Medical product intelligent business card system and method based on artificial intelligence

Abstract

The invention discloses a medical product intelligent business card system and method based on artificial intelligence, and belongs to the technical field of artificial intelligence and medical information. The system comprises a client layer, an AI engine layer, a digital twin layer and a data layer. According to the method, user information is collected through a client, intelligent analysis is performed by using a privately deployed AI large model to identify user requirements, then the requirements are matched with a medicine product database, and a digital twin engine is called to generate a corresponding three-dimensional visual model for interactive display. The defects that a traditional medicine product display mode is single, interactivity is poor and information updating is not timely are overcome, personalized and immersive dynamic display of medicine product information is achieved through the intelligent requirement recognition and digital twinborn visualization technology, meanwhile, data safety and compliance are guaranteed, and user experience is improved. And the information transmission efficiency and the user experience are remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of interdisciplinary application technology of artificial intelligence and information technology in the pharmaceutical field, specifically to an intelligent business card system and method for pharmaceutical products based on artificial intelligence. Background Technology

[0002] With the accelerating digitalization of the pharmaceutical industry, the methods of displaying and promoting pharmaceutical product information urgently need innovation. Traditional models mainly rely on paper materials or static electronic documents, which have inherent defects such as limited display formats, lagging information updates, and lack of interactivity. They are unable to clearly convey the characteristics and professional data of complex drugs and cannot meet the needs of medical professionals for in-depth and dynamic information.

[0003] Existing electronic solutions, such as basic digital business cards or customer relationship management systems, while achieving a certain degree of information aggregation, have significant shortcomings in terms of intelligence. They generally lack the ability to accurately identify users' professional intentions and cannot effectively integrate multi-source heterogeneous information such as internal product data, external market intelligence, and real-time interaction logs. This results in static and one-sided user profiles that are difficult to support truly personalized content recommendations and in-depth interactions.

[0004] Furthermore, on the supply chain and production side, existing methods mostly focus on supplier management or rigid production processes, failing to efficiently and accurately map the personalized needs of front-end users into product display solutions or production guidance. This presents a bottleneck in meeting highly customized customer needs and ensuring efficient information transmission. Therefore, there is an urgent need for an integrated solution that can intelligently understand needs, dynamically visualize them, and ensure data security and compliance. Summary of the Invention

[0005] In view of the above-mentioned technical problems in related technologies, the present invention proposes an intelligent business card system and method for pharmaceutical products based on artificial intelligence, which can overcome the above-mentioned shortcomings of the prior art.

[0006] To achieve the above-mentioned technical objectives, the technical solution of the present invention is implemented as follows: A smart business card system for pharmaceutical products based on artificial intelligence; This AI-based smart business card system for pharmaceutical products includes: The client layer is used to receive user input information; The AI ​​engine layer communicates with the client layer. The AI ​​engine layer has a privately deployed large AI model, which is used to intelligently identify and process the user input information in order to obtain user needs. The digital twin layer, which is communicatively connected to the AI ​​engine layer, is used to generate a 3D visualization model of the corresponding pharmaceutical product based on the user's needs. The data layer is communicatively connected to the AI ​​engine layer and the digital twin layer. The data layer includes a pharmaceutical product database to provide data support for the AI ​​engine layer and the digital twin layer.

[0007] Furthermore, the client layer includes H5 mobile applications and the official website; the AI ​​engine layer includes: The intelligent interaction module, deployed in the client layer, is used to collect the user input information; The AI ​​demand identification module, based on the privately deployed AI large model, is used to analyze the user input information and identify the user demand. The product matching module is used to match the identified user needs with the pharmaceutical product database to determine the target pharmaceutical product. The digital twin generation module is used to generate corresponding three-dimensional visualization model generation instructions based on the target pharmaceutical product.

[0008] Furthermore, the AI ​​demand recognition module introduces a pre-training mechanism enhanced by medical knowledge graphs, constructs domain-specific word vectors by integrating multi-source medical structured data, and optimizes intent recognition, entity extraction, and drug relationship reasoning tasks using a multi-task learning framework. The digital twin layer includes a pharmaceutical product digital twin generation and display module, a digital twin editing module, and a digital twin rendering module. The pharmaceutical product digital twin generation and display module has pre-set three-dimensional templates for various drug dosage forms and automatically adapts the size, material, and color according to the product data. It uses WebGL and a lightweight mesh compression algorithm to generate the three-dimensional visualization model.

[0009] Furthermore, the data layer also includes a user interaction database and a multi-source heterogeneous data integration module; the multi-source heterogeneous data integration module is used to standardize data from different channels and then store it in the data layer.

[0010] Furthermore, the AI ​​engine layer incorporates a federated learning framework and / or supports differential privacy processing, and the system's external interface complies with preset medical data compliance standards.

[0011] Furthermore, the product matching module introduces a reinforcement learning-driven recommendation strategy to dynamically adjust matching weights based on user interaction behavior; the system also has an A / B testing platform for optimizing the AI ​​recommendation model based on conversion rate data.

[0012] According to another aspect of the present invention, a method for implementing a smart business card for pharmaceutical products based on artificial intelligence is provided; This AI-based method for creating smart business cards for pharmaceutical products, applied to the aforementioned system, includes the following steps: User input information is collected through the client-side layer; The AI ​​large model deployed in a private location is used to intelligently analyze the user input information, identify user needs, and generate a structured description of the needs. The structured requirements description is matched with a pharmaceutical product database to filter out target pharmaceutical products; The digital twin engine is invoked to generate a three-dimensional visualization model of the target pharmaceutical product; The 3D visualization model is interactively displayed at the client layer.

[0013] Furthermore, the intelligent analysis of user input information includes: based on a pre-trained model enhanced with a medical knowledge graph, a multi-task learning framework is used for intent recognition, entity extraction, and drug relationship reasoning; The process of selecting target pharmaceutical products includes: introducing a reinforcement learning-driven recommendation strategy and dynamically adjusting matching weights based on user interaction behavior; The process of generating a 3D visualization model includes: calling a pre-set 3D template of a drug dosage form, automatically adapting the size, material, and color according to the data of the target pharmaceutical product, and rendering it using WebGL and a lightweight mesh compression algorithm.

[0014] Furthermore, the method also includes: Collect user interaction data with the 3D visualization model; Based on the interaction data, optimize the AI ​​big model and / or product matching algorithm.

[0015] Furthermore, in the intelligent analysis process, a federated learning framework is used for model iteration and / or differential privacy processing is used to inject controllable noise into the data.

[0016] The beneficial effects of this invention are as follows: By using a domain-specific AI model deployed privately, user needs are intelligently identified and accurately analyzed. Furthermore, digital twin technology is used to generate interactive 3D visualized product models, transforming the presentation of pharmaceutical product information from static and two-dimensional to dynamic, three-dimensional, and deeply interactive, effectively enhancing the intuitiveness of information delivery and user participation. This, in turn, provides medical professionals with a real-time, accurate, and personalized product understanding experience while fully ensuring the security and compliance of pharmaceutical data, strongly supporting the professional promotion and decision-making of pharmaceutical products. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a schematic diagram of the structure of an artificial intelligence-based smart business card system for pharmaceutical products according to an embodiment of the present invention. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention are within the scope of protection of the present invention.

[0020] It should be understood that in the description of the embodiments of the present invention, the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," and "counterclockwise," etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings. They are used only for the convenience of describing the embodiments of the present invention and for simplifying the description, and do not indicate or imply that the referred device or element must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the embodiments of the present invention. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, features defined with "first" and "second" may explicitly or implicitly include one or more of the stated features. In the description of the embodiments of the present invention, "several" means two or more, unless otherwise explicitly specified.

[0021] like Figure 1 As shown in the embodiment of the present invention, a smart business card system for pharmaceutical products based on artificial intelligence includes: The client layer is used to receive user input information; The AI ​​engine layer communicates with the client layer. The AI ​​engine layer has a privately deployed large AI model, which is used to intelligently identify and process the user input information in order to obtain user needs. The digital twin layer, which is communicatively connected to the AI ​​engine layer, is used to generate a 3D visualization model of the corresponding pharmaceutical product based on the user's needs. The data layer is communicatively connected to the AI ​​engine layer and the digital twin layer. The data layer includes a pharmaceutical product database to provide data support for the AI ​​engine layer and the digital twin layer.

[0022] According to an embodiment of the present invention, a smart business card system for pharmaceutical products based on artificial intelligence, in a specific implementation, includes an H5 mobile application and an official website in the client layer; and an AI engine layer including: The intelligent interaction module, deployed in the client layer, is used to collect the user input information; The AI ​​demand identification module, based on the privately deployed AI large model, is used to analyze the user input information and identify the user demand. The product matching module is used to match the identified user needs with the pharmaceutical product database to determine the target pharmaceutical product. The digital twin generation module is used to generate corresponding three-dimensional visualization model generation instructions based on the target pharmaceutical product.

[0023] According to an embodiment of the present invention, an AI-based intelligent business card system for pharmaceutical products is described. In a specific implementation, the AI ​​demand identification module introduces a pre-training mechanism enhanced by pharmaceutical knowledge graphs. It constructs domain-specific word vectors by fusing multi-source pharmaceutical structured data and optimizes intent recognition, entity extraction, and drug relationship reasoning tasks using a multi-task learning framework. The digital twin layer includes a pharmaceutical product digital twin generation and display module, a digital twin editing module, and a digital twin rendering module. The pharmaceutical product digital twin generation and display module has pre-set three-dimensional templates for various drug dosage forms and automatically adapts the size, material, and color according to the product data. It uses WebGL and a lightweight mesh compression algorithm to generate the three-dimensional visualization model.

[0024] According to an embodiment of the present invention, a smart business card system for pharmaceutical products based on artificial intelligence is provided. In a specific implementation, the data layer further includes a user interaction database and a multi-source heterogeneous data integration module. The multi-source heterogeneous data integration module is used to standardize data from different channels and then store it in the data layer.

[0025] According to an embodiment of the present invention, a smart business card system for pharmaceutical products based on artificial intelligence is provided. In a specific implementation, the AI ​​engine layer introduces a federated learning framework and / or supports differential privacy processing, and the system's external interface complies with preset medical data compliance standards.

[0026] According to an embodiment of the present invention, an AI-based smart business card system for pharmaceutical products is provided. In a specific implementation, the product matching module introduces a reinforcement learning-driven recommendation strategy to dynamically adjust the matching weights based on user interaction behavior. The system also includes an A / B testing platform for optimizing the AI ​​recommendation model based on conversion rate data.

[0027] Secondly, according to an embodiment of the present invention, a method for implementing a smart business card for pharmaceutical products based on artificial intelligence includes: User input information is collected through the client-side layer; The AI ​​large model deployed in a private location is used to intelligently analyze the user input information, identify user needs, and generate a structured description of the needs. The structured requirements description is matched with a pharmaceutical product database to filter out target pharmaceutical products; The digital twin engine is invoked to generate a three-dimensional visualization model of the target pharmaceutical product; The 3D visualization model is interactively displayed at the client layer.

[0028] According to an embodiment of the present invention, a method for implementing a smart business card for pharmaceutical products based on artificial intelligence, in a specific implementation, the intelligent analysis of user input information includes: based on a pre-trained model enhanced with a pharmaceutical knowledge graph, using a multi-task learning framework to perform intent recognition, entity extraction, and drug relationship reasoning; The process of selecting target pharmaceutical products includes: introducing a reinforcement learning-driven recommendation strategy and dynamically adjusting matching weights based on user interaction behavior; The process of generating a 3D visualization model includes: calling a pre-set 3D template of a drug dosage form, automatically adapting the size, material, and color according to the data of the target pharmaceutical product, and rendering it using WebGL and a lightweight mesh compression algorithm.

[0029] According to an embodiment of the present invention, a method for implementing a smart business card for pharmaceutical products based on artificial intelligence, in a specific embodiment, the method further includes: Collect user interaction data with the 3D visualization model; Based on the interaction data, optimize the AI ​​big model and / or product matching algorithm.

[0030] According to an embodiment of the present invention, a method for implementing a smart business card for pharmaceutical products based on artificial intelligence is described. In a specific implementation, during the intelligent analysis process, a federated learning framework is used for model iteration and / or differential privacy processing is used to inject controllable noise into the data.

[0031] The present invention will be further described in detail below with reference to embodiments, but the implementation of the present invention is not limited thereto.

[0032] I. System Overall Architecture To achieve the above objectives, the system of the present invention adopts a layered architecture design. (See attached diagram.) Figure 1 The system architecture diagram shown mainly includes a client layer, an AI engine layer, a digital twin layer, and a data layer.

[0033] The client layer serves as the user interaction entry point, used to receive user input information. In a specific embodiment, the client layer can be implemented as an H5 mobile application or an official website, providing users with an access interface and handling data interaction with backend services.

[0034] The AI ​​engine layer, which communicates with the client layer, is its core intelligent processing unit. This layer is privately deployed with a dedicated large-scale AI model to ensure data security and domain expertise in data processing. The AI ​​engine layer is responsible for intelligently recognizing, semantically understanding, and analyzing the user input submitted by the client layer to accurately obtain user intent.

[0035] The digital twin layer communicates with the AI ​​engine layer and is responsible for visualization. This layer receives the processing results from the AI ​​engine layer and generates a 3D visualization model (i.e., a digital twin) of the corresponding pharmaceutical product, transforming complex pharmaceutical product information into an intuitive and interactive three-dimensional display.

[0036] The data layer serves as the data foundation of the entire system, maintaining communication connections with both the AI ​​engine layer and the digital twin layer. The data layer includes at least a pharmaceutical product database, used to store and manage structured information about various pharmaceutical products, providing data support for AI analysis, product matching, and model generation.

[0037] II. Detailed Structure and Interaction of Each Layer Detailed Structure of the AI ​​Engine Layer The AI ​​engine layer further integrates multiple functional modules to work collaboratively: Intelligent Interaction Module: Deployed on the client side, it is responsible for collecting information input by users through multimodal methods such as text, voice, and images.

[0038] The AI-powered demand identification module, based on the aforementioned privately deployed large language model, performs in-depth analysis of the collected information. This module employs a pre-training mechanism enhanced with a pharmaceutical knowledge graph, constructing domain-specific word vectors by integrating multi-source structured data such as the Chinese Pharmacopoeia, drug instructions, and clinical guidelines. During model training, a multi-task learning framework can be used to simultaneously optimize intent recognition, entity extraction, and drug relationship reasoning tasks, significantly improving the accuracy of semantic understanding within a pharmaceutical professional context.

[0039] The product matching module matches and filters the structured user needs output by the AI-powered demand identification module against a pharmaceutical product database in the data layer to determine the most suitable target product. This module can incorporate reinforcement learning strategies to dynamically adjust matching weights based on users' historical interaction behavior.

[0040] Digital twin generation module: Based on the target product information selected by the product matching module, it generates corresponding 3D visualization model generation instructions and sends them to the digital twin layer.

[0041] Detailed Composition of Digital Twin Layer The digital twin layer receives and executes instructions from the AI ​​engine layer: The pharmaceutical product digital twin generation and display module is the core model generation unit. This module can pre-load a parametric 3D template library for various drug dosage forms and automatically adapt the template's geometric attributes, materials, and colors based on specific product data such as size and properties. Employing WebGL technology and a lightweight mesh compression algorithm, it effectively controls model data volume while ensuring high-quality visual rendering, supporting smooth loading and real-time interaction on mobile platforms.

[0042] Digital Twin Editing Module: Provides users with the ability to edit the generated digital twin model, such as adjusting product appearance attributes and changing surface materials, to meet customized display needs.

[0043] Digital Twin Rendering Module: Responsible for outputting the finalized digital twin model in various formats, such as formats for web page interaction, VR models, or promotional videos.

[0044] Detailed structure of the data layer In addition to the basic pharmaceutical product database, the data layer may also include: User interaction database: Used to store user actions, session history, and other data within the system.

[0045] Multi-source heterogeneous data integration module: Responsible for connecting and standardizing data from different internal and external channels. This data can include biomedical data, medical and health data, and patient vital sign data collected through IoT devices, as well as professional information obtained from expert databases, market sentiment, and other sources. After integration, it is stored in the corresponding database to support the construction of dynamic and complete user profiles and product knowledge graphs.

[0046] III. System Collaborative Working Mechanism and Process The various layers of the system work together through defined interfaces and protocols. The core process for displaying intelligent business cards for pharmaceutical products is as follows: User needs collection: Users input their needs through the client layer, such as target symptoms, usage scenario preferences, etc.

[0047] AI-powered intelligent analysis: The client sends user input to the AI ​​engine layer. A private AI big data model parses the information, identifies user intent, and generates a structured description of user needs.

[0048] Product matching: The product matching module of the AI ​​engine layer compares structured requirements with the pharmaceutical product database and uses matching algorithms to select the most relevant target pharmaceutical products.

[0049] Digital twin generation: Based on the target product information, the digital twin layer calls the template library and rendering engine to automatically generate a 3D visualized digital twin model of the product.

[0050] Interactive Display: The generated digital twin model is pushed back to the client layer for display. Users can interact with the model through operations such as rotation, scaling, and sectioning, and view related detailed product information, clinical data, and other content.

[0051] Feedback and Optimization: The system collects user interaction data throughout the process, including clicks, dwell time, and reviews. This data is used to continuously optimize the analytical capabilities of the AI ​​model and the accuracy of the product matching algorithm, forming a closed loop of "interaction-feedback-optimization" to continuously improve the system's intelligence level.

[0052] IV. Extended Technical Features and Safeguard Measures The system of this invention also includes the following extended features to enhance its capabilities and compliance: Privacy and Compliance: A federated learning framework can be introduced at the AI ​​engine layer, enabling iterative updates to the model without requiring centralized access to raw user data. Simultaneously, the system supports differential privacy technology, adding controllable noise during data query and analysis to prevent individual data from being used for reverse identification. All system interfaces comply with internationally recognized medical data security and privacy protection standards such as HIPAA and GDPR.

[0053] System Scalability and Integration: The system can establish a unified task scheduling layer and resource management layer to efficiently coordinate tasks and resources across modules. It adopts a distributed big data storage and computing architecture to support the processing of massive amounts of data. Furthermore, the system provides standard API interfaces to support secure integration with third-party enterprise systems, enabling data sharing and business collaboration.

[0054] Example 1 This invention provides an intelligent business card system for pharmaceutical products based on artificial intelligence. For example... Figure 1 As shown, the system includes a client layer, an AI engine layer, a digital twin layer, and a data layer.

[0055] The client layer is used to receive user input information. In one specific embodiment, the client layer may specifically include an H5 mobile application and an official website, used to interact with users and receive their input information, while also interacting with the AI ​​engine layer for data exchange.

[0056] The AI ​​engine layer communicates with the client layer and has a privately deployed large AI model for intelligently recognizing and processing user input information to obtain user needs. Specifically, the AI ​​engine layer may further include an intelligent interaction module, an AI needs recognition module, a product matching module, and a digital twin generation module. The intelligent interaction module is deployed on the H5 page and official website to collect user input information; the AI ​​needs recognition module analyzes user input information and identifies user needs based on the privately deployed large language model; the product matching module matches the identified user needs with a pharmaceutical product database to determine the target pharmaceutical product; and the digital twin generation module generates corresponding 3D visualization model generation instructions based on the target pharmaceutical product.

[0057] The digital twin layer communicates with the AI ​​engine layer and is used to generate corresponding 3D visualization models of pharmaceutical products according to user needs. In one specific embodiment, the digital twin layer includes a pharmaceutical product digital twin generation and display module, a digital twin editing module, and a digital twin rendering module. The pharmaceutical product digital twin generation and display module is used to generate and display a 3D visualization model of the target pharmaceutical product in response to the 3D visualization model generation command. The digital twin editing module allows users to perform secondary editing and customization of the generated digital twin model, such as adjusting product attributes and changing materials. The digital twin rendering module is responsible for outputting the edited digital twin model in different formats, including VR models and videos.

[0058] The data layer is communicatively connected to the AI ​​engine layer and the digital twin layer, providing data support for both. The data layer includes at least a pharmaceutical product database. In a preferred embodiment, the data layer may further include a user interaction database and a multi-source heterogeneous data integration module. The multi-source heterogeneous data integration module is used to standardize data from different channels and store it in the database. This data may include, but is not limited to: biodiversity data, biomedical data, and medical and health data acquired through the Internet of Things (IoT); vital sign data acquired through IoT terminal devices; health data acquired through smart devices; and professional data acquired from an expert database.

[0059] Furthermore, the system can also include a task scheduling layer to manage collaborative tasks between modules and to schedule system resources through a resource management layer. The system can adopt a distributed big data storage and computing architecture, including a data source layer, a data transmission layer, and a data storage layer, to support different types of data storage and transmission methods.

[0060] The system ensures consistency between the digital twin model and product data through a real-time data synchronization mechanism. It supports multi-round information interaction, allowing users to interact with the digital twin model and train and optimize the model based on user feedback. Furthermore, the system provides API interfaces, allowing integration with third-party systems to achieve data sharing and collaborative functions.

[0061] Example 2 This embodiment provides a method for implementing a smart business card for pharmaceutical products based on the above system, specifically including the following steps: Step S1: User Needs Collection. User input information is collected through the client-side layer. This input may include usage scenarios, target symptoms, drug type preferences, etc., and users can input information through various methods such as text input, voice recognition, or image recognition.

[0062] Step S2: AI Intelligent Analysis. The user input information is transmitted to a privately deployed AI large-scale model for intelligent analysis. The AI ​​model can use deep learning algorithms and natural language processing technology to perform semantic analysis on the user input information, identify the user's specific needs, and generate a structured description of those needs.

[0063] Step S3: Product Matching. The structured requirement description is matched against the pharmaceutical product database in the data layer. The most suitable target pharmaceutical product is selected based on the matching algorithm. During the matching process, multiple user requirement factors can be considered, and different weights are assigned to each factor.

[0064] Step S4: Digital Twin Generation. A digital twin engine is invoked to generate a 3D visualization model of the target pharmaceutical product. During the generation process, a geometric model of the product can be constructed first, and then material properties and physical parameters can be assigned to it, giving the model a realistic and real-time appearance. Simultaneously, specific attributes of the product, such as expiration date, specifications, and applicable population, can be customized according to user needs.

[0065] Step S5: Interactive Display. The generated digital twin model of the pharmaceutical product is displayed on the client-side, allowing users to interact with it. Users can interact with the model via touchscreen, voice commands, etc., to view detailed product information, usage instructions, clinical data, etc.

[0066] Step S6: Feedback Optimization. Collect user interaction data with the digital twin model, such as user reviews and usage feedback. Based on this interaction data, optimize the AI ​​big data model and / or product matching algorithm to improve the system's intelligence level and user satisfaction.

[0067] Through the above steps, intelligent and personalized display of pharmaceutical products is achieved, improving user experience while ensuring the security and accuracy of pharmaceutical data.

[0068] Example 3 This embodiment further explains the AI ​​engine layer and its key technical details in the above system.

[0069] The AI ​​demand identification module can incorporate a pre-training mechanism enhanced with pharmaceutical knowledge graphs. Specifically, it constructs domain-specific word vectors by integrating multi-source structured pharmaceutical data, including the Chinese Pharmacopoeia, drug instruction databases, and clinical guidelines. During training, a multi-task learning framework is employed to simultaneously optimize three tasks: intent recognition, entity extraction, and drug relationship reasoning, thereby improving the model's semantic understanding accuracy within a pharmaceutical professional context.

[0070] The product matching module can incorporate a reinforcement learning-driven recommendation strategy. The system dynamically adjusts matching weights based on user interaction behaviors such as dwell time, click hotspots, and query history. Simultaneously, the system can establish an A / B testing platform, allowing for testing of display strategies against different user groups, and automatically optimizing the AI ​​recommendation model based on conversion rate data, forming a closed loop of "interaction-feedback-optimization."

[0071] In terms of digital twin generation, the pharmaceutical product digital twin generation and display module can incorporate a parametric template library and a lightweight real-time rendering engine. The system pre-sets 3D templates for various drug dosage forms, which can automatically adapt to size, material, and color based on product data. Employing WebGL technology and a lightweight mesh compression algorithm, it effectively controls model size while ensuring high-quality visual effects, supporting smooth loading and interaction on mobile devices.

[0072] Regarding data security and privacy protection, the AI ​​engine layer can incorporate a federated learning framework, allowing model iteration without decrypting user data. The system supports differential privacy processing, injecting controllable noise during data querying and analysis to prevent user behavior data from being reverse-engineered. All external interfaces comply with medical data compliance standards such as HIPAA / GDPR.

[0073] In summary, by utilizing the technical solutions described above, a domain-specific AI model deployed privately can intelligently identify and accurately analyze user needs. Furthermore, digital twin technology generates interactive 3D visualized product models, transforming the presentation of pharmaceutical product information from static and planar to dynamic, three-dimensional, and deeply interactive. This effectively enhances the intuitiveness of information delivery and user engagement. Ultimately, while fully ensuring pharmaceutical data security and compliance, it provides medical professionals with a real-time, accurate, and personalized product understanding experience, strongly supporting the professional promotion and decision-making of pharmaceutical products.

[0074] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A smart business card system for pharmaceutical products based on artificial intelligence, characterized in that, include: The client layer is used to receive user input information; The AI ​​engine layer communicates with the client layer. The AI ​​engine layer has a privately deployed large AI model, which is used to intelligently identify and process the user input information in order to obtain user needs. The digital twin layer, which is communicatively connected to the AI ​​engine layer, is used to generate a 3D visualization model of the corresponding pharmaceutical product based on the user's needs. The data layer is communicatively connected to the AI ​​engine layer and the digital twin layer. The data layer includes a pharmaceutical product database to provide data support for the AI ​​engine layer and the digital twin layer.

2. The intelligent business card system for pharmaceutical products based on artificial intelligence according to claim 1, characterized in that, The client layer includes H5 mobile applications and the official website; the AI ​​engine layer includes: The intelligent interaction module, deployed in the client layer, is used to collect the user input information; The AI ​​demand identification module, based on the privately deployed AI large model, is used to analyze the user input information and identify the user demand. The product matching module is used to match the identified user needs with the pharmaceutical product database to determine the target pharmaceutical product. The digital twin generation module is used to generate corresponding three-dimensional visualization model generation instructions based on the target pharmaceutical product.

3. The intelligent business card system for pharmaceutical products based on artificial intelligence according to claim 2, characterized in that, The AI ​​demand recognition module introduces a pre-training mechanism enhanced by medical knowledge graphs. It constructs domain-specific word vectors by integrating multi-source medical structured data and adopts a multi-task learning framework to optimize intent recognition, entity extraction, and drug relationship reasoning tasks. The digital twin layer includes a pharmaceutical product digital twin generation and display module, a digital twin editing module, and a digital twin rendering module. The pharmaceutical product digital twin generation and display module has pre-set three-dimensional templates for various drug dosage forms and automatically adapts the size, material, and color according to the product data. It uses WebGL and a lightweight mesh compression algorithm to generate the three-dimensional visualization model.

4. The intelligent business card system for pharmaceutical products based on artificial intelligence according to claim 1, characterized in that, The data layer also includes a user interaction database and a multi-source heterogeneous data integration module; the multi-source heterogeneous data integration module is used to standardize data from different channels and then store it in the data layer.

5. The intelligent business card system for pharmaceutical products based on artificial intelligence according to claim 2, characterized in that, The AI ​​engine layer incorporates a federated learning framework and / or supports differential privacy processing, and the system's external interfaces comply with preset medical data compliance standards.

6. The intelligent business card system for pharmaceutical products based on artificial intelligence according to claim 2, characterized in that, The product matching module introduces a reinforcement learning-driven recommendation strategy, which dynamically adjusts the matching weights based on user interaction behavior; the system also has an A / B testing platform for optimizing the AI ​​recommendation model based on conversion rate data.

7. A method for implementing a smart business card for pharmaceutical products based on artificial intelligence, characterized in that, The method, applied to the system as described in any one of claims 1 to 6, comprises: User input information is collected through the client-side layer; The AI ​​large model deployed in a private location is used to intelligently analyze the user input information, identify user needs, and generate a structured description of the needs. The structured requirements description is matched with a pharmaceutical product database to filter out target pharmaceutical products; The digital twin engine is invoked to generate a three-dimensional visualization model of the target pharmaceutical product; The 3D visualization model is interactively displayed at the client layer.

8. The method for implementing a smart business card for pharmaceutical products based on artificial intelligence according to claim 7, characterized in that, The intelligent analysis of user input information includes: based on a pre-trained model enhanced with a medical knowledge graph, a multi-task learning framework is used for intent recognition, entity extraction, and drug relationship reasoning; The process of selecting target pharmaceutical products includes: introducing a reinforcement learning-driven recommendation strategy and dynamically adjusting matching weights based on user interaction behavior; The process of generating a 3D visualization model includes: calling a pre-set 3D template of a drug dosage form, automatically adapting the size, material, and color according to the data of the target pharmaceutical product, and rendering it using WebGL and a lightweight mesh compression algorithm.

9. The method for implementing a smart business card for pharmaceutical products based on artificial intelligence according to claim 7, characterized in that, The method further includes: Collect user interaction data with the 3D visualization model; Based on the interaction data, optimize the AI ​​big model and / or product matching algorithm.

10. The method for implementing a smart business card for pharmaceutical products based on artificial intelligence according to claim 7, characterized in that, In the intelligent analysis process, a federated learning framework is used for model iteration and / or differential privacy processing is used to inject controllable noise into the data.

Images