User Tools

Site Tools


projects:workgroups:minutes

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:minutes [2015/10/09 15:51]
anu_gururaj
projects:workgroups:minutes [2015/10/22 19:19]
anu_gururaj
Line 9: Line 9:
   -IRB for use of clinical text   -IRB for use of clinical text
   -Clinical text data storage and representation schema   -Clinical text data storage and representation schema
-    - Presentation by Dr. Noemie ElhadadTitle: NLP schemas and clinical NLP tools in ShARe+    - Presentation by Dr. Noemie Elhadad 
 +    - Title: NLP schemas and clinical NLP tools in ShARe, {{:​projects:​workgroups:​15ohdsi_nlp_share.pdf|File}} 
 +        - output of converted unstructured text could be in the form of structured data, bag of words and word embedding. Structured data and bag of words are the most useful in the current context. 
 +        - the ShARe schema for structured output combines many initiatives such as SHARP, THYME etc.
     - Discussion – Next steps     - Discussion – Next steps
 +        - Table structure for storing concept level NLP outputs to be determined
 +        - It is sufficient to start with structured output
 +        - A concept table with concept ID in each row and note IDs should be generated
 +        - OMOP vocabulary is to be used to aggregate concept to a higher level to manage and condense the number of concepts
 +        - Next step is to go through all the columns exhaustively for all attributes, merge them and then decide the attributes that should be used in the table
   -NLP tools/​pipelines for ETL   -NLP tools/​pipelines for ETL
   -Use cases, e.g, phenotyping for cohort selection using NLP outputs   -Use cases, e.g, phenotyping for cohort selection using NLP outputs
     - Presentation by Dr. Jon Duke, Title: Regenstrief NLP platform and approach to validation of phenotypes     - Presentation by Dr. Jon Duke, Title: Regenstrief NLP platform and approach to validation of phenotypes
 +        - the NLP platform is composed of a state machine with Regex based system
 +        - the NLP data analysis tool is currently being used for data analytics at Regenstrief.
 +        - the tool has text search capabilities and was demonstrated at the meeting
     - Discussion – Next steps     - Discussion – Next steps
 +        - need to determine if the API for keyword search based on Solr or ElasticSearch can be shared ​
   -Discussion   -Discussion
projects/workgroups/minutes.txt · Last modified: 2015/10/29 18:36 by anu_gururaj