You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@ctakes.apache.org by Matthew Vita <ma...@gmail.com> on 2017/09/02 03:24:04 UTC
Re: ICD10 Dictionary Issue [EXTERNAL]
Sean,
I am pleased to inform you that I created a YouTube video for creating
ICD10 dictionaries for cTAKES: https://www.youtube.com/watch?v=4aOnafv-NQs.
Thank you again for the tips!
I haven't actually hooked up my cTAKES to use it (
https://github.com/tmills/ctakes-docker/blob/master/ctakes-as-pipeline/Dockerfile
is
what I will be using to pull it in), but I'm pretty sure I am on the right
track! Please let me know if I've missed anything.
Thanks,
Matthew Vita
www.matthewvita.com
On Mon, Aug 28, 2017 at 8:48 AM Finan, Sean <
Sean.Finan@childrens.harvard.edu> wrote:
> Hi Matthew,
>
> The dictionary creator uses a 'full' umls installation to create the
> custom dictionary. After that any installations of ctakes that contain
> your custom dictionary do not need the umls data.
>
> Sean
>
> -----Original Message-----
> From: Matthew Vita [mailto:matthewvita48@gmail.com]
> Sent: Sunday, August 27, 2017 9:20 PM
> To: dev@ctakes.apache.org
> Subject: Re: ICD10 Dictionary Issue [EXTERNAL]
>
> Sean,
>
> Nice to meet you. Thank you for your high-quality work on cTAKES.
>
> I will create the dictionary with the latest ICD10 data and contribute
> back my documentation + artifacts. My only question is with step 3:
> *"Select a UMLS installation directory. This is the directory containing
> the META/ subdirectory (which contains RRF files). After selecting the UMLS
> installation directory, the available vocabularies are gathered."* -
> basically, I have been providing my UMLS user/pass to the cTAKES Docker
> solution and letting it take care of the rest. I suppose I'm going to have
> to download and configure a custom UMLS dataset installation. This is
> correct?
>
> Thanks,
>
> Matthew Vita
> www.matthewvita.com
>
>
> On Sun, Aug 27, 2017 at 9:05 PM Finan, Sean <
> Sean.Finan@childrens.harvard.edu> wrote:
>
> > Hi Matthew,
> >
> > One of the updates that happened with ctakes 4.0 was the long-awaited
> > update to use the current version of hypersql (hsqldb) used by
> > components such as the dictionary lookup modules. That update made
> > all of the older dictionary databases obsolete as their format does
> > not match that used by the current hsqldb.
> >
> > Your best option is probably to create an updated '17 icd dictionary
> > of your own. Have a look at the wiki introduction to the dictionary
> > creator
> > gui:
> > https://urldefense.proofpoint.com/v2/url?u=https-3A__cwiki.apache.org_
> > confluence_display_CTAKES_Dictionary-2BCreator-2BGUI&d=DwIFaQ&c=qS4goW
> > BT7poplM69zy_3xhKwEW14JZMSdioCoppxeFU&r=fs67GvlGZstTpyIisCYNYmQCP6r0bc
> > pKGd4f7d4gTao&m=rn9A67xy_cGUI60xNsFsR3RpAKFa3Sqcq7CFhFkHdaQ&s=QnFljs_2
> > PkWy6ow43XdWzMHG9KW_4czYGAwy3gR3Oh0&e=
> >
> > Sean
> >
> > -----Original Message-----
> > From: Matthew Vita [mailto:matthewvita48@gmail.com]
> > Sent: Sunday, August 27, 2017 3:29 PM
> > To: dev@ctakes.apache.org
> > Subject: ICD10 Dictionary Issue [EXTERNAL]
> >
> > Hi cTAKES Maintainers,
> >
> > My name is Matthew Vita, a healthcare software developer and one of
> > the OpenEMR project administrators (this is a popular open source EHR
> > mainly used outside of the US).
> >
> > I am interested in integrating cTAKES with the EMR using Docker and a
> > friendly web frontend. Fortunately, Dr. Timothy Miller has provided an
> > excellent Docker pipeline solution that I have been using and enhancing.
> > However, I'm having an issue with ICD10 dictionary support.
> > Introducing
> > ctakesicd2015 simply has no effect.
> >
> > When observing the SNOMED/RXNORM dictionary, the structure is as follows:
> >
> > - sno_rx_16ab.xml
> > - sno_rx_16ab
> > - sno_rx_16ab.properties
> > - sno_rx_16ab.script
> >
> > However, when one pulls down the
> >
> > https://urldefense.proofpoint.com/v2/url?u=https-3A__sourceforge.net_p
> > _ctakesresources_code_HEAD_tree_trunk_ctakes-2Dresources-2Dsnomed-2Drw
> > ord-2Dhsqldb-2D2011ab_src_main_resources_org_apache_ctakes_dictionary_
> > lookup_fast_ctakesicd2015&d=DwIBaQ&c=qS4goWBT7poplM69zy_3xhKwEW14JZMSd
> > ioCoppxeFU&r=fs67GvlGZstTpyIisCYNYmQCP6r0bcpKGd4f7d4gTao&m=bmYyxtv1U77
> > r3grgB76TXATo01ayCe_Gu1kOvL5VvD8&s=XWhnGpcbuQvkPzwgSYg9uuRlzBBm_K_GwEp
> > rLsJ0QJk&e=
> > , the structure is:
> >
> > - ctakesicd2015.script
> > - ctakesicd2015.properties
> >
> > cTAKES does not pick up these files. I wonder if it's because the
> > ctakesicd2015.xml manifest is missing. Can anyone point me to a proper
> > download of this file?
> >
> > I very much appreciate the hard work behind this project and look
> > forward to hearing back.
> >
> > Thanks,
> >
> > Matthew Vita
> > www.matthewvita.com
> >
>
Re: ICD10 Dictionary Issue [EXTERNAL]
Posted by "Finan, Sean" <Se...@childrens.harvard.edu>.
Hi Matthew,
That is great! I am trying to be on vacation but can't wait to check out the video!
Sean
________________________________________
From: Matthew Vita <ma...@gmail.com>
Sent: Friday, September 1, 2017 11:24 PM
To: dev@ctakes.apache.org
Subject: Re: ICD10 Dictionary Issue [EXTERNAL]
Sean,
I am pleased to inform you that I created a YouTube video for creating
ICD10 dictionaries for cTAKES: https://urldefense.proofpoint.com/v2/url?u=https-3A__www.youtube.com_watch-3Fv-3D4aOnafv-2DNQs&d=DwIBaQ&c=qS4goWBT7poplM69zy_3xhKwEW14JZMSdioCoppxeFU&r=fs67GvlGZstTpyIisCYNYmQCP6r0bcpKGd4f7d4gTao&m=yytpXZalgLQI4oY7ELBurCxIbP-GKkjS3AeHNoehsxQ&s=eCy04Y5cdHQPEPFN4J2CTVmZXIVpk06JYPvd9czyh7o&e= .
Thank you again for the tips!
I haven't actually hooked up my cTAKES to use it (
https://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_tmills_ctakes-2Ddocker_blob_master_ctakes-2Das-2Dpipeline_Dockerfile&d=DwIBaQ&c=qS4goWBT7poplM69zy_3xhKwEW14JZMSdioCoppxeFU&r=fs67GvlGZstTpyIisCYNYmQCP6r0bcpKGd4f7d4gTao&m=yytpXZalgLQI4oY7ELBurCxIbP-GKkjS3AeHNoehsxQ&s=j1_OFYmNeIpUdo0jZkbUEjBMd6vEKYQ5nnUEnaGqYrU&e=
is
what I will be using to pull it in), but I'm pretty sure I am on the right
track! Please let me know if I've missed anything.
Thanks,
Matthew Vita
www.matthewvita.com
On Mon, Aug 28, 2017 at 8:48 AM Finan, Sean <
Sean.Finan@childrens.harvard.edu> wrote:
> Hi Matthew,
>
> The dictionary creator uses a 'full' umls installation to create the
> custom dictionary. After that any installations of ctakes that contain
> your custom dictionary do not need the umls data.
>
> Sean
>
> -----Original Message-----
> From: Matthew Vita [mailto:matthewvita48@gmail.com]
> Sent: Sunday, August 27, 2017 9:20 PM
> To: dev@ctakes.apache.org
> Subject: Re: ICD10 Dictionary Issue [EXTERNAL]
>
> Sean,
>
> Nice to meet you. Thank you for your high-quality work on cTAKES.
>
> I will create the dictionary with the latest ICD10 data and contribute
> back my documentation + artifacts. My only question is with step 3:
> *"Select a UMLS installation directory. This is the directory containing
> the META/ subdirectory (which contains RRF files). After selecting the UMLS
> installation directory, the available vocabularies are gathered."* -
> basically, I have been providing my UMLS user/pass to the cTAKES Docker
> solution and letting it take care of the rest. I suppose I'm going to have
> to download and configure a custom UMLS dataset installation. This is
> correct?
>
> Thanks,
>
> Matthew Vita
> www.matthewvita.com
>
>
> On Sun, Aug 27, 2017 at 9:05 PM Finan, Sean <
> Sean.Finan@childrens.harvard.edu> wrote:
>
> > Hi Matthew,
> >
> > One of the updates that happened with ctakes 4.0 was the long-awaited
> > update to use the current version of hypersql (hsqldb) used by
> > components such as the dictionary lookup modules. That update made
> > all of the older dictionary databases obsolete as their format does
> > not match that used by the current hsqldb.
> >
> > Your best option is probably to create an updated '17 icd dictionary
> > of your own. Have a look at the wiki introduction to the dictionary
> > creator
> > gui:
> > https://urldefense.proofpoint.com/v2/url?u=https-3A__cwiki.apache.org_
> > confluence_display_CTAKES_Dictionary-2BCreator-2BGUI&d=DwIFaQ&c=qS4goW
> > BT7poplM69zy_3xhKwEW14JZMSdioCoppxeFU&r=fs67GvlGZstTpyIisCYNYmQCP6r0bc
> > pKGd4f7d4gTao&m=rn9A67xy_cGUI60xNsFsR3RpAKFa3Sqcq7CFhFkHdaQ&s=QnFljs_2
> > PkWy6ow43XdWzMHG9KW_4czYGAwy3gR3Oh0&e=
> >
> > Sean
> >
> > -----Original Message-----
> > From: Matthew Vita [mailto:matthewvita48@gmail.com]
> > Sent: Sunday, August 27, 2017 3:29 PM
> > To: dev@ctakes.apache.org
> > Subject: ICD10 Dictionary Issue [EXTERNAL]
> >
> > Hi cTAKES Maintainers,
> >
> > My name is Matthew Vita, a healthcare software developer and one of
> > the OpenEMR project administrators (this is a popular open source EHR
> > mainly used outside of the US).
> >
> > I am interested in integrating cTAKES with the EMR using Docker and a
> > friendly web frontend. Fortunately, Dr. Timothy Miller has provided an
> > excellent Docker pipeline solution that I have been using and enhancing.
> > However, I'm having an issue with ICD10 dictionary support.
> > Introducing
> > ctakesicd2015 simply has no effect.
> >
> > When observing the SNOMED/RXNORM dictionary, the structure is as follows:
> >
> > - sno_rx_16ab.xml
> > - sno_rx_16ab
> > - sno_rx_16ab.properties
> > - sno_rx_16ab.script
> >
> > However, when one pulls down the
> >
> > https://urldefense.proofpoint.com/v2/url?u=https-3A__sourceforge.net_p
> > _ctakesresources_code_HEAD_tree_trunk_ctakes-2Dresources-2Dsnomed-2Drw
> > ord-2Dhsqldb-2D2011ab_src_main_resources_org_apache_ctakes_dictionary_
> > lookup_fast_ctakesicd2015&d=DwIBaQ&c=qS4goWBT7poplM69zy_3xhKwEW14JZMSd
> > ioCoppxeFU&r=fs67GvlGZstTpyIisCYNYmQCP6r0bcpKGd4f7d4gTao&m=bmYyxtv1U77
> > r3grgB76TXATo01ayCe_Gu1kOvL5VvD8&s=XWhnGpcbuQvkPzwgSYg9uuRlzBBm_K_GwEp
> > rLsJ0QJk&e=
> > , the structure is:
> >
> > - ctakesicd2015.script
> > - ctakesicd2015.properties
> >
> > cTAKES does not pick up these files. I wonder if it's because the
> > ctakesicd2015.xml manifest is missing. Can anyone point me to a proper
> > download of this file?
> >
> > I very much appreciate the hard work behind this project and look
> > forward to hearing back.
> >
> > Thanks,
> >
> > Matthew Vita
> > www.matthewvita.com
> >
>