The LightRAG and AI Agent architecture

 

Category: *Precision Health & Smart Medical

Exhibitor: NATIONAL SUN YAT-SEN UNIVERSITY

Booth No: N526

 

Characteristic

 

This system proposes an intelligent medical assistant based on the LightRAG and AI Agent architecture, integrated with LINE Bot as the user interaction interface. Patients can use the LINE Bot to ask questions related to diseases, medications, and medical guidelines, while the system provides responses through Large Language Models (LLMs) and Retrieval-Augmented Generation (RAG) technology. To address the issue of version conflicts in traditional RAG systems caused by outdated medical documents, the proposed system incorporates Metadata-based version control and a Graph-based Retrieval architecture, combining knowledge graphs with vector search to enhance cross-document medical knowledge retrieval. In addition, the system supports contextual memory, multi-turn conversations, RAG knowledge base version management, Prompt Template management, and model management. The backend also allows administrators to upload documents and build new medical knowledge bases, thereby improving the accuracy and practicality of medical question answering.

 

Other Products

 

• 

  Large-area Afterglow Plasma for Wound Healing Applications

• 

  KTU Microbial and Probiotic Identification Solution

• 

  Rapid screening method for functional microalgae strains in aquaculture

 

 

 

Back