Intelligent system for foreign students preparation for postgraduate exams in surgery

Abstract

Objective . To develop an Intelligent System (IS) for foreign students preparation for surgery exams to confirm diplomas obtained in foreign medical universities. Materials and methods . Development of the system consisted of two stages: 1) creation of a Decentralized Knowledge Graph (DKG); 2) integration of DKG with the technology of Large Language Models (LLM), digital twins and a chat-bot. The materials for the developed IS were structured data: textbooks on surgery, which are used for preparation of postgraduate examinations in India (FMGE), educational materials of the Department of Surgical Diseases No. 3 of the Gomel State Medical University, web pages https://cyb.ai/@gsmu-by/brain. Results . The created IS is deployed on a local server/computer and includes the DeepSeek LLM, a chatbot on the AnythingLLM platform and Telegram; it works on the RAG technology (Retrieval-Augmented Generation), which combines information search and text generation based on the found data. The IS was tested during surgery classes with foreign fourth year students of the Gomel State Medical University. Conclusion . The developed information system for preparation for postgraduate examinations in surgery allows better personalization of training and can improve the efficiency of preparing students for postgraduate examinations in their home countries.

Ähnliche Arbeiten