Phenotype WG - Nigam Shah
Who can test models for two conditions that we have built?
Who can include text data into the CDM v5, and by when, to build models for two more phenotypes?
Who can volunteer to build models without text data, and quantify the gain/loss in performance?
LAERTES - Rich Boyce
CDM Development - Christian Reich Vocabulary Improvement - Christian Reich
CMS/Medicare - Mark Danese
Heracles WG - Questions for the group:
Have you have built cohorts through Circe or other means?
How are you currently exploring your cohorts? What are the major use cases?
Have you tried installing HERACLES (either via OLYMPUS or directly from GitHub)?
If there is anything else you would like to discuss, please reply to this thread.
I will not be able to make this meeting but I look forward to a follow up discussion on Heracles. I invite those who are working through the cohort analysis process (using Heracles or otherwise) to post on the forums and let us know what their experience has been.
All, I am a new member joining in today, and hope to contribute to all types of predictive/descriptive modeling efforts (Plato/Homer/…) and participate in the Phenotype WG.
Maura has already posted my profile on the OHDSI collaborators list, more gory details at http://hercules.ece.utexas.edu/ghosh/
Welcome @joydeep to OHDSI! Note that in addition to the Phenotype WG lead by @nigam, there is also some NLP-based phentoype activities happening within @Hua Xu’s new NLP WG.
Thanks for the welcome, Jon. Hua is within driving distance and I met him
at AMIA SF.
I wrote several statistical NLP papers 15 years back, but at this point I
mostly apply existing NLP techniques rather than develop new ones, as this
area has taken a life of its own. So this is a natural partnership.
regards,
Joydeep Ghosh
Schlumberger Centennial Chaired Professor
Dept of Electrical and Computer Engineering
University of Texas at Austin,
1 Univ. Station C0803, Austin, TX 78712