You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@ctakes.apache.org by "Savova, Guergana" <Gu...@childrens.harvard.edu> on 2017/01/31 21:50:35 UTC
RE: gold standard annotations for cTAKES [SUSPICIOUS]
Thank you, Sean!
Yes, absolutely -- we welcome volunteers for the gold annotations!
Regards,
--Guergana
-----Original Message-----
From: Finan, Sean [mailto:Sean.Finan@childrens.harvard.edu]
Sent: Tuesday, January 31, 2017 4:08 PM
To: dev@ctakes.apache.org
Subject: RE: gold standard annotations for cTAKES [SUSPICIOUS]
Hi all,
I just have a couple of notes to expand upon what Guergana wrote.
Anafora requires a schema for annotation and it requires text files to be in a certain structure. I just checked in text files for annotation and the schema that we plan to use in ctakes-examples-res src/main/resources/org/apache/ctakes/examples/annotation/ .
Everybody is obviously welcome to use the schema and notes, or to create annotations using another tool for all to share.
As a disclaimer ... Anafora is not associated with ctakes. My opinion is that the ctakes devlist should not be over-used for anafora q/a.
Thanks,
Sean
-----Original Message-----
From: Savova, Guergana [mailto:Guergana.Savova@childrens.harvard.edu]
Sent: Tuesday, January 31, 2017 3:42 PM
To: dev@ctakes.apache.org
Subject: gold standard annotations for cTAKES [SUSPICIOUS]
A while ago our physician colleague John Green created 16 realistically looking (but fake) clinical notes. Many thanks again, John!
These notes are in ctakes-examples/data/notes. We now volunteer to annotate them with gold annotations. The main elements with their attributes are:
Medications, Attributes ::= span associatedCode change_status_model conditional dosage_model duration_model end_date form_model frequency_model generic negation_indicator route_model start_date strength_model subject uncertainty_indicator
Signs/Symptoms, Attributes ::= associated_code body_location conditional course duration end_time generic historyOf negation_indicator relative_temporal_context severity start_time subject uncertainty_indicator
Anatomical Sites, Attributes ::= associatedCode conditional generic negation_indicator subject uncertainty_indicator
Disease/DisordersAttributes ::= associated_code body_location conditional course duration end_time generic historyOf negation_indicator relative_temporal_context severity start_time subject uncertainty_indicator
Procedures, Attributes ::= associated_code body_location conditional duration end_time generic historyOf method negation_indicator relative_temporal_context start_time subject uncertainty_indicator
We expect to have the gold annotations by end of March. We are using the Anafora annotation tool (https://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_weitechen_anafora&d=DwIFAg&c=qS4goWBT7poplM69zy_3xhKwEW14JZMSdioCoppxeFU&r=fs67GvlGZstTpyIisCYNYmQCP6r0bcpKGd4f7d4gTao&m=klIlU3or2Lr4NKPbcLwbF6pes2n2Ype-qri4zGIW_Xk&s=DE-u9g6s9UaCO6fLztks2ClRi7lrSCi5IkV5jtu3BPc&e= ) and will release the annotations in the xml format.
Regards,
--Guergana
Guergana Savova, PhD, FACMI
Associate Professor
PI Natural Language Processing Lab
Boston Children's Hospital and Harvard Medical School
300 Longwood Avenue
Mailstop: BCH3092
Enders 144.1
Boston, MA 02115
Tel: (617) 919-2972
Fax: (617) 730-0817
Guergana.Savova@childrens.harvard.edu<ma...@childrens.harvard.edu>
Harvard Scholar: https://urldefense.proofpoint.com/v2/url?u=http-3A__scholar.harvard.edu_guergana-5Fk-5Fsavova_biocv&d=DwIFAg&c=qS4goWBT7poplM69zy_3xhKwEW14JZMSdioCoppxeFU&r=fs67GvlGZstTpyIisCYNYmQCP6r0bcpKGd4f7d4gTao&m=klIlU3or2Lr4NKPbcLwbF6pes2n2Ype-qri4zGIW_Xk&s=8AV7t2x3gPeu3zXjyzKyiyi6KUNsNO2Qv2Jmsx2Ys1M&e=
ctakes.apache.org
thyme.healthnlp.org
cancer.healthnlp.org
share.healthnlp.org