New Approach of Artificial Intelligence for Fertility Treatment
summary
Benchmarks Tanit.AI against GPT-3.5-Turbo and GPT-4 using a knowledge graph built from 400 documents (medical guidelines and publications) with medical ontologies, NER and table extraction, embeddings and vector indexes; evaluation uses 180 yes/no questions from a reproductive medicine master’s program; Tanit.AI (GraphRAG architecture) records 75.42% accuracy and 82.61% precision while returning source-linked, explainable answers; moves from pure vector RAG to ontology-linked graphs (Neo4j tools mentioned) and highlights evaluation gaps, privacy concerns, and prototype-scale limits.
AI-written summary · about
chapters (7)
- 0:00Introduction to the problem
- 1:20Generative AI in healthcare
- 2:14Transition to graph RAG
- 3:38Building the knowledge graph
- 4:32Medical logic and ontologies
- 7:08Evaluation and results
- 9:00Future challenges and closing
related talks
01h 28mBuilding a Simple RAG Application: From Fundamentals to Implementation
17mNatural Language Processing in Automation
21mData-Driven NLP: From Statistics to GenAI
21mHow RAG is Redefining AI Capabilities
16mMedQGraph: TMKGs for AI-Driven Healthcare Insights
21mGen AI in Health: What's New in 2026?
44mBuilding Knowledgeable Machines
19mLet's Talk GenAI