You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@ctakes.apache.org by Prakhar Gaur <pr...@infosys.com> on 2017/12/27 07:13:47 UTC

RE: cTAKES UMLS Using Custom Dictionary

Hello,

I created a custom Dictionary using Dictionary creator GUI tool and the XML file for it is created in the folder org/apache/ctakes/dictionary/lookup/fast/foobar.xml as mentioned by Mike.

Post this I make required change in  UmlsLookupAnnotator.xml file for the “DictionaryDescriptor” attribute.

But on running the piper file with Piper-runner tool, the default sno_rx_ab17 file is used.

If I use the ‘-l’ option followed with custom dictionary xml path with piper-runner tool, then the custom dictionary is used for UMLS lookup.

Can someone please help me in getting the solution Mike suggested ie. modifying “DictionaryDescriptor” attribute, to work ?

Regards,

From: Desteny Child [mailto:myshareit@gmail.com]
Sent: Friday, December 15, 2017 2:49 PM
To: Prakhar Gaur <pr...@infosys.com>
Cc: user@ctakes.apache.org
Subject: Re: cTAKES UMLS ICD10 codes lookup

Hi,
For example, if you use AggregatePlaintextUMLSProcessor.xml  you can update apache-ctakes-4.0.0\desc\ctakes-dictionary-lookup-fast\desc\analysis_engine\UmlsLookupAnnotator.xml  and provide there a path to you database for DictionaryDescriptor, in my case this is org/apache/ctakes/dictionary/lookup/fast/icd10.xml
Best regards,
Mike

On Fri, Dec 15, 2017 at 11:08 AM, Prakhar Gaur <pr...@infosys.com>> wrote:
Hello Mike,

How/Where do you specify what Dictionary that you want to use with CVD with a particular AE ?

Regards,

From: Desteny Child [mailto:myshareit@gmail.com<ma...@gmail.com>]
Sent: Friday, December 15, 2017 2:34 PM
To: user@ctakes.apache.org<ma...@ctakes.apache.org>
Subject: cTAKES UMLS ICD10 codes lookup

Hello,

I created a cTAKES custom dictionary from UMLS database with ICD10 codes.

Right now I able to analyze the text by for example disease name, like "Asthma" and annotation index will contain the ICD10 code for this match "code = "J45.90".


Is it possible to configure cTAKES in order to reverse this process in order to look for ICD10 code appearance in the text instead?

Thanks,
Mike