T
his project has constructed a pathology knowledge graph covering 20 common diseases, with data sourced from the relevant clinical guidelines published by the National Health Commission of China. By utilizing graphRAG and large language models, we have structured the diagnostic processes, symptom presentations, examination methods, treatment plans, and more for these diseases. We performed entity recognition and relationship extraction on the textual data, automatically constructing nodes and edges within the knowledge graph. The result is a knowledge graph designed for use in medical research, clinical decision support, and disease management systems, helping researchers explore potential disease correlations and develop innovative treatment methods. By integrating large language models, this system also provides doctors with diagnostic insights, treatment suggestions, and personalized medical plans, improving both diagnostic efficiency and accuracy.
