This shows you the differences between two versions of the page.
Both sides previous revision Previous revision Next revision | Previous revision | ||
projects:workgroups:wg_meeting_03092016 [2016/04/13 16:16] anu_gururaj1 |
projects:workgroups:wg_meeting_03092016 [2016/04/13 16:51] (current) anu_gururaj1 |
||
---|---|---|---|
Line 22: | Line 22: | ||
- Note-type mapping Presentation – Karthik {{:projects:workgroups:ohdsi-nlp-wg1457551276.mp4|}} | - Note-type mapping Presentation – Karthik {{:projects:workgroups:ohdsi-nlp-wg1457551276.mp4|}} | ||
- | - 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. | + | - 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 notes were processed from eMERGE cohort at Columbia with about 60,000 notes encompassing 1700 patients. The original patient number was 3200. | + | - the system allows for comparison of configs across different sites. |
- | - 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. | + | - Vojtech also has looked into using LOINC ontology for note types. {{:projects:workgroups:loinc_note_types_marshall_clinic.pdf|}} |
+ | - 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: | - 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|}} | - existing ontologies from Vanderbilt - Hua - {{:projects:workgroups:document-types.zip|}} | ||
===Action Items=== | ===Action Items=== | ||
- | - Share existing ontologies from Regenstrief (Jon) | + | - Suggestion for OMOP model regarding NLP output - Hua/Noemie |
+ | - 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 | - Share strategies for combining data from different searches - Jon | ||
- | - Report on WG for commenting - Hua | ||
- Wrappers for cTAKES and Metamap - Min | - Wrappers for cTAKES and Metamap - Min | ||
- 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 | - NLP system evaluation across different sites - Discussion |