▶
Spell Correction & Abbreviation Recognition
Clinicians type fast and imprecisely — the Terminology Server catches typos like "hyper tension" and surfaces a suggested correction before executing the search, while abbreviation handling maps shorthand like "T2DM" or "HTN" directly to their full medical concepts and associated ICD codes without any manual lookup. This video demos both features live, then shows how to stack them with domain and vocabulary filters for tightly scoped, accurate results even with messy input.
📚 Documentation: https://nlp.johnsnowlabs.com/docs/en/terminology_server/term_server
Install
on AWS: https://aws.amazon.com/marketplace/pp/prodview-3hta3hebivvrk
on Azure: https://marketplace.microsoft.com/en-us/product/johnsnowlabsinc1646051154808.medical_terminology_server?tab=Overview
🔌 MCP Server (Agent Integration): https://nlp.johnsnowlabs.com/docs/en/terminology_server/features/mcp_server#mcp-server---agent-conn
📓 Code Sample Notebook:
https://github.com/JohnSnowLabs/spark-nlp-workshop/blob/master/products/term_server/terminology_mcp_strands_demo.ipynb
