[Aspell-user] RE: using aspell to identify medical terms

[Denny, Joshua Charles]

I have a medical search engine project at Vanderbilt.  We use the vocabulary/ontology by the NIH Unified Medical Language System (UMLS).  We use the UMLS's Specialist Lexicon as a list of terms (200k terms, most of them medical, but it also includes general English use terms) and UMLS Metathesaurus (a list of 800k unique single and multiword concepts; when broken down into words I think there are about 600k unique words (much of them genes, chemical names, etc).  see http://www.nlm.nih.gov/research/umls/.  Licenses can be had free for research purposes.

 

Our project has a research page that you may find helpful.  http://knowledgemap.mc.vanderbilt.edu/research.  The methodology of identifying medical concepts to the UMLS before searching may be a bit complex for your task, not sure.

 

Josh