Skip to content
forked from OHDSI/NLPTools

[in development] Tools to support Natural Language Processing of freetext to create structured data elements for analysis

License

Notifications You must be signed in to change notification settings

asivura/NLPTools

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

9 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

ohdsi-nlp-wrapper

This repository contains three separate eclipse projects: clamp-wrapper, ctakes-wrapper, metamap-lite-wrapper; Users can import them into eclipse as existing projects.

download dependencies

To run the wrapper, users must download the dependent libraries and resources and then place them in the subfolder. For example, This is a directory of ctakes wrapper:

ctakes-wrapper

--bin

--input

--output

--lib

----ctakes

------config

------desc

------lib

------resources

--resources

----org

--src

cTAKES wrapper setup

  • Download the latest cTAKES binary file: http://apache.osuosl.org//ctakes/ctakes-4.0.0/apache-ctakes-4.0.0-bin.zip
  • Unzip that file
  • Run mkdir -p ctakes-wrapper/lib/ctakes
  • Copy lib directory into classpath: cp -r apache-ctakes-4.0.0/lib <ctakes-wrapper-location>/lib/ctakes/
  • Copy resource files into classpath: cp -r apache-ctakes-4.0.0/resources ctakes-wrapper/lib/ctakes/
  • Download the official cTAKES UMLS resources: https://downloads.sourceforge.net/project/ctakesresources/sno_rx_16ab.zip
  • Run mkdir -p ctakes-wrapper/resources/org/apache/ctakes/dictionary/lookup/fast
  • Unzip that file into the resources directory: unzip sno_rx_16ab.zip -d <ctakes-wrapper-location>/resources/org/apache/ctakes/dictionary/lookup/fast/
  • Setup your UMLS credentials in this file: <ctakes-wrapper-location>/lib/ctakes/resources/org/apache/ctakes/dictionary/lookup/fast/sno_rx_16ab.xml
    • There are two places where you need to replace CHANGE_ME with your UMLS user name and your UMLS password.
  • Now you should be able to run the class in the ctakes-wrapper project. This version of cTAKES makes use of large machine learning models and should be run with at least 2GB RAM. To do this in Eclipse, in the run configuration dialog, under the Arguments tab, add "-mx2g" to the VM Arguments text field.

Here are download links

clamp wrapper resources: https://drive.google.com/open?id=1uzvWZPebq3DYZNPFNm2Ise354IWuyADo

metamap lite wrapper resources: https://drive.google.com/open?id=1sjUad3KYbBtI1iFz3iw_LW4fdtu4q1UZ

For CLAMP wrapper, write them in the java code:

ClampWrapper clamp = new ClampWrapper( "resources/clamp/clamp-ner-attribute.pipeline",

"YOUR_UMLS_USER_NAME",

"YOUR_UMLS_USER_PASSWORD",

"resources/clamp/umls_index/",

"resources/clamp/rxnorm_index/" );

set CLAMP SN

For CLAMP wrapper, users must obtain an SN code from CLAMP website and input their email and SN for the first time when they start CLAMP wrapper.

Please visit this page and submit an request for Clamp Command line version 1.4.0.

https://clamp.uth.edu/get-clamp.php

About

[in development] Tools to support Natural Language Processing of freetext to create structured data elements for analysis

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Java 100.0%