Differences

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

--- projects:workgroups:wg_meeting_03092016 [2016/04/13 16:22] – anu_gururaj1
+++ projects:workgroups:wg_meeting_03092016 [2016/04/13 16:51] (current) – anu_gururaj1
@@ Line 23: / Line 23: @@
   - Note-type mapping Presentation – Karthik {{:projects:workgroups:ohdsi-nlp-wg1457551276.mp4|}}
         - INYP terminology services provides clinical document browser based on LOINC ontology for clinical notes. The paper describing the system is published and the pdf is attached. {{:projects:workgroups:loinc_note_mapping_paper.pdf|}}
-        - 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.
+        - the system allows for comparison of configs across different sites.
-        - the notes were processed from eMERGE cohort at Columbia with about 60,000 notes encompassing 1700 patients. The original patient number was 3200.
+        - Vojtech also has looked into using LOINC ontology for note types. {{:projects:workgroups:loinc_note_types_marshall_clinic.pdf|}}
-        - In theory, a set containing the combination of minimal modifiers can be generated. Practically, can we trust the data enough to add it into OHDSI tables? - only highest confidence data (with maximum PPV) should be added to the tables.
+        - With reference to standard note types, one of the suggestions was to use it in the database for the output of the NLP system.
+        - The note table in the OMOP CDM has a note type field with about 10 or so note types
+        -
         - Next steps:
-          - Look at the note sections to determine the errors.
+          - Comparison of what aspects of LOINC is being used across different institutions.
-          - Work with Sunny to generate the NLP outputs for the phenotyping data
+          - One of the deliverable is to build a system that can map the institutional note type to the LOINC note types.
-          - Evaluate by comparisons with structured data
+          - Update the note type field of the note table in OMOP CDM. Should a subset of LOINC document types be used for this purpose? this would depend on the real-world data from various institutions.
-          - Make the system more robust
+          - Ontologies such as CP, ICD-9 etc. that include more clinical terminologies should also be explored.
-          - Generate a protocol and/or annotation guidelines
-          - Share the data as a Gold standard with manually annotated CUIs
-          - Alex's script is to be tried on different datasets and evaluated across notes from different institutions
-          - Identify minimal set of notes to work with when recommending to the OHDSI community
-          - Identify sets of concepts that are not reliable - negation is a very good example of this idea.
-          - Continue discussion of NLP system evaluation across different sites
    - existing ontologies from Vanderbilt - Hua - {{:projects:workgroups:document-types.zip|}}
@@ Line 42: / Line 38: @@
   - Suggestion for OMOP model regarding NLP output - Hua/Noemie
-  - Share existing ontologies from Regenstrief (Jon)
+  - 2016 OHDSI symposium - plan is to present the search engine, medical record viewer (a chart review tool) from Scott; any other suggestions?
   - 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
   - NLP system evaluation across different sites - Discussion