This n8n workflow automates the process of extracting, processing, and vectorizing medical procedure data for semantic search applications. It starts by retrieving medical procedures from an Oracle database, then formats and splits the text into manageable chunks. Using Google’s Gemini API, it generates embeddings (vector representations) for each medical procedure, which are then stored in a PostgreSQL database equipped with PGVector for efficient semantic querying. The workflow includes detailed notes to guide configuration and ensure proper data handling, making it ideal for healthcare data management and advanced search solutions.
Medical Data Vectorization for Semantic Search
| Node Count | 11 – 20 Nodes |
|---|---|
| Nodes Used | @n8n/n8n-nodes-langchain.documentDefaultDataLoader, @n8n/n8n-nodes-langchain.embeddingsGoogleGemini, @n8n/n8n-nodes-langchain.textSplitterTokenSplitter, @n8n/n8n-nodes-langchain.vectorStorePGVector, code, manualTrigger, n8n-nodes-oracle-database-parameterization.Oracle Database with Parameterization, splitInBatches, stickyNote |
Be the first to review “Medical Data Vectorization for Semantic Search”
Reviews
There are no reviews yet.