System for rapidly providing medicine auxiliary sales and pharmaceutical service

Abstract

The invention relates to a system for rapidly providing medicine auxiliary sales and pharmaceutical services. The system comprises a user interaction module, a pharmaceutical service module and a medical interpretation module based on a large model. The user interaction module receives symptom or disease information input by a user through a graphical interface and transmits the symptom or disease information to the pharmacy service module; the pharmacy service module integrates the pharmacist experience module, the light artificial intelligence module and the medicine recommendation module, and generates a priority-sorted medicine recommendation list through the combination of experience data and an AI algorithm; the medical interpretation module based on the large model utilizes a large model technology to scientifically evaluate and interpret recommendation results, including symptom medical principles, drug action mechanisms and matters needing attention, and provides a basis for artificial recheck. The system supports cross-platform multi-terminal adaptation, adopts a touch screen priority design, comprises a double-layer member system, and customizes an exclusive drug library and a knowledge library for a pharmacy / clinic. According to the invention, rapid, safe and accurate drug recommendation is realized through module cooperation, efficiency and safety are both considered, and the system is suitable for self-service scenes of basic drugstores, clinics and patients.

Description

Technical Field

[0001] The present invention relates to the field of medical information technology, and particularly to a rapid drug-assisted sales and pharmaceutical service system based on the combination of artificial intelligence and pharmacist experience. This system integrates user interaction, pharmaceutical service recommendation, and medical explanation functions through modular design, is applicable to scenarios such as pharmacies and clinics, supports self-operation by practicing personnel or users, and aims to improve the efficiency, accuracy, and safety of drug recommendation. Background Art

[0002] The drug sales and pharmaceutical services in traditional pharmacies or clinics highly rely on the experience of licensed pharmacists and have the following problems.

[0003] 1. Insufficient efficiency: Pharmacists need to ask about symptoms one by one and manually retrieve drug treatment plans, which takes a long time.

[0004] 2. Limited experience: It is difficult for a single pharmacist to cover all complex symptoms and drug compatibility scenarios.

[0005] 3. Insufficient explanation: It is difficult for users to understand the medical principles of the recommended drugs, resulting in low compliance.

[0006] 4. Poor terminal adaptability: Existing systems mostly adopt single algorithms or manual experience, lack a fusion mechanism, and cannot flexibly adapt to the operation requirements of different devices (such as mobile terminals and intelligent terminals).

[0007] In recent years, the development of artificial intelligence technologies (such as large models and lightweight algorithms) has made automated pharmaceutical services possible, but existing solutions still have problems such as slow response speed, low credibility of recommendation results, and lack of multi-dimensional explanations. Therefore, there is an urgent need for a system that integrates artificial experience and AI technology and has both fast response and security. Summary of the Invention

[0008] Object of the Invention: The present invention aims to provide a system for rapidly providing drug-assisted sales and pharmaceutical services, to solve the problems of slow and inaccurate drug recommendation and insufficient pharmaceutical services in grass-roots pharmacies and clinics. By integrating pharmacist experience, artificial intelligence technology, and large model evaluation and explanation, it provides accurate drug recommendations and comprehensive drug knowledge explanations for patients, and at the same time provides an efficient auxiliary tool for pharmacists, ensuring drug safety and improving the quality of grass-roots medical services.

[0009] Technical Solution: The core architecture of this system includes the following modules:

[0010] 1. User Interaction Module

[0011] Function: Receive information on symptoms, diseases, or departments input by users through a graphical interface, and display recommendation results and explanations.

[0012] Design features: Optimized touchscreen design, supporting multi-dimensional interactions such as human body part atlases and symptom lists; real-time synchronous transmission of input information to the pharmaceutical service module; receiving feedback from the large model medical explanation module and presenting the medication principles and precautions in the form of graphics / text or voice.

[0013] 2. Pharmaceutical service module.

[0014] It consists of three sub-modules, forming a closed loop of "experience + AI + recommendation".

[0015] a. Pharmacist experience module: Stores a large amount of clinical experience data, classifies and matches preset medication plans according to symptoms and diseases to ensure quick response.

[0016] b. Lightweight artificial intelligence module: Adopts lightweight models such as Naive Bayes to analyze symptom data to generate recommendations and cross-validate with the results of the pharmacist experience module.

[0017] c. Drug recommendation module: Dynamically integrates the outputs of the previous two modules, supports independent recommendations of a single module or combined recommendations of two modules, and finally generates a prioritized list of drugs. It also supports displaying drug information and charging information, etc. It also supports printing the instructions and precautions, and can also choose to mark the medication taboos in red.

[0018] 3. Large model-based medical explanation module.

[0019] Calls large models (such as DeepSeek, ChatGPT, ERNIE Bot, etc.) to complete the following tasks: Analyze the medical principles of user symptoms and generate easy-to-understand explanations; analyze the mechanism of action, usage, and taboos of recommended drugs; provide reference for the reasoning process of licensed pharmacists to assist in manual review and secondary decision-making.

[0020] 4. Double-layer membership system, supporting personalized data management for institutional members (pharmacies / clinics) and user members.

[0021] 5. Multi-terminal adaptation module.

[0022] Technical implementation: Developed based on cross-platform frameworks such as Flutter / React, supporting adaptive layouts for PCs, mobile devices (HarmonyOS / iOS / Android), and smart terminals.

[0023] Interaction optimization: Prioritize adapting to touchscreen operations, reduce dependence on peripherals, and reduce the risk of cross-infection. Description of the Drawings

[0024] Figure 1 : System architecture diagram, showing the data flow and collaboration relationship between modules.

[0025] Figure 2: Schematic diagram of the user interface, with symptoms or diseases and body part selection areas marked.

[0026] Figure 3 : Workflow diagram of the pharmaceutical care module, showing independent drug recommendations by a single module and joint recommendations by two modules.

[0027] Figures 4 - 6 : In Example 1, the implementation step diagram. DETAILED DESCRIPTION

[0028] Example 1: Pharmacy scenario application.

[0029] Implementation conditions: The pharmacy is equipped with a PC terminal with good performance and connected to the Internet to ensure the real-time update of system data. Pharmacists have received system training and are familiar with the system operation process.

[0030] Implementation steps: The patient goes to the pharmacy for consultation, selects "symptoms" in the user interaction module of the PC terminal, selects "head" for the body part, and selects "flu" according to the displayed symptoms. The user interaction module transmits the symptom information to the pharmacist experience module, the light artificial intelligence module, and the drug recommendation module at the same time. The pharmacist experience module quickly matches the commonly used influenza drug recommendations. Based on the Bayesian model, the light artificial intelligence recommendation module gives preliminary diagnosis candidates within 500ms and recommends corresponding drugs. The system performs conflict detection on the results of the two. If the results are consistent, the large model evaluation and interpretation module evaluates and interprets the recommended drugs, generates medication guidance information, and displays it to the patient through the user interaction module. At the same time, the pharmacist conducts manual review and signs for confirmation. If the results are inconsistent, a review interface pops up, and the pharmacist reviews and confirms again. Finally, the system displays medication guidance information and charging information, and you can choose to print instructions and precautions, or you can choose to mark medication contraindications in red.

[0031] Implementation effect: Patients can obtain accurate medication recommendations and detailed medication instructions in a short time, which improves their satisfaction with purchasing medicines. Pharmacists can use the system to improve their work efficiency and reduce the possibility of recommendation errors.

[0032] Example 2: Clinic scenario application.

[0033] Implementation conditions: The clinic is equipped with mobile smart terminals (such as tablet computers) in the consulting room to ensure network stability. Doctors are familiar with system operation, and the system has a certain data interaction interface with the clinic's internal medical record system.

[0034] Implementation steps: The patient visits the clinic, and the doctor enters the patient's symptoms or disease information into the user interaction module of the mobile intelligent terminal. The pharmacist experience module, the lightweight artificial intelligence module, and the drug recommendation module work together to give drug recommendations. The large model evaluation and explanation module evaluates and explains the recommended drugs, and the doctor conducts manual review and signature confirmation after reference. The system pushes medication guidance knowledge to the patient. The patient can choose to print the instructions and precautions, and synchronize the current diagnosis and drug recommendation information to the clinic medical record system.

[0035] Implementation effects: Doctors can quickly obtain accurate drug recommendations, improving the diagnostic efficiency and accuracy. Patients have a deeper understanding of medication knowledge, which helps them cooperate with the treatment and improves the medical service quality of the clinic.

Claims

1. A system for quickly providing pharmaceutical assisted sales and pharmaceutical services, characterized in that, Including: a. User Interaction Module: including human body parts, symptoms, disease types, and department categories; b. Pharmaceutical Service Module: Pharmacist Experience Module, Lightweight Artificial Intelligence Module, and Drug Recommendation Module; c. Medical Explanation Module Based on Large Model: providing artificial intelligence explanations of medical principles and medication common sense based on an independent knowledge base or a comprehensive network knowledge base; This system is organically combined by the above three major modules a, b, and c, and none of them can be missing, providing fast and complete artificial intelligence pharmaceutical services. This system can be operated by pharmacy clinic practitioners or by customers themselves.

2. The system for quickly providing pharmaceutical assisted sales and pharmaceutical services according to claim 1, characterized in that, The said User Interaction Module adopts a graphical interface design, is used to receive symptom or disease information selected by the user, and transmits them to the Pharmacist Experience Module, the Lightweight Artificial Intelligence Module, and the Drug Recommendation Module respectively; At the same time, it receives the feedback information from the large model medical explanation module and displays it to the user to facilitate the user to understand the relevant explanations of the recommended medications and the results of manual review.

3. The system for quickly providing pharmaceutical assisted sales and pharmaceutical services according to claim 1, wherein The said Pharmaceutical Service Module is the integration of the years of experience summary of practicing pharmacists, namely the Pharmacist Experience Module, the Lightweight Artificial Intelligence Module, and the Drug Recommendation Module. The integration of the three modules can not only ensure the rapid and accurate medication service but also strengthen the guarantee of medication safety.

4. The pharmaceutical service module according to claim 3, wherein The said Pharmacist Experience Module stores a large amount of clinical experience data of pharmacists, covering recommended information on various symptoms or diseases and corresponding medications. This module can directly match effective medication suggestions according to the symptom or disease information input by the user, and ensure the rapidity and effectiveness of medication recommendations by virtue of the rich practical experience of pharmacists.

5. The pharmaceutical service module according to claim 3, characterized in that, The said Lightweight Artificial Intelligence Module uses a fast and accurate artificial intelligence model (such as Naive Bayes, etc., or it can also be a full-version or distilled version of a large model) to analyze and process the input symptom data. This module is linked with the Drug Recommendation Module and can quickly give drug recommendation results, and can complement the Pharmacist Experience Module to jointly ensure the rapidity, effectiveness, and safety of the recommended medications, meeting the needs of grass-roots pharmacies and clinics for rapid and accurate drug recommendations. Safe medication is the focus and advantage of this module. The difference between the two concepts of lightweight and heavyweight lies in whether the system can quickly and accurately complete auxiliary judgment and drug recommendation, and the relative consumption of computing power. With the passage of time and the progress of technology, the currently computationally intensive large model can also become the configuration of the Lightweight Artificial Intelligence Module.

6. The pharmaceutical service module according to claim 3, wherein The said Drug Recommendation Module supports independent drug recommendation by a single module (Pharmacist Experience Module or Lightweight Artificial Intelligence Module), and also supports the combination of these two modules into one, and finally generates a list of medications sorted by priority. The Drug Recommendation Module also includes the display of drug information (such as instructions, medication precautions) and the display of charging information (such as charging QR code), etc. After the instructions and precautions are displayed, the user can choose to print the instructions and precautions, or choose to mark the medication taboos in red.

7. The system for quickly providing pharmaceutical assisted sales and pharmaceutical services according to claim 1, characterized in that, The medical interpretation module based on large models uses the newly emerging large model technologies (including but not limited to deepseek, ChatGPT, Douyin Doubao, Tencent Hunyuan, Alibaba Qianwen, Baidu Wenxin Yiyan, etc.) to perform artificial intelligence analysis on patients' symptoms or diseases, and interpret and evaluate the medication recommendations given by the pharmaceutical service module. On the one hand, in a way of artificial intelligence-assisted education that is easy for customers to accept, it explains the medical knowledge and principles behind symptoms and diseases, and at the same time can also introduce the mechanism of action, usage methods, and precautions of the recommended drugs, etc.; on the other hand, it also provides a more detailed and in-depth reference basis or even the medication reasoning process for store clerks or licensed pharmacists, providing an additional opportunity for pharmacy staff to review and verify, further ensuring medication safety and achieving multiple benefits at once.

8. The system for quickly providing pharmaceutical assisted sales and pharmaceutical services according to claim 1, characterized in that, The difference between the lightweight artificial intelligence module in the pharmaceutical service module and the medical interpretation module based on large models is that the former has higher accuracy and response speed than the latter, while the latter has a better knowledge breadth than the former. The former focuses on quick and accurate judgment, and the latter focuses on making a comprehensive and scientific explanation of the judgment of the former. With the development and iteration of artificial intelligence models and the progress of computing power, the two artificial intelligence modules may replace each other or even be merged into one.

9. The system for quickly providing pharmaceutical assisted sales and pharmaceutical services according to claim 1, characterized in that, The system includes a two-tier membership system. Pharmacies or clinics using this system will register as members of this system, and the system will customize a proprietary drug library and knowledge base for the pharmacies or clinics. Pharmacy customers or clinic patients using this system will automatically register as members of the pharmacy or clinic.

10. The system for quickly providing pharmaceutical assisted sales and pharmaceutical services according to claim 1, characterized in that, The system includes a multi-terminal adaptation module, which realizes the seamless adaptation of the system on the PC side, mobile terminals (HarmonyOS / iOS / Android), and intelligent terminal devices based on a cross-platform development framework, supports adaptive responsive design, and can be customized for individual systems according to different places. In addition to using traditional computer peripherals, this system is especially suitable for using touch screen devices, facilitating the interaction between pharmacists, physicians, and patient customers.

Images