Researchers have introduced MedKGent, a large language model agent framework that constructs a temporally evolving medical knowledge graph from over 10 million PubMed abstracts. The resulting graph contains 156,275 entities and 2,971,384 triples, achieving a 90% validity rate and improving retrieval-augmented generation across seven benchmarks. Concurrently, platforms like Medvolt's MedGraph EDGE leverage structured biomedical knowledge graphs to support clinical decision systems, target discovery, and translational medicine.
Story comments
Loading comments…