Method and apparatus for recommending drug and controlling dispensing based on customized disease prediction using artificial intelligence and dynamic body map with multi-modal data
An AI-driven dynamic body map system using wearable data and vision AI for real-time disease prediction and drug safety analysis addresses misdiagnosis and misuse by enhancing accuracy and safety in drug dispensing.
Patent Information
- Application Number
- KR1020260052160
- Authority / Receiving Office
- KR · KR
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2026-03-23
- Publication Date
- 2026-11-10
- Estimated Expiration
- 2046-03-23
AI Technical Summary
Existing body map-based systems lack the ability to accurately predict diseases and ensure drug safety due to static structures and lack of real-time analysis of user-specific health data and drug interactions, leading to potential misdiagnosis and misuse.
An AI-driven dynamic body map system that integrates wearable biometric data and vision AI for real-time disease prediction, combined with Drug Utilization Review and inventory management to ensure safe and personalized drug dispensing.
Enhances disease prediction accuracy and prevents drug misuse by providing hyper-personalized precision prescribing and real-time safety controls, ensuring secure and efficient drug dispensing.
Smart Images

Figure 112026035204028-PAT00003_ABST
Abstract
Description
Technology Field
[0001] The present invention relates to a method and apparatus for personalized drug recommendation and dispensing control using artificial intelligence-based dynamic body maps and multimodal data. More specifically, it relates to an intelligent drug dispensing system that analyzes a user's biosignals and affected area vision data to generate a personalized dynamic body map interface, diagnoses drug safety according to individual health contexts, and automatically controls the physical dispensing logic of an integrated smart drug storage cabinet. Background Technology
[0002] Recently, due to the entry into an aging society and the spread of the self-medication trend of managing one's own health, the demand for over-the-counter (OTC) drugs available without a doctor's prescription has surged. Consequently, devices in the form of kiosks and vending machines are being introduced to allow users to obtain information about and purchase medications remotely, even outside of pharmacies. Since professional knowledge is required for the general public to select a product optimized for their symptoms from among numerous medications, a method has been proposed in which a body map mimicking the human body shape is utilized on the user interface to select the pain area and receive information on related diseases.
[0003] However, existing body map-based systems remain in a static structure that merely hierarchically lists pre-set images of body parts and disease lists, which has limitations such as a high possibility of misdiagnosis based on the user's subjective judgment and the inability to consider the specific physical conditions or underlying diseases of individual users. In particular, the method of simply selecting a specific body part and matching it with representative diseases and medications is merely at the level of publicly available technology, such as the cited references, and lacks the technical completeness to perform precise disease prediction by comprehensively analyzing the actual condition of the affected area or changes in bio-data through wearable devices.
[0004] Furthermore, although pharmaceuticals can cause fatal side effects if misused, unlike simple consumer goods, existing systems lack safety mechanisms to control physical dispensing devices by verifying the user's allergy history or drug interactions (DUR) with currently taken medications in real time. Therefore, there is an urgent need to develop an advanced system that combines lesion analysis using vision AI with dynamic body map rendering technology based on multimodal data to dramatically improve the accuracy of disease prediction, and can actively unlock smart storage boxes and recommend alternative medications based on personalized medication safety verification results for each user. The problem to be solved
[0005] The objective of the present invention to solve the aforementioned problems is to provide a method for personalized drug recommendation and dispensing control using artificial intelligence-based dynamic body maps and multimodal data.
[0006] Another objective of the present invention to solve the above-mentioned problems is to provide a customized drug recommendation and dispensing control device using artificial intelligence-based dynamic body maps and multimodal data.
[0007] The problems that this application aims to solve are not limited to those described above, and problems not mentioned will be clearly understood by those skilled in the art from this specification and the attached drawings. means of solving the problem
[0008] A method for a computing device executed by at least one server according to an embodiment of the present invention to achieve the above objective, which recommends a customized medicine based on artificial intelligence (AI) and controls dispensing, may include the steps of: providing an interactive body map interface for receiving a symptom area input to a user terminal; acquiring and analyzing user context and multi-modal data including the user's health records and wearable biometric data; predicting a disease based on the symptom area input through the interactive body map interface and the analyzed data, and recommending a customized medicine corresponding to the predicted disease; and comprehensively determining the real-time inventory status of a smart medicine cabinet linked to the recommended customized medicine and the user's medication safety information to automatically control the physical dispensing of the smart medicine cabinet. The means for solving the problem are not limited to the means described above, and unmentioned means of solution will be clearly understood by those skilled in the art to which this application belongs from this specification and the attached drawings.
[0009] delete
[0010] delete
[0011] delete
[0012] delete
[0013] delete
[0014] delete
[0015] delete
[0016] delete Effects of the invention
[0017] According to the present invention, a dynamic body map interface that combines user's wearable device data and real-time affected area vision analysis results to preemptively visualize areas where abnormal signs are detected on a body map can help users intuitively recognize symptoms and minimize input errors.
[0018] Furthermore, by applying a multidimensional inference algorithm that fuses image analysis using vision AI models with individual health contexts, the accuracy of disease prediction can be dramatically improved compared to existing technologies based on simple menu selection, enabling the provision of expert-level, precise drug recommendation services. In particular, through Edge Federated Learning, predictive power for regionally prevalent diseases can be maximized while maintaining data security.
[0019] Furthermore, by utilizing real-time filtering based on Drug Utilization Review (DUR) and allergy information, along with dynamic rearrangement algorithms based on inventory status, this system physically blocks inappropriate drug dispensing and suggests optimal alternative medications, thereby achieving economic and social benefits by preventing drug misuse and abuse accidents in non-face-to-face environments. In particular, it offers innovative effects such as enabling hyper-personalized precision prescribing through real-time PK / PD simulation, and ensuring safety even after administration through multi-unit asynchronous-synchronous dispensing control and closed-loop tracking management. Brief explanation of the drawing
[0020] FIG. 1 is a block diagram of an electronic device performing a method and apparatus for personalized drug recommendation and dispensing control using an artificial intelligence-based dynamic body map and multimodal data according to an embodiment of the present invention. FIG. 2 is a diagram showing terminals performing a method and apparatus for personalized drug recommendation and dispensing control using an artificial intelligence-based dynamic body map and multimodal data according to an embodiment of the present invention. FIG. 3 is a flowchart of a method and apparatus for personalized drug recommendation and dispensing control using an artificial intelligence-based dynamic body map and multimodal data according to an embodiment of the present invention. FIG. 4 to 8 are diagrams showing screens of an electronic device performing a method for providing drug information suitable for a disease based on a body map according to an embodiment of the present invention. FIG. 9 is a diagram showing major and minor categories that a user can select when selecting a body map-based area according to an embodiment of the present invention. Specific details for implementing the invention
[0021] The present invention is susceptible to various modifications and may have various embodiments; specific embodiments are illustrated in the drawings and described in detail in the detailed description. However, this is not intended to limit the invention to specific embodiments, and it should be understood that the invention includes all modifications, equivalents, and substitutions that fall within the spirit and scope of the invention. Similar reference numerals have been used for similar components in the description of each drawing.
[0022] Terms such as first, second, A, B, etc., may be used to describe various components, but said components shall not be limited by said terms. These terms are used solely for the purpose of distinguishing one component from another. For example, without departing from the scope of the present invention, the first component may be named the second component, and similarly, the second component may be named the first component. The term "and / or" includes a combination of a plurality of related described items or any of a plurality of related described items.
[0023] 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.
[0024] The terms used in this application are used merely to describe specific embodiments and are not intended to limit the invention. The singular expression includes the plural expression unless the context clearly indicates otherwise. In this application, terms such as "comprising" or "having" are intended to specify the presence 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.
[0025] Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as generally understood by those skilled in the art to which the present invention pertains. Terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and should not be interpreted in an ideal or overly formal sense unless explicitly defined in this application.
[0026] Hereinafter, preferred embodiments of the present invention will be described in more detail with reference to the attached drawings. In order to facilitate an overall understanding of the present invention, the same reference numerals are used for identical components in the drawings, and redundant descriptions of identical components are omitted.
[0027] Referring to FIG. 1, an electronic device according to one embodiment of the present invention may be implemented based on a general-purpose computing hardware architecture including a processor, memory, a communication unit, and an input / output interface. The electronic device described herein may be understood as a concept encompassing a server, a cloud computing system, or an edge computing device that runs an artificial intelligence (AI) algorithm and operates as a personalized medicine recommendation and dispensing control device by being linked through a network with a user's terminal and a smart medicine cabinet.
[0028] The processor may include at least one of a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), an Application Processor (AP), a Tensor Processing Unit (TPU), or a Neural Processing Unit (NPU). The processor can proactively perform complex operations proposed in the present invention, such as multimodal data acquisition, dynamic body map generation, disease prediction based on a Vision AI model, deep medical history chatbot inference based on a Decision Tree, context-aware dynamic re-ranking of drug priorities, and physical dispensing control based on multi-unit synchronization of a smart drug storage box, by fetching and decoding one or more instructions stored in memory. In particular, it may be desirable for the processor to be equipped with high-performance computing capabilities that support a distributed processing environment, enabling real-time processing of parallel drug recommendation and dispensing requests from multiple users without delay.
[0029] The above processor may be operated by executing one or more instructions stored in memory, including an interface provider, a data analysis unit, a disease prediction and drug recommendation unit, and a dispensing control unit as logical or functional components. Specifically, the interface provider may provide an interactive body map interface to receive symptom area input from a user terminal. The data analysis unit may acquire and analyze user context and multi-modal data including the user's health records and wearable biometric data. The disease prediction and drug recommendation unit may predict a disease based on the symptom area input through the interactive body map interface and the analyzed data, and may recommend a customized drug corresponding to the predicted disease. The dispensing control unit may automatically control the physical dispensing of the smart drug storage unit by comprehensively determining the real-time inventory status of the linked smart drug storage unit and the user's medication safety information regarding the recommended customized drug.
[0030] The above memory can serve as a data repository that stores data processed by a processor, learning parameters of an artificial intelligence model, and software programs for running algorithms in non-volatile and volatile forms. According to an embodiment of the present invention, the memory can systematically build and store health records including a user's past medical treatment and drug purchase history, time-series biometric data collected from a wearable device, a database of drug ingredient cross-analysis and drug safety information, encrypted template data for biometric (face or fingerprint)-based identity authentication, and a real-time inventory status table of an integrated smart drug storage box.
[0031] The communication unit may be responsible for establishing a communication session so that the electronic device can transmit and receive data to and from external user terminals, wearable devices, and smart medicine cabinets via a wired or wireless communication network. The communication unit applies an end-to-end encryption protocol to ensure integrity and security during the transmission and reception of sensitive medical information and multimodal biometric data of the user, and can transmit control signals for the synchronized physical dispensing of the smart medicine cabinet in an ultra-low latency environment.
[0032] The above input / output interface may provide a physical or logical connection channel to allow a user or administrator to input commands into an electronic device and intuitively check the operating status of the system. The input / output interface may be combined with a touchscreen, keypad, camera module, biometric sensor, etc., to receive signals for selecting a symptom area, actual images of the affected area, identity authentication data, etc., in real time.
[0033] Referring to FIG. 2, a customized drug recommendation and dispensing control system according to one embodiment of the present invention includes a user terminal, a wearable device, a smart drug storage box, a server, and a database, and these organically transmit and receive data through a communication network. The server operates as a core entity that integrally drives an artificial intelligence-based multimodal disease prediction, a context-aware drug dynamic re-ranking algorithm, and a storage box physical dispensing control logic.
[0034] The system of the present invention enhances data security and the predictive power of regional epidemic diseases by introducing Edge Federated Learning technology. Specifically, it utilizes multiple storage boxes and edge servers in regions where smart medicine cabinets are deployed as local learning nodes to self-learn the body map input patterns of multiple users and transmits the calculated local disease weight parameters to a central server. Based on this, the central server updates the global model and redistributes it back to the local nodes. Accordingly, during the interactive body map provision stage, based on the redistributed local epidemic disease weights, when a subsequent user accesses the body map, the heatmap of body parts associated with the epidemic disease in that region (e.g., a respiratory virus outbreak in a specific area) is highlighted as a default, and related in-depth medical history is placed at the highest priority, thereby maximizing the intuitiveness of input and the speed of prediction.
[0035] When a user transmits an image of a specific symptomatic area via a terminal, a vision AI model installed on the server performs an in-depth analysis to generate an expected disease scoring result. The server cross-validates the reliability between multi-modality data. Specifically, the cross-validation layer calculates the feature discrepancy between the image of the affected area analyzed by vision AI, wearable biometric data, and response text data from an in-depth medical history chatbot. If the discrepancy exceeds a threshold and contradictory results are obtained, the server executes a dynamic attention algorithm that applies an attention penalty to modalities with low data reliability to lower their weights. Furthermore, it triggers micro-interactions, such as automatically requesting a re-measurement of rPPG based on blood flow or reconfirming the presence of pain via a haptic sensor, to precisely correct the disease prediction score.
[0036] When recommending personalized medicines in response to predicted diseases, the server performs hyper-personalized precision prescriptions that reflect the individual user's internal metabolic environment. Using skin bioimpedance-based fluid content, heart rate variability (HRV), and body temperature data obtained from wearable biometric data as inputs, the server performs real-time pharmacokinetic / pharmacodynamic (PK / PD) simulations to predict the metabolic rate of specific drug components. Subsequently, based on the simulation results and the user's real-time gastrointestinal fluid status and blood flow information, it dynamically selects and recommends the optimal formulation (e.g., liquid, soft capsule, tablet, etc.) and precise dosage in milligrams (mg) to maximize the absorption rate and onset time of drug action.
[0037] When a recommended drug is physically dispensed, the microcontroller (MCU) of the storage unit retrieves physical characteristic data, including the drug's weight, packaging type, and shock resistance, to individually calculate the dispensing tray descent speed and drop trajectory for each unit. Based on this, the dispensing motor for drugs with high shock resistance is preemptively driven asynchronously, and the dispensing speed of drugs at risk of damage is decelerated to delay the drop point. Consequently, a multi-unit asynchronous-to-synchronous dispensing control logic is applied to synchronize the time it takes for the dispensed drugs to reach the final single simultaneous dispensing zone to the microsecond level.
[0038] Following the dispensing control stage, a closed-loop tracking stage is performed to track and manage the patient's condition after medication. The server transmits symptom relief threshold trajectory data, calculated by combining the average onset time of the dispensed medication with user context, to the user's wearable device. The wearable device performs background monitoring from the time of administration; if the biometric data deviates from the threshold trajectory and deteriorates, or if an acute adverse reaction is detected, it sends a warning notification to the terminal and simultaneously requests a connection with a counselor to a remote tele-health system. It also controls the risk in advance by temporarily hard-locking additional dispensing of medications with the same mechanism of action within the corresponding storage box.
[0039] Referring to FIG. 3, in operation S310, a method according to an embodiment of the present invention may be initiated chronologically from the step of a computing device (server) providing an interactive body map interface to a user's terminal. In this step, the server collects and analyzes biometric data streamed from the user's health records and wearable devices in real time. For example, if a rise above a threshold is detected in the collected wearable body temperature data, the server may determine this as an immediate abnormal sign and preemptively map a visual heat map indicating fever on the head or face area of the body map.
[0040] After the dynamic body map is provided, in the S320 operation, the server enters a stage where it precisely predicts the disease by combining the user's symptom area entered through the interface with previously acquired multimodal data. When the user touches and selects an area highlighted by the heatmap or a specific suspected area, and then captures and uploads an image of the actual affected area using a smartphone camera, the server's vision AI model primarily calculates the predicted disease scoring result. Subsequently, triggered by the calculated result, an in-depth medical history chatbot based on a decision tree algorithm is activated, and the final probability of the predicted disease is corrected by receiving feedback on the response data.
[0041] Once precise disease prediction is complete, a step is performed in the S330 operation to recommend customized medicines corresponding to the predicted disease. The server can perform a re-ranking operation to dynamically increase the priority weights of medicines associated with respiratory diseases by analyzing collected current weather information (fine dust, seasonal changes, etc.). Additionally, the server can perform dynamic re-ranking of customized priorities by further analyzing the user's medicine request time zone context, such as placing medicines without drowsiness-inducing side effects at the top during daytime hours and prioritizing the recommendation of medicines containing sleep-inducing ingredients during nighttime hours. Furthermore, if it is confirmed that the re-ranked first-priority medicine is not in stock in the inventory, the server can call the medicine ingredient database to perform a cross-analysis and automatically switch to recommend a second-priority alternative medicine with chemically / pharmacologically equivalent efficacy to the user.
[0042] Once the drug recommendation is finally confirmed, a step is performed in operation S340 to automatically control the physical dispensing of the recommended drugs from the smart drug cabinet. Prior to generating the dispensing control signal, the server receives facial or fingerprint biometric data to perform identity authentication. The authentication result is encrypted within the server and combined with the physical dispensing control signal packet, thereby fundamentally preventing the unauthorized dispensing of age-restricted drugs. When the authenticated control signal is transmitted to the cabinet, the cabinet's microcontroller (MCU) activates the previously described multi-cabinet collaborative dispensing mechanism to accurately transport the corresponding drugs to a single simultaneous dispensing zone.
[0043] Alternatively, for example, one embodiment may provide body map-based drug information through the screen of an electronic device. For example, the electronic device may represent the computing device or server described above. For example, at least some of the contents described below and at least some of the contents described above may be organically combined and operated.
[0044] FIGS. 4 to 8 are drawings showing the screen of an electronic device in which a method for providing drug information suitable for a disease based on a body map according to an embodiment of the present invention is performed.
[0045] The electronic device can determine the disease area based on a body map comprising a full-body image of the human body and multiple partial images of the human body. Here, the body map is composed of a full-body image of the human body and multiple partial images of the human body, and can assist the user in easily visually identifying and selecting the area by representing major and minor classification areas as images. The major and minor classification areas will be described in detail later with reference to FIG. 8.
[0046] To explain in more detail, the electronic device may provide a full-body image of a human body including a plurality of preset major categories, and when a selection signal for a first human body position is obtained in the full-body image of the human body, a specific major category corresponding to the first human body position among the plurality of major categories may be determined. Here, referring to FIG. 4, the full-body image of the human body can be viewed, and an area may be provided at the bottom of the screen that allows the full body of the human body to be rotated at various angles according to the user's selection signal in order to display both the front and back sides of the human body.
[0047] Subsequently, the electronic device may derive and provide a specific human body part image that includes a plurality of pre-set sub-categories belonging to a specific major category among a plurality of human body part images, and when a selection signal for a second human body location is obtained in the specific human body part image, the specific sub-categories corresponding to the second human body location among the plurality of sub-categories may be determined as the disease area. Here, referring to FIG. 5, a human body part image (face image) can be seen, and in order to display both the front and back sides of the human body part, an area that can rotate the human body part at various angles according to the user's selection signal may be provided at the bottom of the screen, similar to FIG. 4.
[0048] The electronic device can derive and provide at least one disease that may occur in a determined disease site. In other words, when a specific site is determined as in FIG. 6, it can derive all diseases that may occur, be identified, or cause pain in that site, and can list and provide the names of these diseases.
[0049] When the electronic device acquires a selection signal for a specific disease among at least one disease, it may derive and provide symptom information corresponding to the specific disease and at least one medicine. Referring to FIG. 7, the electronic device may derive at least one medicine and provide an image of the medicine's packaging, thereby assisting the user in easily discovering the medicine among several medicines.
[0050] When the electronic device acquires a selection signal for a specific drug among at least one drug, it may derive and provide at least one of the ingredient information, efficacy information, and dosage information of the specific drug. Referring to FIG. 8, the electronic device may provide detailed information about the selected specific drug. For example, the detailed information may include ingredient information, efficacy information, and dosage information. To provide detailed information about the drug, the electronic device may acquire and provide detailed information about the drug from a relevant server of the Ministry of Food and Drug Safety or another external server. When a selection signal for a specific drug is acquired, the operation of acquiring detailed information about the specific drug from a relevant server of the Ministry of Food and Drug Safety or another external server may be performed, but detailed information about the drug may be acquired in advance and stored in a storage device, and then detailed information about the specific drug may be derived from the storage device and provided.
[0051] FIG. 9 is a diagram showing major and minor categories that a user can select when selecting a body map-based area according to an embodiment of the present invention.
[0052] Referring to FIG. 9, one embodiment may construct a body map based on a full-body image of a human body and a plurality of partial images of a human body, the full-body image of a human body may represent a plurality of major classification regions, and the plurality of partial images of a human body may correspond to the plurality of major classification regions and represent a plurality of minor classification regions belonging to each major classification region.
[0053] For example, as a first human body location is selected in a full-body image of the human body, a specific major classification region corresponding to the first human body location among a plurality of major classification regions may be determined, and the plurality of major classification regions may include at least one of the head, neck, chest, back, abdomen, arms, hands, legs, and feet.
[0054] In addition, when a specific major category region is determined, a specific human body part image corresponding to that specific major category region can be derived among multiple human body part images. Here, each of the multiple human body part images may correspond to multiple major category regions, and each human body part image may be represented including multiple minor category regions belonging to the corresponding major category region.
[0055] For example, if the major classification area is the head, the plurality of minor classification areas may include at least one of the back of the head, scalp, forehead, eyes, nose, face, mouth, jaw, and ears. If the major classification area is the neck, the plurality of minor classification areas may include at least one of the back of the neck and the front of the neck. If the major classification area is the chest, the plurality of minor classification areas may include at least one of the upper chest, breast, and sternum. If the major classification area is the back, the plurality of minor classification areas may include at least one of the upper back, lower back, flank, and tailbone area. If the major classification area is the abdomen, the plurality of minor classification areas may include at least one of the sternum area, upper abdomen, and lower abdomen. If the major classification area is the arm, the plurality of minor classification areas may include at least one of the shoulder, lower arm, upper arm, elbow, and armpit. If the major classification area is the hand, the plurality of minor classification areas may include at least one of the hand, wrist, and fingers. If the major classification area is the leg, the multiple minor classification areas may include at least one of the thigh, buttocks, shin, knee, and popliteal fossa. If the major classification area is the foot, the multiple minor classification areas may include at least one of the foot, ankle, and toes.
[0056] The operation according to the embodiments of this specification may be implemented as a computer-readable program or code on a computer-readable recording medium. A computer-readable recording medium includes all types of recording devices in which data that can be read by a computer system is stored. Additionally, a computer-readable recording medium may be distributed across networked computer systems, allowing computer-readable programs or code to be stored and executed in a distributed manner.
[0057] When the embodiment is implemented in software, the above-described technique may be implemented as a module (process, function, etc.) that performs the above-described function. The module may be stored in memory and executed by a processor. The memory may be located inside or outside the processor and may be connected to the processor by various well-known means.
[0058] In addition, computer-readable recording media may include hardware devices specifically configured to store and execute program instructions, such as ROM, RAM, and flash memory. Program instructions may include machine code, such as that generated by a compiler, as well as high-level language code that can be executed by a computer using an interpreter, etc.
[0059] Some aspects of the invention have been described in the context of a device, but may also be described according to a corresponding method, wherein a block or device corresponds to a method step or a feature of a method step. Similarly, aspects described in the context of a method may also be described according to a corresponding block or item or a feature of a corresponding device. Some or all of the method steps may be performed by (or using) a hardware device, such as, for example, a microprocessor, a programmable computer, or an electronic circuit. In some embodiments, one or more of the most important method steps may be performed by such a device.
[0060] In the embodiments, a programmable logic device (e.g., a field-programmable gate array) may be used to perform some or all of the functions of the methods described herein. In the embodiments, the field-programmable gate array may operate with a microprocessor to perform one of the methods described herein. Generally, it is preferable that the methods be performed by some hardware device.
[0061] Although the present invention has been described above with reference to preferred embodiments, those skilled in the art will understand that various modifications and changes can be made to the invention without departing from the spirit and scope of the invention as described in the following claims.
Claims
Claim 1 A method for a computing device to recommend and control the dispensing of customized medicines based on artificial intelligence (AI), comprising: providing an interactive body map interface to receive a symptom area input to a user terminal; acquiring and analyzing user context and multi-modal data including the user's health records and wearable biometric data; predicting a disease based on the symptom area input through the interactive body map interface and the analyzed data, and recommending a customized medicine corresponding to the predicted disease. A method for recommending and dispensing customized medicines, comprising the step of automatically controlling the physical dispensing of the smart medicine cabinet by comprehensively determining the real-time inventory status of the smart medicine cabinet linked to the recommended customized medicines and the user's medication safety information, wherein the step of recommending the customized medicines comprises: dynamically re-ranking the priority of the optimal medicine among a plurality of medicine candidate groups for the predicted disease by synthesizing context including the real-time inventory status of the smart medicine cabinet, the user's age, allergy and drug use contraindication (DUR) information, time zone, and current weather information; and, if it is determined that the inventory of the first-ranked medicine according to the re-ranked priority is not in the smart medicine cabinet, performing a control logic to automatically switch and recommend a second-ranked alternative medicine having equivalent efficacy through ingredient cross-analysis, wherein the dynamic re-ranking of the priority comprises calculating whether it is a change of season or a fine dust index from the current weather information, and dynamically adjusting the priority weight of medicines related to respiratory diseases upward in response to the calculation result. Claim 2 A method for personalized drug recommendation and dispensing control according to claim 1, wherein the step of providing the interactive body map interface comprises collecting the user’s wearable device biometric data or past medical treatment and drug purchase history included in the multimodal data in real time, and generating a dynamic body map that preemptively highlights and renders body parts where abnormal signs are detected in the form of a visual heat map on the body map based on the collected data. Claim 3 A method for personalized drug recommendation and dispensing control according to claim 1, wherein the step of predicting the disease includes acquiring an actual image of the affected area of the corresponding area that the user inputs after selecting a specific symptom area in the interactive body map, analyzing the acquired actual image of the affected area with a Vision AI model to produce an expected disease scoring result, and executing an inference algorithm that improves disease prediction accuracy by providing a decision tree-based deep questioning chatbot based on the scoring result. Claim 4 delete Claim 5 A method for personalized drug recommendation and dispensing control according to claim 2, wherein generating the dynamic body map includes visualizing the abnormal signs by mapping the heat map indicating fever in the head or face area when an increase in body temperature is detected in the wearable device biometric data.
Citation Information
Patent Citations
System and application for providing information of health examinations
KR1020160115269A
Apparatus and system for prescribing personal-ustomized drugs or nutraceuticals using artificial intelligence
KR1020200143157A
Method and apparatus for providing drug information suitable for disease based on body map
KR1020250104380A
System to collect and identify skin conditions from images and expert knowledge
US20210133970A1
