Enhancing Medical Data Processing With Neo4j
summary
Neo4j graph DB indexes ATC hierarchies and MeSH to turn 350,000-term taxonomies and deep disease trees into fast traversable graphs, enabling linear traversal across five ATC levels and child/sibling expansion for MeSH nodes. The production system links >2 million regulatory documents from ~50 sources, maps ~250 approved drugs, and resolves disease synonyms (up to ~2,500 for broad terms) so searches that once took teams weeks now return results in seconds with low latency. Neo4j graph DB 350,000 MeSH terms 2 million documents 2,000 nodes per disease
AI-written summary · about
chapters (7)
- 0:00Introduction to Vivpro
- 0:32The role of RIA
- 2:23Importance of structured data
- 2:55ATC and MeSH classification
- 9:24Modeling ATC with Neo4j
- 12:37Modeling MeSH with Neo4j
- 13:46Conclusion
related talks
15mKnowledge Graph of Drugs Data for Swiss Healthcare System
24mUnleashing the power of FHIR data with knowledge graphs
16mMedQGraph: TMKGs for AI-Driven Healthcare Insights
30mPerformant GraphRAG Improving Neo4j to Drive Autonomous AI ...
01h 00mGraph Mastery to AI
29mSo... Graph Databases
26mKnowledge Combined Common Data Model: A Graph-Based Integration of Reactome and OMOP CDM
30mInterconnected Capital: Visualizing WAM, PE, RE, and HF Data with Neo4j