Hi Erica, You are correct: ctakes normalizes to CUI. If you fetch Snomed codes then you will get all snomed codes that match the CUI, which may be > 1/1.
The reason for this is that those snomed codes are synonymous (or not truly unique) as determined by somebody who contributes to the umls. This does however mean that unique texts have been deemed equal, and they may not be for your purposes. I'm sure that somebody can come up with a cute way to break things back down to unique snomeds - whether that is a post-process that does string matching when there are 2 or more possible snomeds or some other method. Sean -----Original Message----- From: Scotty, Erica J [mailto:scotty.er...@mcrf.mfldclin.edu] Sent: Tuesday, December 20, 2016 12:49 PM To: dev@ctakes.apache.org Subject: CUI vs SNOMED Hi all, I am having trouble tracking down if cTAKES tags the CUI or the SNOMED code directly. From what I am looking at, it looks like the CUI is tagged then the output contains all SNOMED codes that map to that CUI. Is that correct? If so, is there a way to map to the SNOMED code directly? Thanks, Erica ______________________________________________________________________ The contents of this message may contain private, protected and/or privileged information. If you received this message in error, you should destroy the e-mail message and any attachments or copies, and you are prohibited from retaining, distributing, disclosing or using any information contained within. Please contact the sender and advise of the erroneous delivery by return e-mail or telephone. Thank you for your cooperation.