Search

CN-121983257-A - Artificial intelligence teaching system for guided learning and application thereof

CN121983257ACN 121983257 ACN121983257 ACN 121983257ACN-121983257-A

Abstract

The invention relates to an artificial intelligence teaching system for guided learning and application thereof, which can provide an integrated case-centered intelligent collaborative learning platform, can build complete clinical situations for learners, help the learners to establish direct connection between morphological characteristics and disease diagnosis, thereby clinically integrating thinking capability in a culture period, and simultaneously, enable the learners to be converted into active and collaborative problem solvers from passive observers. In addition, the system aims to solve the problem of lack of real-time professional guidance in collaborative learning, and creatively introduces an AI intelligent guiding module, so that the interactive behavior of a learning group is analyzed in real time, and prompts, problems and feedback highly related to the situation are provided, so that students are actively guided to focus on key features, potential error cognition is corrected, and the immersive and efficient school effect of simulating the side guidance of experts is achieved.

Inventors

  • CHEN HAISHENG
  • Feng Menle

Assignees

  • 佛山大学

Dates

Publication Date
20260505
Application Date
20251231

Claims (10)

  1. 1. The artificial intelligence teaching system for guided learning is characterized by comprising a database module, a case learning module, a collaborative work space module, an AI intelligent guiding module and a user interface, wherein the database module is used for storing, managing and retrieving image files, the case learning module is used for associating the image files with real data to form a teaching case, the collaborative work space module is used for allowing a plurality of system users to enter the same virtual room, carrying out synchronous, shared and collaborative analysis on the same image files in the teaching case, outputting an interactive data packet to the AI intelligent guiding module, receiving and displaying real-time feedback of the AI intelligent guiding module, the AI intelligent guiding module is used for analyzing the interactive data packet output by the collaborative work space module in real time to generate and feed back guiding information, and the user interface is used for inputting instructions of the system users and displaying the guiding information.
  2. 2. The artificial intelligence teaching system according to claim 1, wherein the database module comprises a metadata database and a file server, and the relational mapping mechanism of the database module comprises storing the image file in the file server, generating an identifier, and storing descriptive information, the identifier, and a storage path of the image file in a metadata record, wherein the metadata record is stored in the metadata database.
  3. 3. The artificial intelligence teaching system of claim 1, wherein the image file is associated with the real data by a many-to-many relationship.
  4. 4. The artificial intelligence teaching system according to claim 1, wherein the synchronization, sharing and collaboration analysis comprises a synchronization view, a sharing annotation and an integrated chat, wherein the synchronization view comprises that operation parameters of a view of an image file by a system user are broadcast to the rest system users in the same virtual room in real time through a WebSocket protocol, and application program logic of a client is triggered to update the view so as to synchronize the view; The sharing annotation comprises the steps that operation parameters of annotation of an image file by a system user are broadcast to the rest system users in the same virtual room in real time through a WebSocket protocol, so that the annotation is synchronously displayed; The integrated chat includes a discussion by system users in the same virtual room through a WebSocket protocol based instant-time chat box.
  5. 5. The artificial intelligence teaching system according to claim 1, wherein the outputting the interaction data packet to the AI intelligent guiding module comprises generating the interaction data packet when the system user views or marks the image file, sending the interaction data packet to the application server through a WebSocket protocol, and forwarding the interaction data packet to the AI intelligent guiding module for real-time analysis; the real-time feedback of the receiving and displaying AI intelligent guiding module comprises the steps that the AI intelligent guiding module carries out real-time analysis, guiding information is generated, and the guiding information is fed back to the collaborative work space module through a WebSocket protocol by an application server to display.
  6. 6. The artificial intelligence teaching system according to claim 1, wherein the AI intelligent guidance module comprises a multitasking deep learning visual analysis model, a parameter calculation algorithm, a dialog generation model; The multi-task deep learning visual analysis model comprises a convolutional neural network and a plurality of parallel solution wharfs, wherein the convolutional neural network is used for extracting features and outputting the features to the decoding head, the decoding head is used for generating a segmentation mask based on the features, and the segmentation mask is used for extracting target parameters through a parameter calculation algorithm and transmitting the target parameters to the dialogue generation model; The dialog generation model comprises a unified language model interface layer, wherein the unified language model interface layer is used for calling dialog generation capability to generate guide information based on the target parameters, and the unified language model interface layer is used for decoupling an AI intelligent guide logic of an upper layer and a language model of a bottom layer.
  7. 7. The artificial intelligence teaching system of claim 6, wherein the AI intelligent guidance logic comprises a group attention detection and critical area deviation analysis algorithm, the execution of the group attention detection and critical area deviation analysis algorithm comprising the steps of predefining critical diagnostic areas, group attention real-time aggregation, deviation analysis and triggering algorithms, generating guidance information; When the teaching cases are stored in the metadata base, at least 1 area containing key features is marked out as a key diagnosis area of the teaching cases and is stored in the metadata base; The group attention real-time aggregation comprises the steps of receiving the interaction data packet, generating a dynamic two-dimensional array, simulating to form an attention thermodynamic diagram, and quantifying the observed time length and frequency of a key diagnosis area; Calculating the accumulated attention degree obtained by each key diagnosis area, comparing the accumulated attention degree with the total attention degree of the whole image file in the teaching case to obtain the attention degree duty ratio of the key diagnosis area, and triggering an internal instruction aiming at the key diagnosis area according to a triggering rule; the generating the guidance information includes generating guidance information according to the internal instruction.
  8. 8. A morphological teaching and/or learning method, characterized in that the artificial intelligence teaching system according to any of claims 1-7 is used for teaching and/or learning, the real data comprising real world research data.
  9. 9. A computer readable storage medium, characterized in that the storage medium stores a computer program which, when executed by the processor, implements the method of claim 8.
  10. 10. A computer device comprising a memory storing a computer program and a processor implementing the method of claim 8 when the processor executes the computer program.

Description

Artificial intelligence teaching system for guided learning and application thereof Technical Field The invention relates to the technical field of medical education, in particular to an artificial intelligence teaching system for guided learning and application thereof. Background Morphology is a cornerstone of medical disciplines such as pathology, histology, and hematology, and requires a learner to have a high level of image recognition and diagnostic capabilities. Traditional teaching methods rely primarily on optical microscopy and glass sectioning. The learner must observe limited specimens in a fixed laboratory for a limited period of time, which is difficult to meet the demands of modern medical education for efficient and flexible learning. With the development of information technology, digital teaching has become a trend. With the development of digital microscopic imaging technology, a typical lesion area and key morphological features in a physical slice specimen can be photographed as high-resolution digital images by equipping a conventional optical microscope with a high-resolution digital camera. These carefully selected digital images are collected together, which results in a large number of digital morphology databases and online teaching maps. The platforms allow students to access a large number of case images anytime and anywhere, and preliminarily solve the problem of resource scarcity. The most similar schemes at present can be divided into three categories: 1. virtual microscope/digital map platform the core function of such a platform is to store and display a large number of digital slice images. The user can zoom, pan, etc. the digital image as the real microscope is operated. Platforms are typically provided with simple graphic annotations that illustrate key morphological features. Such platforms solve the "look at" problem, but are essentially unidirectional, personal learning patterns. 2. Generic collaboration/gamification teaching platform-for example, utilizing existing commercial gamification platform Kahoot | to assist in histological teaching. The platform stimulates the learning interest and participation of students in the forms of questions and answers, contests and the like, and supports team cooperation. However, such platforms are versatile tools that do not themselves contain specialized, structured medical case content. The teacher needs to prepare morphological pictures and problems by himself and import the morphological pictures and problems into the universal platforms, so that the content and the collaboration function are separated, and deep collaboration for the digital slice itself cannot be realized. Summarizing, the existing teaching system has the following defects that the learning mode is single passively, the existing digital map platform mainly solves the problem of accessibility of images, but the learning mode still stays in a passive mode of browsing-memorizing, effective guidance and interaction are lacking, and clinical diagnosis reasoning capability of students is difficult to exercise. Content and collaboration functionality are disjoint-although the general collaboration platform can promote interaction, its functionality does not match the requirements of professional morphological observations. Students cannot make real-time, synchronous labeling and discussion on a platform for a specific region of a digital slice image, and the depth and efficiency of collaboration are greatly limited. The lack of complete clinical situations is that most platforms only provide isolated morphological images, but lack complete clinical case information such as medical history, laboratory examination and the like matched with the platforms, and the comprehensive ability of combining morphological discovery with clinical practice of students is not facilitated. The existing AI teacher system is usually in a single mode, and AI interacts with a single student, so that collaborative dynamics among multiple users cannot be understood and intervened. Otherwise, the collaboration platform is completely free of AI participation. This results in the student not being able to obtain the benefits of companion collaboration and real-time guidance of expert AI in one environment at the same time, the learning experience being split. One-way and passive limitations of synchronization mechanisms in addition, existing multi-screen synchronization techniques (such as operation synchronization in teleconferencing) essentially broadcast and reproduce the presenter's operating instructions, which system itself does not have any ability to analyze and understand user behavior, but is simply a passive "command microphone". The mechanism can be used for remote demonstration, but can not intervene or guide the learning process in the teaching scene, and can not identify whether the learning group deviates from the correct diagnosis thought, so that the application of the mech