Remote medication guidance scheme pushing method and system based on AI multi-model fusion
The remote medication guidance system, which integrates AI multi-model fusion and professional pharmacist review, solves the problems of outdated drug instructions and inconsistencies between AI answers, enabling efficient delivery of simplified medication guidance and improving the accuracy and convenience of medication use.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-08
- Publication Date
- 2026-03-13
AI Technical Summary
Existing drug instructions are outdated and incomplete, making them difficult for ordinary patients to understand. Inconsistent answers from AI models lead to inaccurate medication guidance, and professional guidance is time-consuming and laborious, leaving patients unable to obtain accurate medication guidance.
Using AI multi-model fusion technology, and after review by professional pharmacists, drug information is integrated to generate simplified medication guidance text, which is then revised based on the original manufacturer's instructions and pushed remotely via a mobile application.
It improves the efficiency and accuracy of medication guidance, enabling non-professionals to understand drug instructions and achieving convenient and accurate medication guidance.
Smart Images
Figure CN121662330A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of medical and health technology, and in particular to a method and system for remote medication guidance program delivery based on AI multi-model fusion. Background Technology
[0002] Currently, drug guidance mainly relies on drug instructions, but drug instructions are often difficult to understand, especially foreign drug instructions. When necessary, guidance from a professional doctor or pharmacist is required. However, relying on a doctor or pharmacist is time-consuming, expensive, laborious, and inefficient.
[0003] Statistics show that drug instructions are updated on average only once every 10 years. The content of these instructions is outdated and incomplete, often out of touch with actual clinical needs. For example, many instructions only provide adult dosages, failing to address questions such as how to calculate dosages for children, how to administer the medication, the duration of treatment, the appropriate use for special populations, and dosage adjustments. These instructions simply state to use the medication as directed by a doctor or under the guidance of a pharmacist, causing significant confusion for patients.
[0004] With the proliferation of AI big data models, patients can now ask these models questions to obtain information on medication usage. However, different AI models provide varying answers due to different question formats, and even the same model can offer inconsistent medication guidance. Ordinary patients, lacking medical expertise, are unable to discern the correctness of these answers. Furthermore, the diverse question formats lead to wildly different AI responses, leaving many patients without accurate medication guidance because they don't know how to ask the right questions. Summary of the Invention
[0005] The purpose of this invention is to provide a method and system for remote medication guidance based on AI multi-model fusion. It integrates and leverages the advantages of existing large models to translate complex drug instructions into a form that non-professionals can understand. By integrating with the drugs purchased by patients, it provides remote medication guidance in language that patients can easily understand through a mobile application, thereby improving the efficiency and accuracy of medication guidance.
[0006] To achieve the above objectives, this invention provides a method for remote medication guidance scheme delivery based on AI multi-model fusion, comprising the following steps: S1. Collect multi-dimensional information on drugs circulating in the market, break down and organize the information according to indications, usage and dosage, precautions and medication use for special populations, and build a standardized family medication guidance database. S2. For a single drug, set questions based on indications, dosage, precautions, contraindications, interactions, adverse reactions, and references to ask existing AI models, obtain preliminary medication guidance answers, and output simplified drug instructions. S3. Professional pharmacists compare the initial answers from various AI models and review and revise them in conjunction with the original manufacturer's instructions for the corresponding drugs. When the AI answers are inconsistent with the instructions, the instructions shall prevail, and a medication guidance text shall be formed. S4. Enter the medication instructions into the system according to the corresponding modules, and upload the original drug instructions to the system; S5. For drugs with the same generic name, identify the key issues and break them down into indications, dosage and administration, course of treatment, special populations, lifestyle, and precautions. Specifically, for dosage and administration, break it down into usage, timing of administration, dosage adjustment recommendations, and course of treatment. Ask questions to the existing AI model and ask it to generate a simplified version of the drug instruction manual based on the original manufacturer's instructions. S6. Professional pharmacists compare the answers given by the AI model to form a preliminary medication guidance text. Then, professional pharmacists refer to the drug instructions for different dosage forms and specifications of the same generic drug to check whether the answers given by each model are consistent with the instructions. If there are any inconsistencies, the drug instructions shall prevail and the text shall be revised again. After confirming that there are no errors, the final medication guidance text shall be compiled. S7. Enter the organized medication text content into the corresponding modules in the system according to module classification; push it to users through the single-drug guidance module and multi-drug guidance module of the mobile APP and mini program. The multi-drug guidance module supports viewing multiple drug information by swiping left and right and generating and forwarding multi-drug combination medication guidance sheets.
[0007] Preferably, in S1, the multi-dimensional information of the drug includes drug image, original manufacturer's instructions, name, specifications, manufacturer, medical insurance type, prescription / OTC attribute, and national standard code.
[0008] Preferably, in S5, the dosage adjustment recommendations include subdivided dimensions such as postoperative thrombosis prevention dosage, dosage for patients over 65 years of age, dosage for patients with liver or kidney dysfunction, and rules for adjusting the dosage after missed doses.
[0009] Preferably, in S5, the special population includes pregnant women, children, the elderly, and those with liver or kidney dysfunction, and the final medication guidance text clearly specifies the rules for contraindication / caution in medication use and corresponding dosage adjustment suggestions for such populations.
[0010] This invention also provides a remote medication guidance scheme push system based on AI multi-model fusion, including: Drug Information Integration Module: Used to collect images, original manufacturer's instructions, names, specifications, manufacturers, medical insurance types, prescription / OTC attributes and national standard codes of drugs circulating in the market, and to disassemble, organize and build a standardized family medication guidance database according to preset modules; AI Multi-Model Analysis Module: Used to ask questions to multiple large AI models about drug-related issues and obtain preliminary answers; Manual review and correction module: This module is used by professional pharmacists to compare the initial answers from various AI models, and then review and correct them in conjunction with the original drug manufacturer's instructions to form the final medication guidance text. System storage module: used to store the final medication guidance text and multi-dimensional basic information about the drug, categorized by module; Mobile terminal display and push module: integrated into mobile APP and mini program, including single drug guidance unit and multi-drug guidance unit. The single drug guidance unit displays complete medication guidance information for a single drug. The multi-drug guidance unit supports sliding viewing of multi-drug information, generation and forwarding of multi-drug combination medication guidance sheets, realizing remote push and convenient viewing of medication guidance plans.
[0011] Preferably, the drug information integration module also supports dynamic updates of drug information. When the original drug manufacturer's instructions are updated or new drugs are added to the market, the corresponding information in the standardized family medication guidance database is updated simultaneously.
[0012] Preferably, the mobile terminal display and push module supports the function of saving medication guidance texts, so that users can quickly retrieve and view the medication guidance information of the target drug later.
[0013] Therefore, the present invention adopts the above-mentioned remote medication guidance scheme push method and system based on AI multi-model fusion, which integrates and leverages the advantages of existing large models, translates complex drug instructions into a form that non-professionals can understand, and integrates them with the drugs purchased by patients, using language that patients can easily understand, to remotely guide medication use through a mobile application, thereby improving the efficiency and accuracy of medication guidance.
[0014] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description
[0015] Figure 1 This is a multi-dimensional information diagram of the drug according to an embodiment of the present invention; Figure 2 This is the usage and dosage interface of the medication guidance text in this embodiment of the invention; Figure 3 This is the precautions interface of the medication guidance text in this embodiment of the invention; Figure 4 This refers to all existing and circulating ezetimibe tablets involved in the embodiments of this invention; Figure 5 This is the final medication guidance text of the embodiments of the present invention; Figure 6 This is the usage and dosage input interface of this invention embodiment; Figure 7 This is a special population input interface according to an embodiment of the present invention; Figure 8 This is a mobile instruction manual demonstration of an embodiment of the present invention; Figure 9 This is an embodiment of the present invention that generates AI-based medication guidance for ezetimibe tablets; Figure 10 This is the multi-drug guidance interface of an embodiment of the present invention; Figure 11 This is the multi-drug guidance sliding viewing interface according to an embodiment of the present invention; Figure 12 This is a flowchart of an embodiment of the present invention. Detailed Implementation
[0016] The technical solution of the present invention will be further described below with reference to the accompanying drawings and embodiments.
[0017] Unless otherwise defined, the technical or scientific terms used in this invention shall have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms "first," "second," and similar terms used in this invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed following the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.
[0018] Example 1 This invention provides a method for remote medication guidance scheme delivery based on AI multi-model fusion, the overall process of which is as follows: Figure 12 As shown, it includes the following steps: S1. Collect multi-dimensional information on market-circulated drugs, break down and organize the information according to indications, dosage, precautions, and medication use for special populations, and construct a standardized family medication guidance database; the multi-dimensional information of drugs includes drug images, original manufacturer's instructions, name, specifications, manufacturer, medical insurance type, prescription / OTC attribute, and national standard code, specifically as follows: Figure 1 As shown.
[0019] S2. For a single drug, this embodiment takes ezetimibe tablets as an example. According to the indications, dosage, precautions, contraindications, interactions, adverse reactions and references, questions are set to ask the existing AI model to obtain preliminary medication guidance answers and output a simplified drug instruction manual.
[0020] S3. Professional pharmacists compare the initial answers from various AI models and review and revise them in conjunction with the original manufacturer's instructions for the corresponding drugs. When the AI answers are inconsistent with the instructions, the instructions shall prevail, and a medication guidance text shall be generated. Figure 2 and Figure 3 The page displays the system's output, including usage instructions, dosage, and precautions.
[0021] S4. Enter the medication instructions into the system according to the corresponding modules, and upload the original drug instructions to the system.
[0022] S5. Identify and address issues related to drugs with the same generic name, using ezetimibe tablets as an example. Figure 4 The image shows all currently available ezetimibe tablets, broken down into indications, dosage and administration, course of treatment, special populations, lifestyle, and precautions. The dosage and administration section is further broken down into usage, timing of administration, dosage adjustment recommendations, and course of treatment. Questions are posed to the existing AI model, requiring it to generate a simplified version of the drug instructions based on the original manufacturer's instructions. The dosage adjustment recommendations are further subdivided into several dimensions, including postoperative thrombosis prevention dosage, dosage for patients over 65 years of age, dosage for patients with hepatic or renal insufficiency, and rules for adjusting missed doses. Special populations include pregnant women, children, the elderly, and patients with hepatic or renal insufficiency. The final medication guidance text clearly specifies the rules for contraindications / cautions and corresponding dosage adjustment recommendations for these populations.
[0023] S6. Professional pharmacists compare the answers provided by the AI model to create a preliminary medication guidance text. Then, the professional pharmacists refer to the package inserts for different dosage forms and specifications of the same generic drug to verify that the answers provided by each model are consistent with the package inserts. If there are discrepancies, the package inserts prevail, and the text is revised again. After confirming that everything is correct, the final medication guidance text is compiled. Figure 5 As shown.
[0024] S7. Enter the organized medication text content into the corresponding modules in the system according to module classification; Figure 6 and Figure 7 The interface for inputting usage, dosage, and information on special populations is displayed. This is also available through the single-drug guidance module in the mobile app and mini-program. Figure 8 This is a demonstration of the instruction manual for mobile devices. Figure 9 The demonstration showcased AI-generated medication guidance for ezetimibe tablets and a multi-drug guidance module. Figure 10 As shown, the multi-drug guidance module pushes information to users, allowing them to view multiple drug information by swiping left and right. Figure 11 (as shown) and the generation and forwarding of multidrug combination therapy instructions.
[0025] The remote medication guidance system based on AI multi-model fusion used in this method includes: The drug information integration module is used to collect images, original manufacturer's instructions, names, specifications, manufacturers, medical insurance types, prescription / OTC attributes, and national standard codes of drugs circulating in the market. It is then broken down and organized according to preset modules to build a standardized family medication guidance database. The drug information integration module also supports dynamic updates of drug information. When the original manufacturer's instructions are updated or new drugs are added to the market, the corresponding information in the standardized family medication guidance database is updated synchronously.
[0026] AI Multi-Model Analysis Module: Used to ask questions to multiple large AI models about drug-related issues and obtain preliminary answers; Manual review and correction module: This module is used by professional pharmacists to compare the initial answers from various AI models, and then review and correct them in conjunction with the original drug manufacturer's instructions to form the final medication guidance text. System storage module: used to store the final medication guidance text and multi-dimensional basic information about the drug, categorized by module; Mobile display and push module: Integrated into mobile apps and mini-programs, including single-drug guidance units and multi-drug guidance units. The single-drug guidance unit displays complete medication guidance information for a single drug, while the multi-drug guidance unit supports swiping to view information on multiple drugs, generating and forwarding multi-drug combination medication guidance sheets, enabling remote push and convenient viewing of medication guidance plans. It also supports a function to save medication guidance texts, allowing users to quickly retrieve and view medication guidance information for target drugs later.
[0027] Therefore, the present invention adopts the above-mentioned remote medication guidance scheme push method and system based on AI multi-model fusion, which integrates and leverages the advantages of existing large models, translates complex drug instructions into a form that non-professionals can understand, and integrates them with the drugs purchased by patients, using language that patients can easily understand, to remotely guide medication use through a mobile application, thereby improving the efficiency and accuracy of medication guidance.
[0028] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the technical solutions of the present invention, and these modifications or equivalent substitutions cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.
Claims
1. A method for remote medication guidance program delivery based on AI multi-model fusion, characterized in that, Includes the following steps: S1. Collect multi-dimensional information on drugs circulating in the market, break down and organize the information according to indications, usage and dosage, precautions and medication use for special populations, and build a standardized family medication guidance database. S2. For a single drug, set questions based on indications, dosage, precautions, contraindications, interactions, adverse reactions, and references to ask existing AI models, obtain preliminary medication guidance answers, and output simplified drug instructions. S3. Professional pharmacists compare the initial answers from various AI models and review and revise them in conjunction with the original manufacturer's instructions for the corresponding drugs. When the AI answers are inconsistent with the instructions, the instructions shall prevail, and a medication guidance text shall be formed. S4. Enter the medication instructions into the system according to the corresponding modules, and upload the original drug instructions to the system; S5. For drugs with the same generic name, identify the key issues and break them down into indications, dosage and administration, course of treatment, special populations, lifestyle, and precautions. Specifically, for the dosage and administration section, break it down into usage, timing of administration, dosage adjustment recommendations, and course of treatment. Ask questions to the existing AI model and ask it to generate a simplified version of the drug instruction manual based on the original manufacturer's instructions. S6. Professional pharmacists compare the answers given by the AI model to form a preliminary medication guidance text. Then, professional pharmacists refer to the drug instructions for different dosage forms and specifications of the same generic drug to check whether the answers given by each model are consistent with the instructions. If there are any inconsistencies, the drug instructions shall prevail and the text shall be revised again. After confirming that there are no errors, the final medication guidance text shall be compiled. S7. Enter the organized medication text content into the corresponding modules in the system according to module classification; push it to users through the single-drug guidance module and multi-drug guidance module of the mobile APP and mini program. The multi-drug guidance module supports viewing multiple drug information by swiping left and right and generating and forwarding multi-drug combination medication guidance sheets.
2. The method for remote medication guidance program delivery based on AI multi-model fusion according to claim 1, characterized in that, In S1, the multi-dimensional information of a drug includes a drug image, original manufacturer's instructions, name, specifications, manufacturer, medical insurance type, prescription / OTC attribute, and national standard code.
3. The method for remote medication guidance program delivery based on AI multi-model fusion according to claim 1, characterized in that, In S5, the dosage adjustment recommendations are further subdivided into the following dimensions: dosage for postoperative thrombosis prevention, dosage for patients over 65 years of age, dosage for patients with liver or kidney dysfunction, and rules for adjusting the dosage after a missed dose.
4. The method for remote medication guidance program delivery based on AI multi-model fusion according to claim 1, characterized in that, In S5, the special populations include pregnant women, children, the elderly, and those with liver or kidney dysfunction. The final medication guidance text clearly defines the rules for contraindications / cautions for these populations and corresponding dosage adjustment recommendations.
5. A remote medication guidance plan push system based on AI multi-model fusion, characterized in that, include: Drug Information Integration Module: Used to collect images, original manufacturer's instructions, names, specifications, manufacturers, medical insurance types, prescription / OTC attributes and national standard codes of drugs circulating in the market, and to disassemble, organize and build a standardized family medication guidance database according to preset modules; AI Multi-Model Analysis Module: Used to ask questions to multiple large AI models about drug-related issues and obtain preliminary answers; Manual review and correction module: This module is used by professional pharmacists to compare the initial answers from various AI models, and then review and correct them in conjunction with the original drug manufacturer's instructions to form the final medication guidance text. System storage module: used to store the final medication guidance text and multi-dimensional basic information about the drug, categorized by module; Mobile terminal display and push module: integrated into mobile APP and mini program, including single drug guidance unit and multi-drug guidance unit. The single drug guidance unit displays complete medication guidance information for a single drug. The multi-drug guidance unit supports sliding viewing of multi-drug information, generation and forwarding of multi-drug combination medication guidance sheets, realizing remote push and convenient viewing of medication guidance plans.
6. The remote medication guidance scheme push system based on AI multi-model fusion according to claim 5, characterized in that, The drug information integration module also supports dynamic updates of drug information. When the original drug manufacturer's instructions are updated or new drugs are added to the market, the corresponding information in the standardized family medication guidance database is updated simultaneously.
7. The remote medication guidance scheme push system based on AI multi-model fusion according to claim 5, characterized in that, The mobile display and push module supports the function of saving medication guidance texts, making it convenient for users to quickly retrieve and view the medication guidance information for the target drug later.