions for performing the steps of: collecting patient information including information regarding the patient's primary indication; generating a prediction prompt requesting the prediction of the probability of occurrence of a derivative indication associated with the primary indication based on the patient information; processing the generated prediction prompt as input to a pre-trained artificial intelligence model to obtain information on the probability of occurrence of at least one derivative indication corresponding to the primary indication; generating prescription information regarding the patient's primary indication and the derivative indication based on the information on the probability of occurrence of the derivative indication; and transmitting the prescription information to at least one of a pre-configured server and a terminal.

[0029] The method and system for responding to derivative indications linked to a primary indication using artificial intelligence according to the present invention analyzes patient information based on artificial intelligence, enabling personalized prescription and management that considers the individual patient's condition. Through this, it is possible to predict in advance the likelihood of derivative indications linked to the patient's primary indication to perform preventive measures and manage the patient's health status more precisely.

[0030] Furthermore, the method and system for responding to derivative indications linked to a primary indication using artificial intelligence according to the present invention can probabilistically calculate the probability of occurrence of derivative indications correlated with the primary indication by integrating and analyzing the patient's medical data, biometric information, prescription history, etc. Through this, changes in the patient's condition can be quantitatively evaluated, and a treatment plan for additionally occurring indications can be established based on the predicted results, thereby enabling the preemptive prevention of the progression of indications.

[0031] Furthermore, the method and system for responding to derivative indications linked to a primary indication using artificial intelligence according to the present invention can adjust prescription information, such as the type of medication, dosage, treatment cycle, and monitoring cycle, based on the probability of occurrence of derivative indications predicted by the artificial intelligence model. Through this, medical staff can receive support for optimized prescription decisions tailored to individual patient prediction results and implement a prevention-oriented, personalized treatment system.

[0032] Furthermore, the method and system for responding to derivative indications linked to the primary indication using artificial intelligence according to the present invention can prevent or alleviate derivative indications by providing a patient-customized exercise therapy program in response to predicted derivative indications, thereby simultaneously improving the patient's recovery efficiency and treatment continuity.

[0033] FIG. 1 is a conceptual diagram illustrating a derivative indication response system linked to a primary indication using artificial intelligence according to the present invention.

[0034] FIG. 2 is a flowchart for explaining, in general, a method for responding to derivative indications linked to a primary indication using artificial intelligence according to the present invention.

[0035] FIGS. 3a to 3d are conceptual diagrams illustrating patient information and the process of collecting patient information according to the present invention.

[0036] FIG. 4 is a conceptual diagram illustrating the process of generating a prediction prompt according to the present invention.

[0037] FIG. 5 is a conceptual diagram illustrating the process of generating information on the possibility of occurrence of derivative indications using an artificial intelligence model according to the present invention.

[0038] FIGS. 6a and FIGS. 6b are conceptual diagrams illustrating the process of generating prescription information according to the present invention and transmitting it to at least one of a pre-configured server and terminal.

[0039] FIGS. 7a and FIGS. 7b are conceptual diagrams for explaining the process of adjusting an exercise therapy program according to the present invention.

[0040] Figure 8 is a conceptual diagram illustrating prescription information provided to a medical staff terminal.

[0041] FIG. 9 is a block diagram illustrating a computing system in which the present invention can be implemented.

[0042] FIGS. 10 and FIGS. 11 are block diagrams illustrating an embodiment of a computing device according to the present invention.

[0043] Hereinafter, embodiments disclosed in this specification will be described in detail with reference to the attached drawings. Identical or similar components are assigned the same reference number regardless of the drawing symbols, and redundant descriptions thereof will be omitted. The suffixes "module" and "part" used for components in the following description are assigned or used interchangeably solely for the ease of drafting the specification and do not have distinct meanings or roles in themselves. Furthermore, in describing the embodiments disclosed in this specification, if it is determined that a detailed description of related prior art could obscure the essence of the embodiments disclosed in this specification, such detailed description will be omitted. Additionally, the attached drawings are intended only to facilitate understanding of the embodiments disclosed in this specification; the technical concept disclosed in this specification is not limited by the attached drawings, and it should be understood that they include all modifications, equivalents, and substitutions that fall within the spirit and technical scope of the present invention.

[0044] Terms including ordinal numbers, such as first, second, etc., may be used to describe various components, but said components are not limited by said terms. These terms are used solely for the purpose of distinguishing one component from another.

[0045] When it is stated that one component is "connected" or "connected" to another component, it should be understood that while it may be directly connected or connected to that other component, there may also be other components in between. On the other hand, when it is stated that one component is "directly connected" or "directly connected" to another component, it should be understood that there are no other components in between.

[0046] A singular expression includes a plural expression unless the context clearly indicates otherwise.

[0047] In this application, terms such as “comprising” or “having” are intended to specify the existence of the features, numbers, steps, actions, components, parts, or combinations thereof described in the specification, and should be understood as not precluding the existence or addition of one or more other features, numbers, steps, actions, components, parts, or combinations thereof.

[0048] The present invention relates to a method and system for responding to derivative indications linked to a patient's primary indication using artificial intelligence. Specifically, the present invention relates to a method and system that enables personalized treatment and preventive management for a patient by predicting the probability of occurrence of derivative indications associated with the primary indication based on patient information through an artificial intelligence model, and by generating (or adjusting) prescription information regarding the patient's primary indication and derivative indications according to the prediction results.

[0049] According to the present invention, “indication” refers to a symptom or clinical situation requiring specific treatment or examination, which may be understood as the user’s disease or symptom. Specifically, “primary indication” may refer to a disease, symptom, or clinical condition that is initially diagnosed in the patient or is currently the subject of primary treatment. Additionally, according to the present invention, “derived indication” may refer to complications, secondary diseases, or associated symptoms that may occur secondarily or appear in association with the progression of the primary indication.

[0050] For the sake of convenience of explanation, the present invention focuses on “indications related to musculoskeletal disorders,” but is not necessarily limited thereto. As an example, the indications described in the present invention may include all indications arising from various diseases, including musculoskeletal disorders (e.g., cancer, diabetes, hypertension, etc.). The term “patient” described in the present invention may refer to a user experiencing pain due to an indication, and in the present invention, the terms “patient” and “user” may be used interchangeably.

[0051] Specifically, the present invention can predict the probability of a derivative indication associated with a primary indication based on patient information by utilizing a pre-trained artificial intelligence model. Here, the pre-trained artificial intelligence model may refer to an artificial intelligence model that generates information on the probability of occurrence of at least one derivative indication corresponding to the primary indication. Furthermore, the pre-trained artificial intelligence model can generate prescription information regarding the patient's primary indication and derivative indication based on the information on the probability of occurrence of the derivative indication.

[0052] More specifically, in the present invention, patient information can be collected from at least one of a pre-established database and a user terminal (10), and a prompt can be generated requesting the prediction of the likelihood of a derivative indication associated with a primary indication from the patient information.

[0053] In the present invention, a pre-trained artificial intelligence model generates information related to the probability of occurrence of a derivative indication corresponding to an input prompt, and based on the information on the probability of occurrence of the derivative indication, can generate prescription information regarding the patient's primary indication and derivative indication. Furthermore, in the present invention, the generated prescription information is transmitted to at least one of a pre-configured server and terminal, allowing medical staff or the patient terminal to view the prescription information in real time and perform the establishment of a treatment plan and prescription adjustment to address the derivative indication.

[0054] In the foregoing, a method for responding to derivative indications linked to a primary indication using artificial intelligence according to the present invention has been generally described, and this can be implemented by a derivative indication response system described below. Below, with reference to FIG. 1, a derivative indication response system linked to a primary indication using artificial intelligence according to the present invention will be described in detail. FIG. 1 is a conceptual diagram for explaining a derivative indication response system linked to a primary indication using artificial intelligence according to the present invention.

[0055] As illustrated in FIG. 1, a derivative indication response system linked to a primary indication using artificial intelligence according to the present invention (hereinafter referred to as the “derivative indication response system,” 100) may include at least one of a communication unit (110), a storage unit (120), and a control unit (130). At this time, the derivative indication response system (100) according to the present invention is not limited to the components described above and may further include components that perform the same or similar roles as the functions described in the specification.

[0056] Additionally, the derivative indication response system (100) according to the present invention may be composed of a device that performs the learning and inference described in the present invention. For example, the derivative indication response system (100) according to the present invention may be a machine learning model(s) that is trained by one or more machine learning algorithms on learning data. Furthermore, the machine learning model(s) trained in the inference stage of the derivative indication response system (100) may receive input data including one or more inference / prediction requests.

[0057] In this case, the machine learning algorithm or machine learning model(s) according to the present invention may include a deep learning algorithm or model utilizing a neural network. Furthermore, the learned machine learning model(s) according to the present invention may provide one or more inferences and / or prediction(s) as outputs in response to an inference / prediction request.

[0058] Accordingly, the trained machine learning model(s) according to the present invention may include one or more models of one or more machine learning algorithms. In one embodiment, the trained machine learning model(s) according to the present invention may use output inference models(s) and / or prediction(s) as input feedback. Furthermore, the trained machine learning model(s) may use past inference models as inputs for generating new inference models.

[0059] Meanwhile, the derivative indication response system (100) according to the present invention may be implemented as an application or software. The treatment program providing system (100) implemented as software may be downloaded via a program (e.g., Play Store) that allows the application to be downloaded on an electronic device (or terminal), or implemented via an initial installation program on an electronic device. In this case, the communication unit (110), storage unit (120), and control unit (130) according to the present invention may be utilized as components of the electronic device. The electronic device according to the present invention may refer to at least one of a user terminal (10) owned by a patient (or user) and a medical staff terminal (30) owned by medical staff.

[0060] In the present invention, a terminal may also be referred to as a 'mobile terminal' or 'electronic device,' and the terminals described in this specification may include mobile phones, smartphones, smart TVs, laptop computers, digital broadcasting terminals, PDAs (personal digital assistants), PMPs (portable multimedia players), navigation systems, slate PCs, tablet PCs, ultrabooks, wearable devices (e.g., smartwatches, smart glasses, head-mounted displays), etc.

[0061] In the present invention, an electronic device (or terminal) can be understood to mean an application installed on at least one of a user terminal (10) and a medical staff terminal (30). Such an application (or software) can be understood as a component of a derivative indication response system (100) according to the present invention. More specifically, the electronic device according to the present invention is not limited to an electronic device in which an application is activated, but may mean an electronic device connected to an electronic device in which an application is activated. As an example, based on the fact that the electronic device according to the present invention is a smartphone, the electronic device may mean a smart watch connected to said smartphone.

[0062] As another example, the electronic device according to the present invention may refer to a medical device connected to a smartphone. Specifically, the electronic device may refer to a medical device comprising at least one of an electrocardiogram (ECG) measuring device, a pulse oximeter (SpO₂) measuring device, a blood pressure monitor, a thermometer, an electromyogram (EMG) measuring device, a skin electrical response (GSR) sensor device, a heart rate variability (HRV) measuring device, a respiration rate detector, a weighing scale, a muscle strength measuring device, a range of motion (ROM) measuring device, an infrared temperature detector, a skin temperature detector, and an ultrasonic sensor for imaging.

[0063] Meanwhile, the derivative indication response system (100) may exist inside a server (hereinafter referred to as the server) established to perform a specific purpose (e.g., a function for responding to derivative indications linked to the main indication) separately from the electronic device, or it may exist as a system separate from the server. When the derivative indication response system (100) exists inside the server, it may provide various services related to the present invention (e.g., generating prescription information regarding the patient's main indication and derivative indications) through at least one component located inside the server or configuration modules that perform functions similar to said components. In this case, the application may provide various services related to the present invention on the electronic device on which the application is installed through communication with the server.

[0064] In addition, the derivative indication response system (100) can be linked with a central server or an external server to implement a derivative indication response method linked to a primary indication using artificial intelligence according to the present invention, and can transmit the generated prescription information to a pre-configured server and terminal (or electronic device).

[0065] Meanwhile, the patient can check prescription information regarding the main indication and the derivative indication through an application or webpage provided by the derivative indication response system (100) according to the present invention, and receive an exercise therapy program according to the prescription information.

[0066] At this time, the patient may have a user account registered in the derivative indication response system (100). For convenience of explanation, the account of the patient user is referred to as the “patient account” or “user account”.

[0067] The “account” described above can be created through a page linked to the derivative indication response system (100) examined above. Alternatively, the “account” can also be created in at least one other system linked to the derivative indication response system (100).

[0068] Alternatively, an ‘account’ may be created on at least one other server (e.g., a medical staff server, 20) linked to the derivative indication response system (100) according to the present invention. In this case, the “medical staff (D)” described in the present invention is a person employed at a medical institution (e.g., a hospital) and may include, for example, at least one of a doctor, a nurse, or a physical therapist. For convenience of explanation, the present invention describes doctors and physical therapists as examples of medical staff. However, medical staff are not limited to the examples described, and any user employed at a medical institution or related institution to provide prescription information regarding the primary indication and derivative indication may be considered as medical staff according to the present invention.

[0069] A medical professional according to the present invention may possess a medical professional account already registered in the derivative indication response system (100) according to the present invention. In this specification, an electronic device logged in with a medical professional account is described as a medical professional terminal (30). As an example, the derivative indication response system (100) according to the present invention may receive prescription information prescribed by a medical professional (D) to a user (U) by linking with a medical professional server (20).

[0070] Accordingly, in this specification, without distinguishing the server on which the account was issued, all accounts based on the derivative indication response system (100) according to the present invention are referred to as “accounts already registered in the derivative indication response system (100) according to the present invention.”

[0071] Meanwhile, according to the present invention, the communication unit (110) may be connected via a wireless or wired network to a user terminal (10), a medical staff server (20), a medical staff terminal (30), an artificial intelligence server (140), a central server, a device, and at least one network, and configured to receive or transmit overall data and information necessary for the operation of the derivative indication response system (100) according to the present invention. Specifically, the communication unit (110) may receive patient information of a user (U, or patient) from at least one of the medical staff server (20) and the medical staff terminal (30) in order to generate prescription information for the main indication and the derivative indication.

[0072] For example, the communication unit (110) may receive the patient's biometric information from the user terminal (10). At this time, the biometric information may refer to information related to the patient's physical condition collected using at least one sensor provided in the user terminal (10). As an example, the user terminal (10) according to the present invention may be equipped with at least one sensor among a heart rate sensor, a pulse sensor, a body temperature sensor, an oxygen saturation (SpO₂) sensor, a heart rate variability (HRV) sensor, a respiratory rate sensor, a blood pressure sensor, a galvanic skin response (GSR) sensor, an accelerometer, a gyroscope, a posture detection sensor, a skin temperature sensor, and a step count detection sensor.

[0073] Furthermore, the communication unit (110) can receive user queries input through the user terminal (10). Specifically, the communication unit (110) can receive user queries input through a service page provided to the user terminal (10) in order to collect patient information. Here, “receiving user queries” may mean receiving an input signal (or selection signal) corresponding to a user query based on the user query being input by the user through the user terminal (10).

[0074] According to the present invention, a user query may include at least one of a document, text, an image (or video), and voice. In this case, the derived indication response system (100) may further include a module for converting voice into text. For example, the derived indication response system (100) may include a voice recognition model capable of analyzing voice data corresponding to voice received through a microphone provided in a user terminal (10). For example, the voice recognition model may convert voice into text based on at least one of a STT (Speech-to-Text) algorithm, HMM (Hidden Markov Model), HMM-GMM (Hidden Markov Model-Gaussian Mixture Model), a CTC (Connectionist Temporal Classification) based model, a Beam Search based model, a DNN (Deep Neural Network) based model, an RNN (Recurrent Neural Network) based model, a Seq2Seq (Sequence-to-Sequence) model, and a Transformer based model.

[0075] Furthermore, the communication unit (110) can transmit the generated prescription information for the main indication and derivative indication to at least one of a pre-configured server and terminal. For example, the communication unit (110) can transmit the generated prescription information for the main indication and derivative indication to at least one of a medical staff server (20) and a medical staff terminal (30). As another example, the communication unit (110) can transmit the generated prescription information for the main indication and derivative indication to a user terminal (10).

[0076] Meanwhile, the communication unit (110) may include at least one communication module capable of wireless communication and wired communication between the derivative indication response system (100) and the communication target. Additionally, the communication unit (110) may include a communication module that connects the derivative indication response system (100) to at least one network.

[0077] Meanwhile, the communication unit (110) can support various communication methods depending on the communication standard of the communicating device. For example, the communication unit (110) may be configured to perform communication using at least one of the following technologies: WLAN (Wireless LAN), Wi-Fi (Wireless-Fidelity), Wi-Fi (Wireless Fidelity) Direct, DLNA (Digital Living Network Alliance), WiBro (Wireless Broadband), WiMAX (World Interoperability for Microwave Access), HSDPA (High Speed ​​Downlink Packet Access), HSUPA (High Speed ​​Uplink Packet Access), LTE (Long Term Evolution), LTE-A (Long Term Evolution-Advanced), 5G (5th Generation Mobile Telecommunication), Bluetooth (Bluetooth™ Frequency Identification), Infrared Communication (Infrared Data Association; IrDA), UWB (Ultra-Wideband), ZigBee, NFC (Near Field Communication), Wi-Fi Direct, and Wireless USB (Wireless Universal Serial Bus).

[0078] Next, the storage unit (120) may be configured to store various information related to the present invention. In the present invention, the storage unit (120) may be provided in the derivative indication response system (100) itself, or alternatively, at least a part of the storage unit (120) may mean a database (Database: DB, 200).

[0079] The storage unit (120) may include one or more non-transient computer-readable storage media that can be read and / or accessed by at least one processor. One or more computer-readable storage media may include volatile and / or non-volatile storage components such as optical, magnetic, organic, or other memory or disk storage devices. In some examples, the storage unit (120) may be implemented using a single physical device (e.g., one optical, magnetic, organic, or other memory or disk storage device), whereas in other examples, the storage unit (120) may be implemented using multiple physical devices.

[0080] The storage unit (120) may include computer-readable instructions and additional data. The storage unit (120) may include a storage necessary to perform at least some of the methods and techniques described herein and / or at least some of the functions of the device and network.

[0081] Furthermore, at least a portion of the storage unit (120) may be a cloud storage or a cloud server. That is, the storage unit (120) is sufficient as long as it is a space where information necessary for the operation of the derivative indication response system (100) according to the present invention is stored, and it can be understood that there are no restrictions on the physical space. Accordingly, the storage unit (120) and the database (200) may be used interchangeably without being separately distinguished below.

[0082] Meanwhile, patient information including at least one of medical data and biometric information may be stored in the storage unit (120). As an example, the medical data according to the present invention may include at least one of the patient's (or user's) age, the user's gender, the user's physical information (e.g., height, weight, etc.), the user's medical history, information on the main indication, medical imaging data, and prescription information.

[0083] Specifically, the prescription information stored in the storage unit (120) may include information regarding an exercise therapy program. In relation to the exercise therapy program, the storage unit (120) may store at least one of the following: the name of each of a plurality of exercise items constituting the exercise therapy program, the number of exercises, the timing of the exercises, and the difficulty level of the exercises, as well as an exercise video and an exercise description corresponding to each of the plurality of exercise items.

[0084] Additionally, the storage unit (120) according to the present invention may store biometric information received from the user terminal (10). At this time, the biometric information may be very diverse and may include all physiological and biological signals that can be sensed in relation to the patient's physical condition. As an example, the biometric information according to the present invention may include physiological signals such as heart rate, blood pressure, body temperature, respiratory rate, oxygen saturation, electrocardiogram, electromyogram, movement data, sleep patterns, or skin conductivity.

[0085] The bio-information according to the present invention is not limited to the examples described above and may include all forms of measurable signals or data capable of representing a patient's physical functions, physiological responses, or behavioral characteristics. Furthermore, the bio-information may be extended to various forms of data collected from newly developed sensors, wearable devices, or medical measuring devices (or medical devices) for monitoring the patient's condition.

[0086] Meanwhile, the database (DB, 200) may be configured to store various information for generating prescription information for primary indications and derivative indications. Specifically, the database (200) may store patient information for each of multiple users (or patients). For example, the database (200) may store at least one of medical data and biometric information corresponding to each of multiple users.

[0087] Furthermore, the database (200) may store exercise motion content (e.g., exercise name, number of exercises, timing of exercises, difficulty of exercises, etc.) corresponding to each of multiple body parts (e.g., shoulder, elbow, wrist & hand, hip & pelvis, knee, ankle & foot, neck, back, waist, abdomen) related to the main indication. Furthermore, the database (200) may store at least one of an exercise video and an exercise description corresponding to each exercise motion content.

[0088] Additionally, the database (200) may store derivative indication information for at least one derivative indication that is pre-matched to a primary indication. Specifically, the database (200) may store derivative indication information including at least one of a list of derivative indications whose association with each of a plurality of primary indications has been pre-verified, information on the probability of occurrence of derivative indications, correlation coefficients, risk levels, and information on exercise items corresponding to each of the types of multiple derivative indications.

[0089] In this case, the derivative indications whose association with the primary indication has been verified in advance in the present invention, and the exercise items corresponding to each type of multiple derivative indications, may refer to pairs of indications whose clinical evidence and statistical correlation have been verified by a group of experts, such as medical specialists, clinical data analysts, and medical statistics experts. That is, the association between the primary indication and the derivative indications can be established in advance through medical big data analysis, clinical trial results, electronic medical record (EMR) statistics, or literature-based verification procedures.

[0090] Additionally, the storage unit (120) may store prescription information generation rules matched to pre-set probability intervals associated with the probability value of the occurrence of a derived indication. At this time, the prescription information generation rules matched to each of the multiple probability intervals may be predefined to generate prescription information such that the intensity or content of the prescription for one of the adjustments to the type of medication, the dosage, the treatment cycle, the monitoring cycle, and the exercise therapy program varies.

[0091] In this case, the prescription information generation rule in the present invention may refer to a rule related to prescription adjustment criteria derived by a group of experts, such as medical specialists, clinical data analysts, and medical statistics experts. That is, a prescription information generation rule matched to a pre-established probability interval associated with the probability value of the occurrence of a derived indication may be established in advance through medical big data analysis, clinical trial results, electronic medical record (EMR) statistics, or literature-based verification procedures.

[0092] In the present invention, the storage unit (120) is described as existing separately from the database (200), but is not limited thereto, and the storage unit (120) may include the database (200).

[0093] Meanwhile, data and commands necessary for the operation of the derivative indication response system (100) according to the present invention may be stored in the storage unit (120). Specifically, commands for the operation of the prompt generation unit (131) may be stored in the storage unit (120).

[0094] For example, the prompt generation unit (131) according to the present invention may mean a module capable of generating a prediction prompt that requests the prediction of the likelihood of occurrence of a derivative indication associated with a primary indication based on patient information. As another example, the prompt generation unit (131) according to the present invention may mean a module capable of generating a prescription information generation prompt that requests the generation of prescription information for the primary indication and the derivative indication based on information regarding the likelihood of occurrence of the derivative indication.

[0095] For example, the prompt generation unit (131) may include at least one of a large language model based on T5 (Text-to-Text Transfer Transformer), BART (Bidirectional and Auto-Regressive Transformer), GPT (Generative Pre-trained Transformer), or LLaMA (Language Model for Many Applications), a rule-based template matching algorithm, a conditional prompting technique, a contextual embedding selection module, or a few-shot prompt generator.

[0096] At this time, the model or algorithm included in the prompt generation unit (131) according to the present invention is not limited to the examples described above, and may further include artificial intelligence models and algorithms having the same function. The prompt generation unit (131) according to the present specification is not limited in type and method as long as it is a module capable of generating a prediction prompt requesting the prediction of the probability of occurrence of a derivative indication associated with a primary indication based on patient information, and a prescription information generation prompt requesting the generation of prescription information for the primary indication and the derivative indication based on the probability information of occurrence of the derivative indication.

[0097] Furthermore, the storage unit (120) according to the present invention may store commands for the operation of a pre-trained artificial intelligence model (132). Here, the pre-trained artificial intelligence model (132) according to the present invention may refer to an artificial intelligence model that generates information corresponding to a prompt based on an input prompt. For example, the pre-trained artificial intelligence model (132) according to the present invention may refer to a deep learning-based neural network model that is trained to predict the probability of occurrence of a derivative indication by taking patient information, biometric information, and medical data as inputs. As an example, the pre-trained artificial intelligence model (132) according to the present invention may refer to a deep learning model composed of at least one of a recurrent neural network (RNN), a long short-term memory network (LSTM), and a gated recurrent unit (GRU) for learning the time-series changes of a patient's biometric information. More specifically, the pre-trained artificial intelligence model (132) according to the present invention may refer to a deep learning model composed of a derivative neural network structure of the RNN family, such as a bidirectional LSTM (Bi-LSTM), a stacked LSTM (Stacked LSTM), a deep GRU (Deep GRU), and an attention-based RNN (Attention-based RNN).

[0098] Additionally, the previously trained artificial intelligence model (132) may include at least one of a convolution-RNN hybrid structure, a transformer-RNN hybrid structure, and a sequence-to-sequence (Seq2Seq) structure as needed, and may refer to a deep learning model trained to predict the likelihood of a derivative indication occurring by taking patient information, biometric information, and medical data as input.

[0099] Alternatively, the pre-trained artificial intelligence model (132) according to the present invention includes a Large Language Model (LLM) structure for processing prediction prompts expressed in natural language, and the LLM (or Large Language Model, Large Language Model) may refer to a model trained with a Transformer-based encoder-decoder structure. For example, the Large Language Model may refer to at least one artificial intelligence model based on at least one of GPT (Generative Pre-trained Transformer), Gemini, T5 (Text-to-Text Transfer Transformer), Vision-Language Models (e.g., CLIP, Flamingo), Multimodal Generative Models (GPT-4 Multimodal, PaLM-E), BERT (Bidirectional Encoder Representations from Transformers), and LaMDA (Language Model for Dialogue Applications).

[0100] As another example, the pre-trained artificial intelligence model (132) according to the present invention may be configured as a multimodal hybrid neural network structure that processes text information of medical data, time-series data of biometric information, and medical image data together. In this case, the pre-trained artificial intelligence model (132) may refer to an artificial intelligence model configured to include at least one of a Transformer-based language model for text analysis, a Recurrent Neural Network (RNN) for time-series pattern analysis, a Long Short-Term Memory Network (LSTM), a Gated Recurrent Unit (GRU), a Convolutional Neural Network (CNN) for image data processing, and a Vision Transformer (ViT).

[0101] As another example, the pre-trained artificial intelligence model (132) according to the present invention may refer to a reinforcement learning-based model that learns to continuously improve the performance of predicting derived indications by using feedback on the patient's treatment response or prediction accuracy as a reward signal. In this case, the pre-trained artificial intelligence model (132) may refer to an artificial intelligence model configured to include at least one reinforcement learning algorithm among Deep Q-Network (DQN), Double Q-Network (DQN), Prioritized Experience Replay, Policy Gradient, Actor-Critic, Proximal Policy Optimization (PPO), Deep Deterministic Policy Gradient (DDPG), Twin Delayed DDPG (TD3), and Advantage Actor-Critic (A2C / A3C).

[0102] As another example, the artificial intelligence model (132) according to the present invention is composed of a hybrid structure combining a large-scale language model (LLM) and a deep learning model for time series prediction, wherein the LLM (or large-scale language model, large language model) interprets medical records and prediction prompts, and the deep learning model analyzes the change patterns of biometric information, and the artificial intelligence model is trained to predict the probability of occurrence of the derivative indication based on the combined result of the LLM and the deep learning model.

[0103] In this case, a large-scale language model may refer to at least one artificial intelligence model based on at least one of GPT (Generative Pre-trained Transformer), Gemini, T5 (Text-to-Text Transfer Transformer), Vision-Language Models (e.g., CLIP, Flamingo), Multimodal Generative Models (GPT-4 Multimodal, PaLM-E), BERT (Bidirectional Encoder Representations from Transformers), and LaMDA (Language Model for Dialogue Applications).

[0104] In addition, a deep learning model for time series prediction may refer to an artificial intelligence model comprising at least one of a recurrent neural network (RNN), a long short-term memory network (LSTM), a gated recurrent unit (GRU), a one-dimensional convolutional neural network (1D-CNN), a time series transformer, a temporal convolutional network (TCN), and an attention-based time series prediction model. Furthermore, the deep learning model for time series prediction may be extended as needed to at least one of a Seq2Seq (Sequence-to-Sequence) structure, a Bi-LSTM (Bidirectional LSTM), a stacked LSTM, and a combined CNN-LSTM structure.

[0105] That is, the pre-trained artificial intelligence model (132) according to the present invention can be implemented in a wide variety of configurations, and the present invention should be understood to include all artificial intelligence models that can be configured by applying various neural network structures or algorithms according to the form of input data, the purpose of prediction, or the learning method, rather than limiting the pre-trained artificial intelligence model (132) to the examples described above. Furthermore, the pre-trained artificial intelligence model (132) can be modified or expanded in various ways within the scope of the technical concept of the present invention.

[0106] Next, the control unit (130) may be configured to control the overall operation of the derivative indication response system (100) related to the present invention. Specifically, the control unit (130) may include at least one of a prompt generation unit (131) and a pre-trained artificial intelligence model (132). The control unit (130) may process signals, data, information, etc. that are input or output through the components described above, or provide or process appropriate information and functions to the user.

[0107] The control unit (130) can control the output of a service page for collecting patient information and providing prescription information through a display unit (or touchscreen) provided on the user terminal (10) and the medical staff terminal (30). Such a service page may be output on the user terminal (10) and the medical staff terminal through an application or web page installed on the user terminal (10) and the medical staff terminal (30). The service page is a page linked to the derivative indication response system (100) according to the present invention and is configured to be controlled by the derivative indication response system (100) according to the present invention.

[0108] Furthermore, when the service page is provided in the form of an application, the service page may be controlled by at least one CPU (Central Processing Unit) among the user terminal (10) and the medical staff terminal (30) on which the application is installed. In this case, at least one CPU among the user terminal (10) and the medical staff terminal (30) may transmit generated prescription information to a pre-configured server and terminal based on information provided by the derivative indication response system (100) according to the present invention.

[0109] Meanwhile, the control unit (130) can collect patient information from the storage unit (120, or database (200)). Furthermore, the control unit (130) can use the prompt generation unit (131) to generate a prediction prompt requesting the prediction of the probability of occurrence of a derivative indication associated with the main indication based on the patient information. Specifically, the prompt generation unit (131) can generate a prediction prompt requesting the calculation of the probability value of occurrence for each derivative indication associated with the main indication.

[0110] Furthermore, the control unit (130) can process the prediction prompt as input to a pre-trained artificial intelligence model (132). The control unit (130) can obtain information on the probability of occurrence of at least one derivative indication corresponding to the primary indication through the pre-trained artificial intelligence model (132). Specifically, the control unit (130) can obtain probability values ​​for occurrence of each derivative indication through the pre-trained artificial intelligence model (132).

[0111] In the present invention, the derivative indication response system (100) is described as including a pre-trained artificial intelligence model (132), but is not limited thereto, and the pre-trained artificial intelligence model (132) according to the present invention may exist within a separate external artificial intelligence server (140). For example, if the pre-trained artificial intelligence model (132) includes a Large Language Model (LLM) structure for processing prediction prompts expressed in natural language, the large language model may be included in an artificial intelligence server (140, e.g., LLM server). At this time, the control unit (130) can calculate the probability of occurrence for each derivative indication associated with the main indication through the pre-trained artificial intelligence model (132) in conjunction with the artificial intelligence server (140).

[0112] That is, it can be understood that in the present invention, the control unit (130) may perform calculations by directly driving a pre-trained artificial intelligence model (132) included in the derivative indication response system (100), or may perform calculations by linking with a pre-trained artificial intelligence model (132) included in an external artificial intelligence server (140) through a network.

[0113] Accordingly, the control unit (130) can selectively utilize a pre-trained artificial intelligence model (132) inside or outside the system, thereby performing one of the local inference and cloud inference methods depending on the operating environment, computational resources, or data processing purpose.

[0114] Additionally, although the present invention describes the artificial intelligence server (140) as existing separately from the derivative indication response system (100), it is not limited thereto, and the derivative indication response system (100) may be configured to include the artificial intelligence server (140). That is, the derivative indication response system (100) and the artificial intelligence server (140) according to the present invention may exist separately, or the artificial intelligence server (140) may be included within the derivative indication response system (100).

[0115] Furthermore, the control unit (130) can generate prescription information for the patient's primary indication and derivative indication based on information regarding the probability of occurrence of derivative indications. Specifically, the prompt generation unit (131) can generate a prescription information generation prompt requesting the generation of prescription information such that the strength or content of the prescription for derivative indications changes based on probability information included in the information regarding the probability of occurrence of derivative indications.

[0116] The control unit (130) processes the prescription information generation prompt as input to a pre-trained artificial intelligence model (132), and can obtain different prescription information according to the probability of occurrence of a derivative indication through the pre-trained artificial intelligence model (132).

[0117] Furthermore, the control unit (130) can transmit the acquired prescription information to at least one of a pre-configured server and terminal. For example, the control unit (130) can transmit the acquired prescription information to at least one of a medical staff server (20) and a medical staff terminal (30). Specifically, the control unit (130) can output the acquired prescription information to a service page provided to the medical staff terminal (30).

[0118] Meanwhile, the derivative indication response system (100) may include one or more processors, and such processors may include one or more general-purpose processors and / or one or more special-purpose processors (e.g., digital signal processors, tensor processing units (TPUs), graphics processing units (GPUs), neural network processing units (NPUs), application integrated circuits, application semiconductors (ASICs), etc.). One or more processors may be configured to execute instructions, computer-readable instructions, and / or other instructions described herein that are stored (or included) in the storage unit (120). Such a derivative indication response system (100) may perform data processing described below in cooperation with memory and at least one processor. The processor may be electrically connected to memory and may perform a series of operations and data processing using data and information stored in memory. Here, “memory” may be a component of the storage unit (120), and “processor” may be used interchangeably with the control unit (130).

[0119] In the above description, the derivative indication response system (100) of the present invention has been described, and it can be implemented based on the derivative indication response method linked to the main indication using artificial intelligence described below.

[0120] Hereinafter, with reference to FIG. 2 together with FIG. 3a to 3d, FIG. 4, FIG. 5, FIG. 6a, FIG. 6a, FIG. 7a, FIG. 7b, and FIG. 8, a method for responding to derivative indications linked to a primary indication using artificial intelligence according to the present invention will be described in more detail. FIG. 2 is a flowchart for generally explaining the method for responding to derivative indications linked to a primary indication using artificial intelligence according to the present invention, and FIG. 3a to 3d are conceptual diagrams for explaining patient information and the process of collecting patient information according to the present invention. FIG. 4 is a conceptual diagram for explaining the process of generating a prediction prompt according to the present invention, and FIG. 5 is a conceptual diagram for explaining the process of generating information on the probability of occurrence of derivative indications using an artificial intelligence model according to the present invention. FIGS. 6a and 6b are conceptual diagrams illustrating the process of generating prescription information according to the present invention and transmitting it to at least one of a pre-configured server and terminal, FIGS. 7a and 7b are conceptual diagrams illustrating the process of adjusting an exercise therapy program according to the present invention, and FIG. 8 is a conceptual diagram illustrating prescription information provided to a medical staff terminal.

[0121] In the present invention, a process of collecting patient information including information on the patient's primary indication may be carried out (S210, see FIG. 2).

[0122] As previously explained, the “indication” according to the present invention refers to a symptom or clinical situation requiring specific treatment or examination, and can be understood as the user’s disease or symptoms. Specifically, the “primary indication” may refer to a disease, symptom, or clinical condition that is initially diagnosed in the patient or is currently the subject of primary treatment.

[0123] The control unit (130) can collect patient information including information on the patient's primary indication from a database (or storage unit (120)). As illustrated in FIG. 3a, the control unit (130) can collect a plurality of patient information corresponding to each of a plurality of patients from at least one of a user terminal (10), a medical staff terminal (30), and a medical staff server (20), and can store the collected plurality of patient information (300) in a database (or storage unit (120)). Here, the patient information (300) may include at least one of medical data (310) and biometric information (320) related to the patient.

[0124] As illustrated in FIG. 3b, medical data (310) according to the present invention may include at least one of the patient's (or user's) name, age, gender, user's physical information (e.g., weight, height, etc.), user's medical history, information on the main indication (330), medical imaging data (340), and prescription information. The medical data (310) according to the present invention is not limited to the examples described above and can be understood to include all forms of quantitative and qualitative data, including medical judgments regarding the patient's physical condition.

[0125] Additionally, the control unit (130) according to the present invention may store biometric information received from at least one sensor provided in the user terminal (10). At this time, the biometric information may be very diverse and may include all physiological and biological signals that can be sensed in relation to the patient's physical condition. For example, the biometric information may include biometric information measurement information related to the user terminal (10) and at least one sensor provided in the user terminal (10). As another example, the biometric information may include item-specific biometric information collected through at least one sensor provided in the user terminal (10).

[0126] As illustrated in FIG. 3c, the control unit (130) can receive biometric information measurement information including at least one of the type of user terminal (10, ex, smart watch, knee band type motion sensor, etc.) for measuring biometric information, the type of at least one sensor provided in the user terminal (10), a wearing area, a sensor item, and a connection method.

[0127] Additionally, the control unit (130) can collect item-specific biometric information (360) through at least one sensor provided in the user terminal (10). As an example, the control unit (130) can collect at least one biometric information (320) among heart rate, blood pressure, heart rate variability (HRV), number of steps (or steps), knee flexion angle, body temperature, respiratory rate, oxygen saturation, electrocardiogram, electromyogram, movement data, sleep pattern, and skin conductivity through at least one sensor provided in the user terminal (10).

[0128] The bio-information according to the present invention is not limited to the examples described above and may include all forms of measurable signals or data capable of representing a patient's physical functions, physiological responses, or behavioral characteristics. Furthermore, the bio-information may be extended to various forms of data collected from newly developed sensors, wearable devices, or medical devices for monitoring the patient's condition.

[0129] The method for collecting patient information (300) including information on the patient's main indication according to the present invention may be very diverse, and the present specification is not limited to any type or method as long as it is a module capable of collecting patient information (300) including information on the patient's main indication.

[0130] For example, the control unit (130) can collect patient information related to the patient from the user terminal (10). As illustrated in FIG. 3d, the control unit (130) can collect patient information related to the patient through a service page (1000, ex, consultation page) provided to the user terminal (10). Specifically, the control unit (130) can collect patient information related to the patient through a service page (1000, ex, consultation page) provided to the user terminal (10) using an artificial intelligence model.

[0131] As an example, the control unit (130) may provide a consultation page for collecting patient information from a patient and, using a large language model-based artificial intelligence model, generate question items (370) for collecting patient information and provide them to the user terminal (10). At this time, the large language model may refer to at least one artificial intelligence model based on at least one of GPT (Generative Pre-trained Transformer), Gemini, T5 (Text-to-Text Transfer Transformer), Vision-Language Models (e.g., CLIP, Flamingo), Multimodal Generative Models (GPT-4 Multimodal, PaLM-E), BERT (Bidirectional Encoder Representations from Transformers), and LaMDA (Language Model for Dialogue Applications).

[0132] At this time, the large language model may be pre-trained and included in the control unit (130) or included in an external artificial intelligence server (140), and the present specification does not specifically limit the method of implementing the operation of the large language model. For example, the large language model may be a lightweight language model executed in a local environment within the control unit (130), or a cloud-based large language model provided in an external artificial intelligence server (140) and communicating with the control unit (130) via a network.

[0133] Specifically, the control unit (130) can generate an answer (390) to a user query (380) entered through a consultation page by using an artificial intelligence model based on a large language model. As an example, the control unit (130) can receive voice received through a microphone provided in the user terminal (10) as a user query (380). At this time, the control unit (130) according to the present invention may further include a module that converts the voice received through the microphone provided in the user terminal (10) into text. For example, the control unit (130) can input voice data received from the user terminal (10) into a voice conversion module. Specifically, the voice conversion module may include at least one voice recognition model based on a STT (Speech-To-Text) algorithm that converts voice data into text. As an example, the speech conversion module may include at least one speech recognition model among a Hidden Markov Model (HMM), a Hidden Markov Model-Gaussian Mixture Model (HMM-GMM), a Connectionist Temporal Classification (CTC)-based model, a Beam Search-based model, a Deep Neural Network (DNN)-based model, a Recurrent Neural Network (RNN)-based model, a Sequence-to-Sequence (Seq2Seq) model, and a Transformer-based model.

[0134] At this time, the control unit (130) can convert voice (or voice data) into text based on a STT (Speech-to-Text) algorithm and process the converted text as a user query (380).

[0135] Furthermore, the control unit (130) can generate an answer (390) to a user query (380) in conjunction with a large language model and provide it to the user terminal (10). Specifically, the control unit (130) can use a prompt generation unit (131) to generate a prompt requesting the generation of an answer (390) for collecting patient information related to the user query (380).

[0136] The control unit (130) can process the generated prompt as input to a large language model to obtain an answer (390) to a user query (380). At this time, the answer (390) to the user query (380) may include a question item requesting the input of patient information related to the user query (380). Through this process, the control unit (130) can collect patient information (300) including information on the patient's primary indication based on the input user query (380).

[0137] In another example, the control unit (130) can collect patient information (300) already stored in a database (or storage unit (120)). Specifically, the control unit (130) can verify the user (or patient) account logged into the user terminal (10). Furthermore, the control unit (130) can collect patient information (300) corresponding to the user account from at least one of the storage unit (120), the database (200), and an external server (e.g., a medical staff server). For example, the control unit (130) can verify user identification information corresponding to the user account in order to collect patient information (300). At this time, the user identification information may include various identification information such as a unique identifier (ID) of the user account, a session token, and a login token, and the control unit (130) can identify (or verify) the user account logged into the user terminal using the user identification information.

[0138] The control unit (130) can retrieve patient information (300) corresponding to the identified user account information from at least one of the storage unit (120), the database (200), and the linked external server based on the identification of the user account. Furthermore, the control unit (130) can collect the retrieved patient information (300) as patient information (300) corresponding to the user account.

[0139] Meanwhile, the control unit (130) can collect bio-information change pattern information in relation to the patient's bio-information (320). Specifically, the control unit (130) can analyze the bio-information (320) collected through at least one sensor provided in the user terminal (10) according to time information and generate bio-information change pattern information that reflects the patient's physical condition change.

[0140] The method for generating bio-information change pattern information by analyzing the bio-information change pattern according to the present invention can be very diverse, and in this specification, any module capable of generating bio-information change pattern information by analyzing the bio-information change pattern is not limited to its type or method. The control unit (130) according to the present invention can generate bio-information change pattern information by analyzing the bio-information change pattern through at least one algorithm.

[0141] For example, the control unit (130) can analyze the change pattern of the bio-information after the occurrence of the main indication. Specifically, the control unit (130) can analyze time-series data of the bio-information collected during a preset period based on the time of occurrence of the main indication to calculate the average value, fluctuation range, upward trend, downward trend, and whether outliers occur in the bio-information. As an example, the control unit (130) can learn the time-series data of the bio-information by applying a machine learning or statistical analysis algorithm and extract the change in the bio-response after the occurrence of the main indication as a feature value. In addition, the control unit (130) can generate bio-information change pattern information by quantifying the change pattern of the bio-information using at least one of time-series analysis techniques based on a moving average, exponential smoothing, Fourier transform, and recurrent neural network (RNN, LSTM, etc.).

[0142] The control unit (130) can store biometric information change pattern information related to a specific patient's biometric information in the storage unit (120, or database (200)). Furthermore, the control unit (130) can retrieve biometric information change pattern information corresponding to user (or patient) identification information from the storage unit (120, or database (200)) to collect biometric information change pattern information related to a specific patient's biometric information.

[0143] Next, in the present invention, a process of generating a prediction prompt requesting the prediction of the probability of occurrence of a derivative indication associated with a primary indication based on patient information may be carried out (S220, see FIG. 2).

[0144] As illustrated in FIG. 4, the control unit (130) can generate a prediction prompt (400) requesting the prediction of the likelihood of a derivative indication associated with the primary indication based on patient information (300) using a prompt generation unit (131). Specifically, the control unit (130) can generate a prediction prompt (400) requesting the prediction of the likelihood of a derivative indication associated with the primary indication based on at least one of medical data (310), bio-information (320), and bio-information change pattern information.

[0145] For example, the prompt generation unit (131) may generate a prediction prompt (400) that requests the calculation of a probability value for occurrence of a derivative indication associated with a primary indication based on at least one of medical data (310), bio-information (320), and bio-information change pattern information. Specifically, the control unit (130) may verify information regarding the patient's primary indication from patient information (300) and retrieve information on at least one derivative indication that is pre-matched to the primary indication from the database (200).

[0146] Furthermore, the control unit (130) can generate a prediction prompt (400) that includes information on derivative indications related to the primary indication of the searched patient by using the prompt generation unit (131). As an example, the control unit (130) can generate a prediction prompt (400) that requests the calculation of a probability value for occurrence of each derivative indication related to the primary indication and a risk grade corresponding to the probability value for occurrence, based on the derivative indication information.

[0147] Next, in the present invention, a process may be carried out to obtain information on the probability of occurrence of at least one derivative indication corresponding to the main indication by processing the generated prediction prompt as input to a pre-trained artificial intelligence model (S230, see FIG. 2).

[0148] As illustrated in FIG. 5, the control unit (130) can generate information on the probability of occurrence of a derivative indication (500) corresponding to a prediction prompt (400) using a previously trained artificial intelligence model (132). At this time, the information on the probability of occurrence of a derivative indication (500) may include probability information (520) based on the probability value of occurrence for each derivative indication (510).

[0149] Specifically, the control unit (130) can calculate the probability value of occurrence for each derivative indication (510) through a pre-trained artificial intelligence model (132). At this time, the pre-trained artificial intelligence model (132) according to the present invention may be configured to analyze the correlation between the change pattern of bio-information and the primary indication and derivative indication, and to predict the possibility of occurrence of derivative indications based on the correlation.

[0150] The method of predicting the probability of a derivative indication occurring through a pre-trained artificial intelligence model (132) according to the present invention may be very diverse, and in this specification, the pre-trained artificial intelligence model (132) is not limited to any type or method as long as it is an artificial intelligence model capable of predicting the probability of a derivative indication occurring based on a prediction prompt (400). The control unit (130) according to the present invention may further include at least one of an artificial intelligence model and a module that perform the same function as the pre-trained artificial intelligence model (132). At this time, the pre-trained artificial intelligence model (132) may be configured in very diverse ways.

[0151] For example, the pre-trained artificial intelligence model (132) according to the present invention may refer to a deep learning-based neural network model trained to predict the probability of occurrence of a derivative indication using patient information, biometric information, and medical data as inputs. In this case, the pre-trained artificial intelligence model (132) may refer to a deep learning model (132a) configured to include at least one of a Transformer-based language model for text analysis, a Recurrent Neural Network (RNN) for time series pattern analysis, a Long Short-Term Memory Network (LSTM), a Gated Recurrent Unit (GRU), a Convolutional Neural Network (CNN) for image data processing, and a Vision Transformer (ViT).

[0152] Specifically, the previously trained artificial intelligence model (132) may refer to an artificial intelligence model that learns the correlation between each input in the hidden layer by using a training data set that includes at least one of a plurality of patient information, a plurality of biometric information, and a plurality of medical data, and receives the feature values ​​of the patient's patient information, biometric information, and medical data together in the input layer.

[0153] In this process, the artificial intelligence model (132) repeatedly learns the relationship between the pattern of change in biological information after the occurrence of the primary indication and whether a derivative indication occurs, and the loss function can be defined to minimize the difference between the predicted probability and the actual occurrence. The learning of the artificial intelligence model can be performed using a supervised learning method, and the weights of the model can be optimized to estimate the probability of occurrence for each derivative indication by using actual clinical results (derivative indication diagnosis history) within a pre-set period as correct labels. Furthermore, the previously trained artificial intelligence model (132) is evaluated for prediction accuracy, sensitivity, and specificity through a validation dataset, and its performance can be corrected through cross-validation or transfer learning.

[0154] As another example, the pre-trained artificial intelligence model (132) according to the present invention includes a Large Language Model (LLM, 132b) structure for processing prediction prompts expressed in natural language, and the LLM (or Large Language Model, Large Language Model, 132b) may mean a model trained with a Transformer-based encoder-decoder structure.

[0155] In this case, the previously trained artificial intelligence model (132) can be trained to receive a prediction prompt expressed in natural language as input, encode the semantic relationships of patient information, biometric information, and medical data included in the prediction prompt, and generate information on the likelihood of occurrence of a derivative indication as output.

[0156] Specifically, during the training process, the AI ​​model can learn the correlation between the change patterns of biometric information, the primary indication, and the said derivative indication by utilizing a training dataset that includes the occurrence status of derivative indications corresponding to multiple patient information. Additionally, the AI ​​model can optimize parameters to convert the contextual meaning within the prediction prompt into a vector form using a Transformer-based encoder-decoder structure, and to calculate the prediction probability for each derivative indication based on the converted vector.

[0157] As previously explained, if the previously trained artificial intelligence model (132) includes a Large Language Model (LLM) structure for processing prediction prompts expressed in natural language, the Large Language Model may be included in an artificial intelligence server (140, ex, LLM server). In this case, the control unit (130) can calculate the probability of occurrence for each derivative indication associated with the main indication through the previously trained artificial intelligence model (132) in conjunction with the artificial intelligence server (140).

[0158] Specifically, the control unit (130) provides a prediction prompt (400) generated through the prompt generation unit (131) as input to a pre-trained artificial intelligence model (132), and the pre-trained artificial intelligence model (132) can interpret the context of the input prediction prompt (400) to generate information on the probability of occurrence of derivative indications, including a probability value, risk grade, and supporting sentences for each derivative indication associated with the main indication. Furthermore, the control unit (130) can convert the information on the probability of occurrence of derivative indications output from the pre-trained artificial intelligence model (132) into a structured data form and store it in a storage unit (120) or a database (200).

[0159] As another example, the pre-trained artificial intelligence model (132) according to the present invention may be configured as a multimodal hybrid neural network structure that processes text information of medical data, time series data of biometric information, and medical image data together.

[0160] In this case, the pre-trained artificial intelligence model (132) may be configured to receive text information of medical data, time-series data of biometric information, and medical image data together to learn the interrelationships. Specifically, during the learning process, the artificial intelligence model (132) may use a medical data set of each of a plurality of patients as training data to extract at least one semantic expression among disease name, symptoms, and prescription history from the text information of the medical data, analyze the change pattern of biometric information from the time-series data of biometric information, and produce an image feature map from the medical image data.

[0161] Specifically, the previously trained artificial intelligence model (132) can encode feature vectors for each of the semantic expression, the change pattern of bio-information, and the image feature map, and generate an integrated expression through a multimodal fusion layer to optimize weights to predict the correlation between the primary indication and the derivative indication. At this time, the training of the artificial intelligence model can be performed in a supervised learning manner, and the model can be trained to minimize loss functions (e.g., L2 loss (Mean Squared Error, MSE), Cross-Entropy Loss, Focal Loss, Cosine Similarity Loss) using label data containing whether the derivative indication occurred as ground truth data.

[0162] The control unit (130) provides at least one of the patient's medical data, biometric information time series data, and medical image data as input to a pre-trained artificial intelligence model (132), and the pre-trained artificial intelligence model (132) can calculate the probability of occurrence and risk grade for each derived indication based on composite features extracted from the input data. At this time, "complex features" may refer to an integrated expression that reflects semantic correlation by combining feature values ​​extracted from the text information of medical data, the time series data of biometric information, and the medical image data, respectively.

[0163] As another example, the pre-trained artificial intelligence model (132) according to the present invention may refer to a reinforcement learning-based model (132c) that learns to continuously improve the performance of predicting derived indications by using feedback on the patient's treatment response or prediction accuracy as a reward signal. In this case, the pre-trained artificial intelligence model (132) may be composed of a reinforcement learning-based artificial intelligence model that continuously improves the performance of predicting derived indications by using feedback on the patient's treatment response and prediction accuracy as a reward signal.

[0164] In this case, during the learning process, the artificial intelligence model may define an environment, configure a state with the patient's biometric information, medical data, and the progression status of the primary indication, and set an action as an adjustment behavior regarding the predicted probability of the occurrence of a derived indication. Here, the "environment" may refer to a space composed of patient information including at least one of the patient's biometric information and medical data.

[0165] Furthermore, the pre-trained artificial intelligence model (132) recognizes a state from the environment and performs an action, and performs the action based on a policy pre-trained through supervised learning, and can update the policy by receiving feedback on the patient's treatment response or prediction accuracy as a reward signal. At this time, it can be understood that information on the patient's treatment response and prediction accuracy can be obtained from medical data created through the medical staff terminal (30) and biometric information continuously collected from the user terminal (10).

[0166] As another example, the artificial intelligence model (132) according to the present invention is composed of a hybrid structure combining a large-scale language model (LLM) and a deep learning model for time series prediction, wherein the LLM (or large-scale language model, large language model) interprets medical records and prediction prompts, and the deep learning model analyzes the change patterns of biometric information, and may mean an artificial intelligence model trained to predict the probability of occurrence of a derivative indication based on the combined result of the LLM and the deep learning model.

[0167] At this time, the LLM (or, large language model, large language model, 132b) interprets medical data and prediction prompts, and the deep learning model (132a) analyzes the change pattern of bio-information, and can predict the likelihood of a derivative indication occurring based on the combined result of the LLM (132b) and the deep learning model (132b).

[0168] Specifically, the control unit (130) may collect at least one of the patient's medical data and biometric information time series data and provide it as input to an artificial intelligence model corresponding to each data type. At this time, the large-scale language model (132b) may receive a prediction prompt and analyze the semantic relationship of the natural language included in at least one of the medical data and biometric information time series data.

[0169] In addition, the deep learning model (132a) can receive time-series data of biological information (e.g., HRV, heart rate, body temperature, SpO₂, etc.), encode the pattern of change of biological information, and learn the correlation of the occurrence of derivative indications according to the progression state of the main indication.

[0170] During the learning process, the control unit (130) may fuse the semantic vector extracted from the LLM (132b) with the time series feature vector calculated from the deep learning model (132a) to form a multimodal integrated representation, and train the model to calculate the probability of the occurrence of a derived indication based on the multimodal integrated representation. At this time, cross-entropy loss, L2 loss, focal loss, etc. may be applied as loss functions, and class-specific weights may be adjusted to correct for imbalances in medical data.

[0171] That is, the pre-trained artificial intelligence model (132) according to the present invention can be implemented in a wide variety of configurations, and the present invention should be understood to include all artificial intelligence models that can be configured by applying various neural network structures or algorithms according to the form of input data, the purpose of prediction, or the learning method, rather than limiting the pre-trained artificial intelligence model (132) to the examples described above. Furthermore, the pre-trained artificial intelligence model (132) can be modified or expanded in various ways within the scope of the technical concept of the present invention.

[0172] Specifically, the control unit (130) can generate information on the probability of occurrence of a derivative indication (500) corresponding to a prediction prompt using a previously trained artificial intelligence model (132). At this time, the information on the probability of occurrence of a derivative indication (500) may include probability information (520) that includes probability values ​​for each of at least one derivative indication (510) related to the main indication.

[0173] The pre-trained artificial intelligence model (132) according to the present invention can be configured to predict the possibility of a derivative indication by reflecting the variation characteristics of bio-information according to the progression state or treatment response of the primary indication and analyzing the correlation between the variation characteristics and the derivative indication. Specifically, the control unit (130) can predict the possibility of a derivative indication by analyzing the change pattern of bio-information after the primary indication has occurred, and processing the change pattern of bio-information and medical data related to the primary indication as input to the pre-trained artificial intelligence model (132).

[0174] More specifically, the control unit (130) can generate information on the possibility of occurrence of a derivative indication by processing a prediction prompt (400) generated based on at least one of bio-information change pattern information and medical data related to the main indication as input to a pre-trained artificial intelligence model (132).

[0175] Next, in the present invention, a process of generating prescription information for a patient's primary indication and derivative indication based on information on the possibility of occurrence of derivative indications may be carried out (S240, see FIG. 2).

[0176] As illustrated in FIG. 6a, the control unit (130) can generate a prescription information generation prompt (600) requesting the generation of prescription information based on the probability information (500) of the occurrence of a derivative indication using the prompt generation unit (131). Specifically, the prompt generation unit (131) can generate a prescription information generation prompt (600) requesting the generation of prescription information based on the probability information (520) included in the probability information (500) of the occurrence of a derivative indication.

[0177] For example, the prompt generation unit (131) can generate a prescription information generation prompt (600) that requests the generation of prescription information based on probability information (520), so that the intensity or content of the prescription for the derived indication in the prescription information changes. Specifically, the prompt generation unit (131) can generate a prescription information generation prompt (600) that requests the generation of prescription information corresponding to a preset probability interval by analyzing probability information (520) that includes probability values ​​for occurrence of each derived indication.

[0178] At this time, the pre-set probability interval according to the present invention may be defined as a plurality of probability intervals for distinguishing the probability of occurrence of a derivative indication. Additionally, a prescription information generation rule is matched to the pre-set probability interval, and the prescription information generation rule matched to each of the plurality of probability intervals may be pre-defined to generate prescription information such that the intensity or content of the prescription varies for one of the types of medication, dosage, treatment cycle, monitoring cycle, and adjustment of the exercise therapy program.

[0179] For example, a plurality of probability intervals may be pre-set, such as a first probability interval (e.g., 0.7 or more to 1.0 or less), a second probability interval (e.g., 0.3 or more to less than 0.7), and a third probability interval (e.g., 0.0 or more to less than 0.3). The plurality of probability intervals according to the present invention are not limited to the examples described above and may be changed according to at least one of information related to the primary indication and derivative indication and patient information.

[0180] As an example, the prompt generation unit (131) can generate a prescription information generation prompt (600) that requests the generation of prescription information such that the strength or content of the prescription for the first derivative indication and the prescription for the second derivative indication are different when the probability of occurrence of the first derivative indication related to the main indication falls within the first probability interval and the probability of occurrence of the second derivative indication related to the main indication falls within the second probability interval.

[0181] Specifically, the prompt generation unit (131) can generate a prescription information generation prompt (600) that requests the generation of prescription information for a first derivative indication, according to a first prescription information generation rule matched to a first probability interval, such that the intensity or content of the prescription for one of the types of medication, dosage, treatment cycle, monitoring cycle, and adjustment of the exercise therapy program is changed.

[0182] Alternatively, the prompt generation unit (131) can generate a prescription information generation prompt (600) that requests the generation of prescription information for a second derivative indication, according to a second prescription information generation rule matched to a second probability interval, such that the intensity or content of the prescription for one of the types of medication, dosage, treatment cycle, monitoring cycle, and adjustment of the exercise therapy program is different.

[0183] As illustrated in FIG. 6b, the control unit (130) processes the generated prescription information generation prompt (600) as input to a pre-trained artificial intelligence model (132) to obtain prescription information (610) regarding the patient's primary indication and the derivative indication through the pre-trained artificial intelligence model (132). Specifically, the control unit (130) processes the prescription information generation prompt (600) as input to the artificial intelligence model (132) to obtain different prescription information (610) depending on the probability of occurrence of the derivative indication.

[0184] As previously explained, the artificial intelligence model (132) that has been trained includes a Large Language Model (LLM) structure for processing prediction prompts expressed in natural language, and the LLM (or Large Language Model, Large Language Model, 132b) may mean a model trained with a Transformer-based encoder-decoder structure.

[0185] In this case, the previously trained artificial intelligence model (132) can be trained to receive a prescription information generation prompt expressed in natural language as input, encode the semantic relationship between the patient condition included in the prompt, the probability value of the occurrence of the derivative indication, and the prescription information generation rule matched to the pre-set probability interval, and based on this, generate prescription information as output related to one of the types of medication, dosage, treatment cycle, monitoring cycle, and adjustment of the exercise therapy program for the main indication and the derivative indication.

[0186] At this time, the training data used may consist of at least one of a patient condition, a probability value of the occurrence of a derivative indication, and a prescription information generation rule matched to a pre-set probability interval, and the prescription information of a medical professional corresponding to the training data may be used as correct answer data so that the artificial intelligence model (132) is configured to learn the relationship between the probability information including the probability value of the occurrence of a derivative indication and the actual prescription information.

[0187] Specifically, the artificial intelligence model (132) can be trained to minimize prediction error by using as correct information information related to at least one of the prescription information generated according to a prescription information generation rule matched to a preset probability interval corresponding to the probability value of the occurrence of a derived indication, and the type of medication prescribed by the medical staff, the dosage, the treatment cycle, the monitoring cycle, and the adjustment of the exercise therapy program.

[0188] At this time, the objective function of learning may include L2 loss, cross-entropy loss, or KL divergence loss, and the loss value may be calculated based on how closely the content of the prescription information generated by the artificial intelligence model (132) matches the prescription information actually selected by the medical staff. Furthermore, the artificial intelligence model (132) may be trained to minimize the loss by repeatedly updating the weights and biases of the network using a back-propagation algorithm based on the difference between the predicted prescription information and the prescription information actually selected by the medical staff.

[0189] At this time, it can be understood that the method of training the artificial intelligence model (132) that generates prescription information based on the large-scale language model according to the present invention is not limited to the example described above, and it is sufficient if it is trained to generate prescription information for the main indication and derivative indication from a prescription information generation prompt expressed in natural language.

[0190] As previously explained, if the previously trained artificial intelligence model (132) includes a Large Language Model (LLM) structure for processing prescription information generation prompts expressed in natural language, the Large Language Model may be included in an artificial intelligence server (140, ex, LLM server). In this case, the control unit (130) may generate prescription information for primary indications and derivative indications through the previously trained artificial intelligence model (132) in conjunction with the artificial intelligence server (140).

[0191] Specifically, the control unit (130) provides the prescription information generation prompt (400) generated through the prompt generation unit (131) as input to a pre-trained artificial intelligence model (132), and the pre-trained artificial intelligence model (132) can generate prescription information for the main indication and derivative indication by interpreting the context of the input prescription information generation prompt (600). Furthermore, the control unit (130) can convert the prescription information output from the pre-trained artificial intelligence model (132) into a structured data form and store it in a storage unit (120) or a database (200).

[0192] The control unit (130) can update existing prescription information to correspond to the prescription information generation prompt (600) by using a previously trained artificial intelligence model (132). Specifically, the control unit (130) can update existing prescription information for the main indication according to the possibility of the occurrence of a derivative indication.

[0193] For example, the control unit (130) may identify a specific derivative indication among a plurality of derivative indications associated with the main indication, wherein the predicted probability is greater than or equal to a preset threshold value, and generate a prescription information generation prompt (600) requesting an update of existing prescription information for the main indication based on the identified specific derivative indication. At this time, the update of existing prescription information may include at least one of the type of medication, dosage, treatment cycle, monitoring cycle, and adjustment of the exercise therapy program.

[0194] Specifically, the pre-trained artificial intelligence model (132) can update existing prescription information by analyzing together the probability information, prescription information, and prescription information generation rules included in the input prescription information generation prompt (600). For example, the pre-trained artificial intelligence model (132) can recognize the relationship between the main indication and each derivative indication in the prescription information generation prompt (600) and extract prescription information generation rules corresponding to the probability interval to which the occurrence probability value of each derivative indication belongs.

[0195] Furthermore, the pre-trained artificial intelligence model (132) can apply the extracted prescription information generation rules to existing prescription items (drug, dosage, treatment cycle, exercise intensity, monitoring frequency, etc.) to update the intensity or content of the prescription according to probability intervals. Accordingly, the control unit (130) can obtain prescription information (610) including the prescription information (620) updated through the pre-trained artificial intelligence model (132).

[0196] In the present invention, it is described that a pre-trained artificial intelligence model (132) generates prescription information based on prescription information generation rules included in a prescription information generation prompt, but even if prescription information generation rules do not exist in the storage unit (120), the pre-trained artificial intelligence model (132) can infer prescription information by interpreting the semantic relationships between input patient information, the probability of occurrence of derivative indications, and medical data through learned clinical knowledge.

[0197] Meanwhile, the control unit (130) can generate prescription information that adjusts the exercise therapy program by updating the existing prescription information for the main indication according to the possibility of the occurrence of a derivative indication in conjunction with a previously learned artificial intelligence model (132).

[0198] As illustrated in FIG. 7a, a pre-trained artificial intelligence model (132) can change at least one of the type (720), intensity (730), frequency (740), and duration (750) of the exercise therapy program (700) to adjust the exercise therapy program (700). Specifically, the pre-trained artificial intelligence model (132) can extract information about the exercise therapy program from a storage unit (120 or database (200)). At this time, the information about the exercise therapy program according to the present invention may include information about exercise items corresponding to each of the types of a plurality of derivative indications.

[0199] For example, the pre-trained artificial intelligence model (132) can determine the exercise type, intensity, frequency, and performance area of ​​the exercise item corresponding to the type of derivative indication based on information regarding the exercise item corresponding to each of the types of the plurality of derivative indications, and based on the probability value of occurrence of the derivative indication associated with the main indication. Specifically, the pre-trained artificial intelligence model (132) can reflect a prescription information generation rule corresponding to the probability interval to which the probability value of occurrence of the derivative indication belongs, based on a prescription information generation prompt, into the exercise therapy program adjustment.

[0200] As another example, the control unit (130) can perform adjustments to the exercise therapy program in conjunction with a pre-trained artificial intelligence model (132) to add exercise items to prevent or alleviate predicted derivative indications. Specifically, the pre-trained artificial intelligence model (132) can update the exercise item information (760) of the exercise therapy program according to prescription information generation rules included in the prescription information generation prompt.

[0201] As illustrated in FIG. 7b, a pre-trained artificial intelligence model (132) can update the exercise type (780), intensity (730), frequency (740), and performance area (790) of each exercise item included in the exercise therapy program (700) by referring to the exercise item adjustment rule included in the prescription information generation rule.

[0202] At this time, the exercise item adjustment rule may refer to a rule established to update the exercise type (780), intensity (730), frequency (740), and performance area (790) of an exercise item to prevent or alleviate a derivative indication. For example, the exercise item adjustment rule may include at least one of a load adjustment rule that adjusts the load on the body part associated with the primary indication according to the probability interval to which the occurrence probability value belongs, a performance area adjustment rule that changes the performance area according to the symptoms or pain area, a frequency adjustment rule that gradually changes the exercise frequency according to the rehabilitation stage (or degree of recovery), a bio-information linkage rule that adjusts the intensity according to bio-information (HRV, heart rate, electromyography, etc.), and an adaptive adjustment rule that adjusts the difficulty of the exercise movement based on patient feedback (e.g., pain score, fatigue level, etc.). These exercise item adjustment rules may exist in a database (200) based on clinical data and empirical evidence from a group of experts, such as medical staff, physical therapists, and clinical data analysts.

[0203] As an example, a previously trained artificial intelligence model (132) can refer to the exercise item adjustment rule included in the prescription information generation prompt to add a first exercise item (e.g., seated leg extension exercise) to the exercise therapy program to prevent or alleviate a first derivative indication (e.g., joint effusion) related to the main indication.

[0204] Next, in the present invention, a process of transmitting prescription information to at least one of a pre-configured server and a terminal may be carried out (S250, see FIG. 2).

[0205] Referring again to FIG. 6b, the pre-configured server and terminal that transmit prescription information in the present invention can be very diverse. For example, the control unit (130) can transmit the generated prescription information to at least one of a central server of a medical institution (or a medical staff server (20)), a cloud-based healthcare server, or a medical staff terminal (e.g., a tablet for doctors, a PC for medical treatment, 30).

[0206] In addition, the control unit (130) can transmit generated prescription information to the patient's user terminal (10, ex, smartphone, smartwatch, tablet, or IoT hub linked with a home rehabilitation device) so that the patient can check and perform their treatment plan and exercise therapy program.

[0207] Furthermore, the control unit (130) may apply a medical data encryption module or a secure communication protocol (TLS / SSL, etc.) during the transmission process to ensure the security and reliability of the prescription information, and may also provide an interactive feedback interface to the medical staff terminal (300) that allows the medical staff to modify or approve the prescription.

[0208] At least one of the configured server and terminal to which prescription information according to the present invention is transmitted is not limited to a single device and can be understood to include all electronic media capable of exchanging medical data, such as a terminal for medical staff, a terminal for patients, a cloud server, and a hospital information system (HIS).

[0209] As illustrated in FIG. 8, the control unit (130) can provide prescription information generated through a previously learned artificial intelligence (132) to a medical staff terminal (30). Specifically, the control unit (130) can provide a prescription information management page (800) containing prescription information to the medical staff terminal (30).

[0210] At this time, the control unit (130) can provide recommended prescription information (810) generated based on the generated prescription information through the prescription information management page (800). Specifically, the control unit (130) can provide recommended prescription information (810) so that medical staff can compare existing prescription information with prescription information generated through a pre-trained artificial intelligence model (132).

[0211] For example, the control unit (130) can visualize and display existing prescription information and prescription information proposed by the artificial intelligence model (132) in a parallel comparison form on the prescription information management page (800), and can be configured so that differences are clearly identified for each item (e.g., type of medication, dosage, treatment cycle, exercise program, etc.).

[0212] At this time, while reviewing the displayed recommended prescription information (810), the medical staff may approve at least some of the multiple prescription items corresponding to the generated prescription information using a pre-trained artificial intelligence model (132). To this end, the control unit (130) may receive a prescription item approval event based on user input applied to the medical staff terminal (30). At this time, user input may be performed in various ways, and the present invention does not limit the method of user input. For example, user input according to the present invention may be performed in at least one of a tap, double tap, long press, click, swipe, drag, pinch in, pinch out, and rotate.

[0213] Furthermore, user input may be input through the voice of a medical professional. In this case, the derived indication response system (100) may further include a module that converts voice into text. For example, the derived indication response system (100) may include a voice recognition model capable of analyzing voice data corresponding to the voice of a medical professional received through a microphone equipped in a medical professional terminal (30). For example, the voice recognition model may analyze the voice of a medical professional based on at least one of a STT (Speech-to-Text) algorithm, HMM (Hidden Markov Model), HMM-GMM (Hidden Markov Model-Gaussian Mixture Model), CTC (Connectionist Temporal Classification) based model, Beam Search based model, DNN (Deep Neural Network) based model, RNN (Recurrent Neural Network) based model, Seq2Seq (Sequence-to-Sequence) model, and Transformer based model, and determine whether to approve at least some of the multiple prescription items.

[0214] The control unit (130) can reflect the approval result for the prescription item entered into the medical staff terminal in the final prescription information and record the approved final prescription information in the storage unit (120).

[0215] Furthermore, the control unit (130) can provide a treatment schedule, medication plan, and exercise therapy program for the main indication and derivative indication according to the final prescription information to the user terminal (10). For example, the control unit (130) can display corresponding exercise videos and exercise description information on a service page of the user terminal (10) for each of the multiple exercise items constituting the medication time, medication name, medication dosage, and exercise therapy program included in the approved final prescription information. Furthermore, the control unit (130) can guide the user to accurately recognize prescription information related to the main indication and derivative indication through at least one of a text notification and a voice guidance function regarding the prescription information update content.

[0216] As described above, the method and system for responding to derivative indications linked to a primary indication using artificial intelligence according to the present invention analyzes patient information based on artificial intelligence, enabling personalized prescription and management that considers the individual patient's condition. Through this, it is possible to predict in advance the likelihood of derivative indications linked to the patient's primary indication to perform preventive measures and manage the patient's health status more precisely.

[0217] Furthermore, the method and system for responding to derivative indications linked to a primary indication using artificial intelligence according to the present invention can probabilistically calculate the probability of occurrence of derivative indications correlated with the primary indication by integrating and analyzing the patient's medical data, biometric information, prescription history, etc. Through this, changes in the patient's condition can be quantitatively evaluated, and a treatment plan for additionally occurring indications can be established based on the predicted results, thereby enabling the preemptive prevention of the progression of indications.

[0218] Furthermore, the method and system for responding to derivative indications linked to a primary indication using artificial intelligence according to the present invention can adjust prescription information, such as the type of medication, dosage, treatment cycle, and monitoring cycle, based on the probability of occurrence of derivative indications predicted by the artificial intelligence model. Through this, medical staff can receive support for optimized prescription decisions tailored to individual patient prediction results and implement a prevention-oriented, personalized treatment system.

[0219] Furthermore, the method and system for responding to derivative indications linked to the primary indication using artificial intelligence according to the present invention can prevent or alleviate derivative indications by providing a patient-customized exercise therapy program in response to predicted derivative indications, thereby simultaneously improving the patient's recovery efficiency and treatment continuity.

[0220] The derivative indication response system (100) linked to the primary indication using artificial intelligence according to the present invention can be implemented through the computing device described below and can perform data processing related to the derivative indication response method linked to the primary indication using artificial intelligence described above.

[0221] Meanwhile, FIG. 9 illustrates an example of a block diagram of a computing system in which the present invention can be implemented.

[0222] Referring to FIG. 9, a computing system (10000) that performs a method for responding to a derivative indication linked to a primary indication using artificial intelligence according to one embodiment of the present invention may include at least one computing device (or computing system). At this time, the at least one computing device may be a single processor or a multi-processor computing device.

[0223] The components of at least one computing device of the present invention may include various hardware components such as one or more processors, memory, other hardware, and a system bus (not shown) that connects various system components so that they can transmit and receive data to and from each other (e.g., telecommutatively connected, physically connected, electrically connected), and the components of at least one computing device are not limited thereto and may be very diverse.

[0224] Meanwhile, at least one computing device included in a computing system (10000) that performs a method for responding to a derivative indication linked to a primary indication using artificial intelligence may be connected to communicate via a network (1070). For example, at least one computing device included in the computing system (10000) may be clustered or part of a local area network (LAN). Additionally, at least one computing device may be part of a wide area network (WAN) or connected to at least one of a client-server network and a peer-to-peer network within the cloud.

[0225] Meanwhile, when at least one computing device is used in at least one of a network environment and a cloud computing environment, the at least one computing device may be connected to at least one of a public and private network through a network interface or adapter. In one embodiment, other communication connection devices, such as a modem, may be used to establish communication through the network. The modem may be at least one of an internal modem and an external modem, and may be connected to a system bus through a network interface or a specific mechanism, etc. A wireless network component consisting of an interface and an antenna may be coupled to the network through a device such as an access point, a peer computer, etc. In the present invention, the method of connecting at least one computing device to communicate through the network (1070) is not limited, and it may be connected to communicate in a manner different from the described example.

[0226] Furthermore, other computer-type devices and / or systems not shown in FIG. 9 may also interact technically with at least one computing device or other system through one or more connections to the network (1070) via a network interface. Here, the network interface may include network interface equipment such as a physical network interface controller (NIC) or a virtual network interface (VIF).

[0227] The network (1070) of the present invention may include various forms such as the Internet, WLAN (Wireless LAN), Wi-Fi (Wireless-Fidelity), Wi-Fi (Wireless Fidelity) Direct, DLNA (Digital Living Network Alliance), WiBro (Wireless Broadband), WiMAX (World Interoperability for Microwave Access), HSDPA (High Speed ​​Downlink Packet Access), HSUPA (High Speed ​​Uplink Packet Access), LTE (Long Term Evolution), LTE-A (Long Term Evolution-Advanced), 5G (5th Generation Mobile Telecommunication), Bluetooth (Bluetooth™ Frequency Identification), Infrared Communication (Infrared Data Association; IrDA), UWB (Ultra-Wideband), ZigBee, NFC (Near Field Communication), Wi-Fi Direct, Wireless USB (Wireless Universal Serial Bus), etc., and in the present invention, data transmission may be performed based on standard communication protocols such as TCP / IP, HTTP, SSL, etc.

[0228] A computing system (10000) that performs a method for responding to a derivative indication linked to a primary indication using artificial intelligence according to the present invention may include at least one of a user computing device (1010, or user computing system), a training computing system (1050, or training computing device), and a server computing system (1030, or server computing device).

[0229] A user computing device (1010) according to the present invention may be understood as a computing device comprising at least one processor (1011) and a memory (1012) that perform a method for responding to a derivative indication linked to a primary indication using artificial intelligence. For example, the user computing device (1010) may include at least one computing device among a smartphone, a smart TV, a laptop computer, a desktop computer, a digital broadcasting terminal, a PDA (personal digital assistants), a PMP (portable multimedia player), a navigation device, a slate PC, a tablet PC, an ultrabook, a wearable device (e.g., a smartwatch, a smart glass, and a head-mounted display).

[0230] At least one processor (1011) constituting the user computing device (1010) may include one or more general-purpose processors and / or one or more special-purpose processors. For example, at least one processor (1011) constituting the user computing device (1010) may be composed of at least one of a central processing unit (CPU), a graphics processing unit (GPU), a tensor processing unit (TPU), a neural network processing unit (NPU), an arithmetic logic unit (ALU), a floating-point arithmetic unit (FPU), an application integrated circuit, an application semiconductor (ASIC), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, and / or electrical units for performing other functions, or a plurality of electrically connected processors.

[0231] Furthermore, at least one processor (1011) may be configured to execute computer-readable instructions contained in memory (1012) and / or other instructions described herein.

[0232] The memory (1012) constituting the user computing system (1010) according to the present invention may include volatile memory, non-volatile memory, fixed media, removable media, magnetic media, optical media, semiconductor media and / or other types of physically durable storage media.

[0233] For example, the memory (1012) may include one or more non-transient / transient computer-readable storage media such as RAM, ROM, HDD (Hard Disk Drive), SSD (Solid State Disk), SSD (Silicon Disk Drive), EEPROM, EPROM, flash memory device, magnetic disk, etc., and may include web storage of a server that performs the storage function of memory on the internet. This memory (1012) may store data and instructions necessary for the at least one processor (1011) to generate prescription information for the patient's primary indication and derivative indication using artificial intelligence.

[0234] A user computing device (1010) may include one or more user input components (1021) that detect user input. For example, the user input component (1021) may also be referred to as a user interface module. The user input component (1021) may include a touch screen, a computer mouse, a keyboard, a keypad, a touchpad, a trackball, a joystick, a voice recognition module, or other similar devices. However, the present invention does not limit the type of user input component (1021). In this case, the user input component (1021) in the present invention does not necessarily mean a hardware means, but can be understood as a channel for receiving input from a user. Meanwhile, the user of the present invention may refer to an automated agent, script, playback software, etc., that operates on behalf of one or more people.

[0235] A user can interact with a computing system (10000) including at least one computing device through input text, touch, voice, movement, computer vision, gestures and / or other forms of input / output using a user input component (1021). For example, the user input component (1021) may include one or more of a command line interface (CLI), a graphical user interface (GUI), a natural user interface (NUI), a voice command interface and / or other user interface (UI) representations.

[0236] Between the user input component (1021) and the user computing device (1010), one or more application programming interface (API) calls may be made based on user input received from the user interface and / or network.

[0237] Here, the expression "based on" may be interpreted to include cases where it is based on the use of a specific configuration, modified from, derived from, influenced by, dependent on, or otherwise derived from a specific configuration. In some embodiments, an API call may be configured for a specific API, which may be interpreted or converted into an API call configured for another API. Here, an API may refer to a defined interface or connection between computers or between computer programs.

[0238] In one embodiment, the user computing device (1010) may store at least one machine learning model (1020). For example, the user computing device (1010) may be various machine learning models, such as a plurality of neural networks (e.g., deep neural networks) that perform a method for responding to a derivative indication linked to a primary indication using artificial intelligence based on user queries and patient information, or other types of machine learning models including non-linear models and / or linear models, and may be composed of a combination thereof.

[0239] According to an embodiment of the present invention, a user computing device (1010) may perform a method for responding to a derivative indication linked to a primary indication using artificial intelligence by using a local or / and external machine learning model (1020). Alternatively, the user computing device (1010) may perform a method for responding to a derivative indication linked to a primary indication using artificial intelligence by using a machine learning model (1040) provided by a server.

[0240] In addition, according to another embodiment of the present invention, a server computing system (1030) communicating with a user computing device (1010) may provide prescription information regarding a primary indication and a derivative indication to the user computing device (1010) via an application or / and the web in accordance with a request from a user received through the user computing device (1010).

[0241] In addition, according to another embodiment of the present invention, at least a part of the user computing device (1010) and the server computing system (1030) are interconnected to perform a method for responding to derivative indications linked to the primary indication using artificial intelligence, thereby providing prescription information for the primary indication and derivative indications to the user.

[0242] Additionally, according to various embodiments of the present invention, a user computing device (1010) and / or a server computing system (1030) can learn machine learning models (1020, 1040) performed in a method for responding to a derivative indication linked to a primary indication using artificial intelligence through interaction with a training computing system (1050) that is communicatedly connected via a network (1070). In this case, the training computing system (1050) may be a computing system separate from the server computing system (1030). Alternatively, in some embodiments, the training computing system (1050) may be part of the server computing system (1030) or part of the user computing device (1010).

[0243] Meanwhile, the server computing system (1030) may include at least one processor (1031) and memory (1032). Here, the processor (1031) may be composed of at least one or a plurality of electrically connected processors among a central processing unit (CPU), a graphics processing unit (GPU), a tensor processing unit (TPU), a neural network processing unit (NPU), an application integrated circuit, an application semiconductor (ASIC), an arithmetic logic unit (ALU), a floating-point arithmetic unit (FPU), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, and / or other electrical units for performing functions. For example, at least one processor (1031) may include a circuit and a transistor configured to execute instructions from memory (1032).

[0244] The memory (1032) constituting the server computing system (1030) according to the present invention may include volatile memory, non-volatile memory, fixed media, removable media, magnetic media, optical media, semiconductor media, and / or other types of physically durable storage media. For example, the memory (1032) may include one or more non-transient / transient computer-readable storage media such as RAM, ROM, HDD (Hard Disk Drive), SSD (Solid State Disk), SSD (Silicon Disk Drive), EEPROM, EPROM, flash memory device, magnetic disk, etc., and combinations thereof, and may include web storage of a server that performs the storage function of memory over the internet. Additionally, the server computing system (1030) may further include a data storage (data store). For example, the data storage may be composed of at least one of a relational database, a NoSQL database, a data warehouse, and a local file system.

[0245] In the memory (1032) constituting the server computing system (1030) according to the present invention, data and instructions necessary for the at least one processor (1031) to perform the operation of an application for responding to a derivative indication linked to a primary indication using artificial intelligence may be stored.

[0246] In one embodiment, the server computing system (1030) may be composed of a single device or a plurality of computing devices, and may be configured to operate according to a sequential or parallel computing architecture. Additionally, a distributed processing system may be configured with a plurality of networked devices.

[0247] Meanwhile, the training computing system (1050) may include at least one processor (1051) and memory (1052). The model trainer (1060) is a logical component that executes the training of at least one machine learning model (1020, 1040) and may be implemented in the form of hardware, firmware, or software. For example, the model trainer (1060) may be executed by the processor (1051) after loading training data (1061) stored in a storage device into memory (1052). For example, the model trainer (1060) may be configured to execute one or more operations (e.g., model training, model reconstruction, model validation, model testing) on ​​at least one machine learning model.

[0248] The machine learning model of the present invention may include at least one of a statistical model, an algorithm, a neural network (NN), a convolutional neural network (CNN), a generative neural network (GNN), a Word2Vec model, a Bag of Words model, a TF-IDF (document frequency-inverse document frequency) model, a GPT (Generative Pre-trained Transformer) model (or other autoregressive models), a PPO (Proximal Policy Optimization) model, a nearest neighbor model (e.g., a k-nearest neighbor model), a linear regression model, a K-means clustering model, a Q-learning model, a TD (Temporal Difference) model, a Deep Adversarial Network model, and all other types of models further described herein.

[0249] Specifically, the model trainer (1060) may execute operations to train a machine learning model, and said operations may include at least one of adding, removing, and modifying model parameters. At this time, the training of the machine learning model may be at least one of supervised learning, semi-supervised learning, and unsupervised learning. In one embodiment, the training of the machine learning model may include the step of repeatedly inputting training data (1061) based on epochs and repeatedly performing the machine learning model training process configured in this way. Here, an epoch may refer to a unit in which the entire set of training data (1061) undergoes forward and backpropagation processing once. In some implementations, different levels of training methods (e.g., supervised learning, semi-supervised learning, unsupervised learning) may be used for different epochs.

[0250] The training data (1061) of the present invention may include input data and / or data previously output from at least one machine learning model (e.g., recursive learning feedback). The parameters of at least one machine learning model may include at least one of a seed value, a model node, a model layer, an algorithm, a function, connections between different machine learning models, connections between parameters, machine learning model constraints, and other digital components that influence the output of the machine learning model. In this case, the model connections between different machine learning models may include or represent relationships between model parameters and / or models, which may be dependent or interdependent, hierarchical, and / or static or dynamic. The combinations and configurations of model parameters described herein may be too complex to be maintained or used by human cognitive abilities.

[0251] In the present invention, the machine learning parameters described according to the embodiments are not limited, and a single machine learning model may further include a plurality of model parameters.

[0252] Meanwhile, FIG. 10 illustrates an example of a block diagram of a computing device (1100) that may be included in a user computing device (1010), a server computing system (1030), and a training computing system (1050), as an embodiment of a computing system (10000) in which the present invention can be implemented.

[0253] As illustrated in FIG. 10, the computing device (1100) may include at least one application (e.g., Application 1 to Application N), and each of the at least one application may include a machine learning library and a model execution environment for performing a method for responding to a derivative indication linked to a primary indication using artificial intelligence. The at least one application included in the computing device (1100) may communicate with the sensor, context manager, device state manager, or additional component(s) within the computing device (1100) via an API (Application Programming Interface). In one embodiment, the at least one application may interface with device components, such as receiving sensor data or state data or transmitting prediction results to an output device via a public or private API.

[0254] Meanwhile, FIG. 11 illustrates an example of a block diagram in another aspect of a computing device (1200), which is one of the components of a computing system (10000) that performs a method for responding to a derivative indication linked to a primary indication using artificial intelligence according to an embodiment of the present invention.

[0255] A computing device (1200) according to the present invention may include at least one application (e.g., Application 1 to Application N), and at least one application may communicate with a central intelligence layer (1210). Each application may interact with a shared model within the central intelligence layer (1210) through an API (e.g., a common API).

[0256] The central intelligence layer (1210) includes one or more machine learning models and may share them among multiple applications or provide them independently to each. In one embodiment, the central intelligence layer (1210) may be integrated as part of an operating system or implemented as a separate logical layer.

[0257] Additionally, the central intelligence layer (1210) can communicate with the central device data layer (1220). The central device data layer (1220) can provide patient information stored within the computing device (1200) as input data required to generate prescription information for primary indications and derivative indications. Each device component (e.g., sensor, state manager, etc.) can communicate with the central device data layer (1220) via a private API, etc.

[0258] The technology described in this specification may be composed of a single or multiple computing devices, and a machine learning model that performs a method for responding to a derivative indication linked to a primary indication using artificial intelligence may be executed sequentially or in parallel on a single component or multiple distributed components. Data storage, machine learning models, and applications may be distributed and operated locally or over a network, and these configurations can be flexibly applied to various system architectures.

[0259] Meanwhile, computer-readable media include all types of recording devices in which data that can be read by a computer system is stored. Examples of computer-readable media include HDD (Hard Disk Drive), SSD (Solid State Disk), SSD (Silicon Disk Drive), ROM, RAM, CD-ROM, magnetic tape, floppy disk, optical data storage device, etc.

[0260] Furthermore, the computer-readable medium may be a server or cloud storage that includes a storage and is accessible to an electronic device via communication. In this case, the computer may download the program according to the present invention from the server or cloud storage via wired or wireless communication.

[0261] Furthermore, in the present invention, the computer described above is an electronic device equipped with a processor, namely a CPU (Central Processing Unit), and no special limitations are placed on its type.

[0262] Meanwhile, the above detailed description should not be interpreted restrictively in all respects but should be considered exemplary. The scope of the invention should be determined by a reasonable interpretation of the appended claims, and all modifications within the equivalent scope of the invention are included within the scope of the invention.

Claims

1. A step of collecting patient information including information on the patient's primary indication; A step of generating a prediction prompt requesting the prediction of the likelihood of occurrence of a derivative indication associated with the primary indication based on the above patient information; A step of processing the generated prediction prompt as input to a pre-trained artificial intelligence model to obtain information on the probability of occurrence of at least one derivative indication corresponding to the primary indication; A step of generating prescription information for the patient's primary indication and the derivative indication based on information on the possibility of occurrence of the derivative indication; and A method for responding to derivative indications linked to a primary indication using artificial intelligence, characterized by including the step of transmitting the above-mentioned prescription information to at least one of a pre-configured server and terminal.

2. In Paragraph 1, The step of obtaining information on the possibility of occurrence of the above-mentioned derivative indication is, Calculate the probability values ​​of occurrence for each of the above-mentioned derivative indications using the above-mentioned previously trained artificial intelligence model, and The information on the possibility of the above-mentioned derivative indications is, A method for responding to derivative indications linked to a primary indication using artificial intelligence, characterized by including probability information based on the probability value of occurrence for each of the derivative indications mentioned above.

3. In Paragraph 2, The step of generating the above prescription information is, Based on the above probability information, generate a prescription information generation prompt requesting the generation of the above prescription information such that the strength or content of the prescription for the above derivative indication changes in the above prescription information, and A method for responding to derivative indications linked to a primary indication using artificial intelligence, characterized by processing the above prescription information generation prompt as input to the above artificial intelligence model to obtain different prescription information according to the probability of occurrence of the above derivative indication.

4. In Paragraph 1, The above patient information is, The medical data of the patient and the biometric information of the patient collected from at least one sensor, In the step of obtaining information on the possibility of occurrence of the above-mentioned derivative indication, After the above primary indication occurs, the pattern of change in the above biological information is analyzed, and By processing the above-mentioned bio-information change patterns and medical data related to the above-mentioned primary indication as input to the above-mentioned artificial intelligence model, the probability of occurrence of the above-mentioned derivative indication is predicted, and A method for responding to derivative indications linked to a primary indication using artificial intelligence, characterized by generating information on the probability of occurrence of the derivative indication based on the predicted probability of occurrence of the derivative indication.

5. In Paragraph 4, The above artificial intelligence model is, Analyze the correlation between the change pattern of the above biological information and the above primary indication and the above derivative indication, and A method for responding to derivative indications linked to a primary indication using artificial intelligence, characterized by being configured to predict the likelihood of occurrence of the derivative indication based on the above correlation.

6. In Paragraph 5, The above artificial intelligence model is, Reflecting the characteristics of fluctuation in the above bio-information according to the progression status or treatment response of the above primary indication, A method for responding to derivative indications linked to a primary indication using artificial intelligence, characterized by being configured to predict the probability of occurrence of the derivative indication by analyzing the correlation between the above-mentioned variation characteristics and the above-mentioned derivative indication.

7. In Paragraph 1, The step of generating the above prescription information is, Update existing prescription information for the primary indication based on the possibility of the occurrence of the above derivative indication, and A method for responding to derivative indications linked to a primary indication using artificial intelligence, characterized in that the above update includes at least one of the adjustment of the type of medication, the dosage of medication, the treatment cycle, the monitoring cycle, and the exercise therapy program.

8. In Paragraph 7, The adjustment of the above exercise therapy program is, A method for responding to derivative indications linked to a primary indication using artificial intelligence, characterized by being configured to change at least one of the type, intensity, frequency, and duration of exercise.

9. In Paragraph 7, In the step of generating the above prescription information Adjustments to the exercise therapy program are made to add the exercise items to prevent or alleviate the predicted derivative indications, and The adjustment of the above exercise therapy program is, A method for responding to derivative indications linked to a primary indication using artificial intelligence, characterized by being configured to determine the exercise type, intensity, frequency, and performance area of ​​the exercise item corresponding to the type of derivative indication.

10. In Paragraph 1, A method for responding to derivative indications linked to a primary indication using artificial intelligence, characterized in that the above artificial intelligence model is a deep learning-based neural network model trained to predict the probability of occurrence of derivative indications using the above patient information, biometric information, and medical data as inputs.

11. In Paragraph 10, The above artificial intelligence model is, A method for responding to a derivative indication linked to a primary indication using artificial intelligence, characterized by being a deep learning model composed of at least one of a recurrent neural network (RNN), a long short-term memory network (LSTM), and a gated recurrent unit (GRU) for learning the time-series changes of the biological information of the patient.

12. In Paragraph 10, The above artificial intelligence model is, It includes a Large Language Model (LLM) structure for processing the above prediction prompt expressed in natural language, and A method for responding to derivative indications linked to a primary indication using artificial intelligence, characterized in that the above LLM is a model trained with a transformer-based encoder-decoder structure.

13. In Paragraph 10, The above artificial intelligence model is, A method for responding to derivative indications linked to a primary indication using artificial intelligence, characterized by being composed of a multimodal hybrid neural network structure that processes text information of the medical data, time-series data of the biometric information, and medical image data together.

14. In Paragraph 10, The above artificial intelligence model is, A method for responding to derivative indications linked to a primary indication using artificial intelligence, characterized by being a reinforcement learning-based model that learns to continuously improve the prediction performance of derivative indications by using feedback on the treatment response or prediction accuracy of the patient as a reward signal.

15. In Paragraph 10, The above artificial intelligence model is, It consists of a hybrid structure combining a large-scale language model (LLM) and a deep learning model for time series forecasting, and The above LLM interprets the above medical data and the above prediction prompt, and A method for responding to a derivative indication linked to a primary indication using artificial intelligence, characterized in that the deep learning model analyzes the change pattern of the biological information and predicts the probability of the derivative indication occurring based on the combined result of the LLM and the deep learning model.

16. In electronic devices, Memory for storing instructions; and It includes at least one processor electrically connected to the memory, and When the above instructions are executed by the at least one processor, the at least one processor, Collect patient information including information on the patient's primary indications, and Based on the above patient information, generate a prediction prompt requesting a prediction of the likelihood of a derivative indication associated with the above primary indication, and The generated prediction prompt is processed as input to a pre-trained artificial intelligence model to obtain information on the probability of occurrence of at least one derivative indication corresponding to the primary indication, and Based on information on the probability of occurrence of the above derivative indication, prescription information for the patient's primary indication and the above derivative indication is generated, and A derivative indication response system linked to a primary indication using artificial intelligence, characterized by transmitting the above-mentioned prescription information to at least one of a pre-configured server and terminal.

17. A program that is executed by one or more processes in an electronic device and stored on a computer-readable recording medium, The above program is, A step of collecting patient information including information on the patient's primary indications; A step of generating a prediction prompt requesting the prediction of the likelihood of occurrence of a derivative indication associated with the primary indication based on the above patient information; A step of processing the generated prediction prompt as input to a pre-trained artificial intelligence model to obtain information on the probability of occurrence of at least one derivative indication corresponding to the primary indication; A step of generating prescription information for the patient's primary indication and the derivative indication based on information on the possibility of occurrence of the derivative indication; and A program stored on a computer-readable recording medium characterized by including instructions that perform the step of transmitting the above-mentioned prescription information to at least one of a pre-configured server and a terminal.

Images