User Tools

Site Tools


projects:workgroups:wg_meeting_02032016

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
Next revision
Previous revision
Last revision Both sides next revision
projects:workgroups:wg_meeting_02032016 [2016/03/07 21:29]
anu_gururaj
projects:workgroups:wg_meeting_02032016 [2016/03/07 22:50]
anu_gururaj
Line 7: Line 7:
 ==== Agenda ==== ==== Agenda ====
  
-      ​-IRB for use of clinical text +{{:​projects:​workgroups:​nlp_wg_meeting_02032016_final.pdf|}} 
-      -Clinical text data storage ​and representation schema + 
-      -NLP tools/​pipelines ​for ETL +  ​Minimal Model Presentation – Alex 
-      -Use cases, e.g, phenotyping ​for cohort selection ​using NLP outputs +  Note-type mapping Presentation – Karthik 
-      -Discussion+  - Share existing ontologies from Vanderbilt (Hua) and Regenstrief (Jon) 
 +  Share strategies ​for combining data from different searches – Jon 
 +  Report on WG for commenting – Hua 
 +  - Wrappers for cTAKES and Metamap – Min 
 +  - Improvements to search engine set up using MT samples – Min 
 +  Textual Data Representation – Discussion 
 +  - Goals of 2016 
 +  - Change of meeting time
  
 ===Minutes=== ===Minutes===
  
-  - General IRB document for use of clinical text and approval from all contributors,​ post online ​Almost completed +  - Minimal model presentation ​Alex {{:projects:​workgroups:​ohdsi_nlp_wg_yahi.pdf|}} 
-  - Collect minimum set of modifiers for all clinical entities that support use of rule to derive clinical conceptsAlex +        - the model is based on the SHARE-N ​model and adapted to the current data structure. This model incorporates other semantic types and all of the modifiers are not available in cTAKES yet
-      * cTAKES ​is being run on clinical notes programmatically. Alex will present ​the minimal ​model in the next meeting+        the notes were processed ​from eMERGE cohort at Columbia with about 60,000 notes encompassing 1700 patients. The original patient number was 3200. 
-  Aggregate and share note-type metadata ​from various sources: Karthik +        - In theory, ​set containing the combination ​of minimal modifiers can be generatedPractically,​ can we trust the data enough ​to add it into OHDSI tables? - only highest confidence data (with maximum PPVshould be added to the tables. 
-      * LOINC note type mapping would be very useful resource. We should generate hierarchical representation ​of note-types as an ontologyKarthink will present his work to date at the next meeting. +        Next steps
-      * Existing ontology for note types to be shared : Vanderbilt ​(Hua) and Regenstrief (Jon+          - Look at the note sections to determine the errors
-  Simple search set up for MT samplesMinPresentation +          - Work with Sunny to generate ​the NLP outputs for the phenotyping data 
-      * Presentation +          - Evaluate by comparisons with structured data 
-      * The interface being developed should present a summary with visualization for patients/​notes+          - Make the system more robust 
-      * We will add Boolean query options ​to improve ​the search +          ​- ​Generate a protocol and/or annotation guidelines 
-      * We will implement a Ranking algorithm +          - Share the data as Gold standard with manually annotated CUIs 
-      * Assign fake patient ID's to the notes to generate the visualization portion. +          ​- Alex's script is to be tried on different datasets and evaluated across notes from different ​institutions 
-      ​* ​Generate a program like Circe to define the patient cohort +          ​- Identify minimal set of notes to work with when recommending to the OHDSI community 
-      * Next steps: How to move the data from textual searches stored in table outside of OMOP to the OMOP? +          Identify sets of concepts that are not reliable - negation is a very good example of this idea. 
-          ​* Structured searches from CDW and textual searches can be combined using existing strategies. Jon will share the slides of his presentation ​on combining data from different ​searches +          Continue discussion of NLP system evaluation across different sites 
-          ​* Run NLP on the ElasticSearch ​to extract information +   ​- The NLP-WG will meet on second Wednesday of every month
-  Wrappers for cTAKES and Metamap ​ +
-  Report on the WG - Hua will generate and share with the members for comments +
-  - The best ways to represent textual data need to be determined+
  
 ===Action Items=== ===Action Items===
  
-  ​- Minimal Model Presentation - Alex +  - Note-type mapping Presentation - Karthik
-  ​- Note-type mapping Presentation - Karthink+
   - Share existing ontologies from Vanderbilt (Hua) and Regenstrief (Jon)   - Share existing ontologies from Vanderbilt (Hua) and Regenstrief (Jon)
   - Share strategies for combining data from different searches - Jon   - Share strategies for combining data from different searches - Jon
Line 45: Line 48:
   - Improvements to search engine set up using MT samples - Min   - Improvements to search engine set up using MT samples - Min
   - Textual Data Representation - Discussion   - Textual Data Representation - Discussion
 +  - NLP system evaluation across different sites - Discussion
projects/workgroups/wg_meeting_02032016.txt · Last modified: 2016/03/09 20:31 by anu_gururaj