ctakes-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From vijay garla <vnga...@gmail.com>
Subject Re: cTAKES Information Extraction
Date Tue, 20 May 2014 13:04:26 GMT
YTEX has recently been added to ctakes (trunk); it will be included in the
coming ctakes 3.1.2 release.  For more information, refer to
https://cwiki.apache.org/confluence/display/CTAKES/YTEX+3.2



On Tue, May 20, 2014 at 8:37 AM, Anirban Chakraborti <
chakraborti.anirban@googlemail.com> wrote:

> Vijay,
>
> Pardon my ignorance.
>
> Is Ytex an add on module for Ctakes. How do I add it.
> whats the additional functionality provided by Ytex.
>
>
> On Tue, May 20, 2014 at 6:02 PM, vijay garla <vngarla@gmail.com> wrote:
>
>> YTEX writes uima annotations and the CAS to a database (MySQL/SQL
>> Server/Oracle); see:
>> https://cwiki.apache.org/confluence/display/CTAKES/cTAKES+3.1.2+YTEX+DBConsumer
>>
>>
>>
>> On Mon, May 19, 2014 at 4:25 PM, Masanz, James J. <Masanz.James@mayo.edu>wrote:
>>
>>>  You could try looking at
>>>
>>>
>>>
>>> org.spin.scrubber.uima.consumer.JDBCCasConsumer
>>>
>>>
>>>
>>> That’s in sandbox, not in trunk:
>>>
>>>
>>> https://svn.apache.org/repos/asf/ctakes/sandbox/ctakes-scrubber-deid/src/main/java/org/spin/scrubber/uima/consumer/JDBCCasConsumer.java
>>>
>>>
>>>
>>>
>>>
>>> There’s also something in the ctakes-gui project, which is also under
>>> sandbox. Pei would be the best person to give you pointers to basing off
>>> that.
>>>
>>>
>>>
>>> -- James
>>>
>>>
>>>
>>> *From:* Anirban Chakraborti [mailto:chakraborti.anirban@googlemail.com]
>>> *Sent:* Saturday, May 17, 2014 11:48 AM
>>> *To:* user@ctakes.apache.org
>>> *Subject:* RE: cTAKES Information Extraction
>>>
>>>
>>>
>>> James,
>>> Is there any code snippet for writing to a database from annotated cas
>>>
>>> Pls point the same out. The one snippet I got had known type info but
>>> what if the type is not known in advance.
>>>
>>> Anirban
>>>
>>> On 17 May 2014 22:08, "Masanz, James J." <Masanz.James@mayo.edu> wrote:
>>>
>>> This was fixed in trunk under
>>>
>>> https://issues.apache.org/jira/browse/CTAKES-288
>>>
>>> and will be available in the next release
>>>
>>>
>>>
>>> (I was able to view the image, thanks for including your link)
>>>
>>>
>>>
>>> Code that changed is here:
>>>
>>>
>>>
>>>
>>> https://svn.apache.org/viewvc/ctakes/trunk/ctakes-template-filler/src/main/java/org/apache/ctakes/template/filler/ae/TemplateFillerAnnotator.java?view=markup&pathrev=1579707
>>>
>>>
>>>
>>> *From:* Tamzidur Rahman [mailto:u5184041@anu.edu.au]
>>> *Sent:* Thursday, May 15, 2014 6:36 AM
>>> *To:* user@ctakes.apache.org
>>> *Subject:* RE: cTAKES Information Extraction
>>>
>>>
>>>
>>> Sorry about the late response. The sentence that I am using is:
>>>
>>>
>>>
>>>  "patient took 50mg of aspirin for severe pain in right knee."
>>>
>>>
>>>
>>> I have also used the template filler as well for this sentence and I can
>>> see it identify the locationOf and degreeOf in the CVD. However, I don't
>>> see it assign the values inside the identified annotation for locationOf
>>> and degreeOf like the ctakes keynote suggests it should.
>>>
>>>
>>>
>>> Here are my results for template filler using the sentence above:
>>>
>>>
>>>
>>> http://i.imgur.com/dvq958l.png
>>>
>>>
>>>
>>> I'm not exactly sure if you guys can see these links, so if you guys
>>> can't let me know.
>>>
>>>
>>>
>>> Also If I had to manually extract and place these relations into the
>>> identified annotations is there someplace I can be led to? As it is, I
>>> don't know where to start looking for the functions that extract these
>>> values (degreeOF and locationOf).
>>>
>>>
>>>
>>>
>>>   ------------------------------
>>>
>>> *From:* Masanz, James J. <Masanz.James@mayo.edu>
>>> *Sent:* Wednesday, 14 May 2014 5:47 AM
>>> *To:* 'user@ctakes.apache.org'
>>> *Subject:* RE: cTAKES Information Extraction
>>>
>>>
>>>
>>> Do you have an example that you can share of the sentences you were
>>> expecting to find locationOf and degreeOf.
>>>
>>>
>>>
>>> Note:
>>>
>>> The relation extractor creates a relation between two
>>> IdentifiedAnnotation, but does not assign values to the locationOf or
>>> degreeOf in any IdentifiedAnnotation.
>>>
>>> The template filler assigns values to the locationOf or degreeOf based
>>> on what the relation extractor finds.
>>>
>>>
>>>
>>> The AggregateTemplateFiller AE contains the relation extractor and the
>>> template filler.
>>>
>>>
>>>
>>>
>>>
>>> *From:* Tamzidur Rahman [mailto:u5184041@anu.edu.au<u5184041@anu.edu.au>]
>>>
>>> *Sent:* Sunday, May 11, 2014 9:16 AM
>>> *To:* user@ctakes.apache.org
>>> *Subject:* RE: cTAKES Information Extraction
>>>
>>>
>>>
>>> Thanks for the clarification Timothy. We could live without some
>>> features, however somethings such as date, dose, even locationOf are
>>> required for our project.
>>>
>>>
>>>
>>> As of right now locationOf or degreeOf are not related to Identified
>>> Annotations by the relation extractor, which was what I expected as results
>>> from this key note.
>>>
>>>
>>>
>>> https://uima.apache.org/downloads/gscl2013/keynote_ctakes.pdf page 68
>>>
>>>
>>>
>>> a smaller link:
>>>
>>>
>>>
>>> http://i.imgur.com/Gs3lYyS.png
>>>
>>>
>>>
>>>
>>>   ------------------------------
>>>
>>> *From:* Anirban Chakraborti <chakraborti.anirban@googlemail.com>
>>> *Sent:* Monday, 12 May 2014 12:00 AM
>>> *To:* user@ctakes.apache.org
>>> *Subject:* Re: cTAKES Information Extraction
>>>
>>>
>>>
>>> that's the desired output. What were you anticipating.
>>>
>>>
>>>
>>> On Sun, May 11, 2014 at 6:57 PM, Tamzidur Rahman <u5184041@anu.edu.au>
>>> wrote:
>>>
>>> I have used
>>> ctakes-clinical-pipeline/desc/analysis_engine/AggregatePlaintextProcessor.xml
>>> for this.
>>>
>>>
>>>
>>> Without UMLS:
>>>
>>> http://i.imgur.com/TWg1kzG.png
>>>
>>>
>>>
>>> With UMLS:
>>>
>>> http://i.imgur.com/btZdNtS.png
>>>
>>>
>>>
>>> I am still having the same issue.
>>>
>>>
>>>   ------------------------------
>>>
>>> *From:* Anirban Chakraborti <chakraborti.anirban@googlemail.com>
>>> *Sent:* Sunday, 11 May 2014 11:02 PM
>>>
>>>
>>> *To:* user@ctakes.apache.org
>>> *Subject:* Re: cTAKES Information Extraction
>>>
>>>
>>>
>>> Startwith AggregatePlainText.xml , you should get annotations on target
>>> text
>>>
>>>
>>>
>>> On Sun, May 11, 2014 at 6:30 PM, Tamzidur Rahman <u5184041@anu.edu.au>
>>> wrote:
>>>
>>> Thanks for the reply,
>>>
>>>
>>>
>>> I haven't created any custom AEs, I've only used the default AE found in
>>> the vanilla install.
>>>
>>>
>>>
>>> The AEs I've used:
>>>
>>>
>>>
>>>
>>> ctakes-clinical-pipeline/desc/analysis_engine/AggregatePlaintextUMLSProcessor.xml
>>>
>>>
>>> ctakes-relation-extractor/desc/analysis_engine/RelationExtractorAggregate
>>>
>>>
>>>
>>> ctakes-template-filler/desc/analysis_engine/AggregateTemplateFiller
>>>
>>>
>>>
>>>
>>>   ------------------------------
>>>
>>> *From:* Anirban Chakraborti <chakraborti.anirban@googlemail.com>
>>> *Sent:* Sunday, 11 May 2014 10:51 PM
>>> *To:* user@ctakes.apache.org
>>> *Subject:* Re: cTAKES Information Extraction
>>>
>>>
>>>
>>> Tam,
>>>
>>>
>>>
>>> What AE engines are you using. There might be an issue there.
>>>
>>>
>>>
>>> On Sun, May 11, 2014 at 2:38 PM, Tamzidur Rahman <u5184041@anu.edu.au>
>>> wrote:
>>>
>>> Hi,
>>>
>>>
>>>
>>> I've been working on a project which uses cTAKES annotations for a
>>> search algorithm in a database. What we would like to extract is
>>> information such as body Laterality, alleviating Factor, etc.
>>> We've been using cTAKES CVD with various AE but we seem to be mostly
>>> producing annotations with no information. I have used an working sample
>>> from a keynote on cTAKES as evidence of this problem.
>>>
>>> Working sample
>>>
>>>     https://uima.apache.org/downloads/gscl2013/keynote_ctakes.pdf page
>>> 68
>>>
>>>
>>>
>>> Running with sample input
>>>
>>>     http://i.imgur.com/3n6WyHE.png
>>>
>>>
>>>
>>> Any help would be appreciated.
>>>
>>>
>>>
>>> Thanks
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>
>>
>

Mime
View raw message