Linking Knowledge to Care: Knowledge Graph-Augmented Medical Follow-Up Question Generation

Researchers developed KG-Followup, a knowledge graph-augmented large language model system that generates medical follow-up questions for pre-diagnostic assessment. The system combines structured medical domain knowledge with LLMs to improve clinical diagnosis efficiency, outperforming existing methods by 5-8% in recall benchmarks.

• →KG-Followup integrates knowledge graphs with LLMs to generate relevant medical follow-up questions for clinical diagnosis.
• →The system addresses LLMs' limited domain knowledge by incorporating structured medical expertise.
• →Experimental results show 5-8% improvement over state-of-the-art methods in recall metrics.
• →The technology aims to reduce time-consuming patient-doctor interactions during pre-diagnostic assessment.
• →Active in-context learning enhances the model's ability to reason with medical domain knowledge.