i2b2 Clinical NLP Data Sets

October 26, 2016
Free for AMIA members; $50 for non-members
Ozlem Uzuner, PhD

This webinar will give a tour of the i2b2 clinical data sets that have been developed for the i2b2 shared tasks since 2006.  The topics covered by the data sets include de-identification, smoking status classification, diagnosis of obesity and its comorbidities, medication extraction, concepts, assertions, and relations, coreference resolution, temporal relations, heart disease risk factors, and research domains for psychiatry.  All data sets are distributed through i2b2.org/NLP with a data use agreement.

Learning Objectives

After participating in this activity, the learner should be better able to:

  • Learn about various data sets
  • Know the different annotation methods utilized for generating the data sets
  • Incorporate the data sets into their research

Speaker Information

Ozlem Uzuner, PhD
Associate Professor
Computer Science Department, University at Albany SUNY

Dr. Uzuner is an associate professor at the University at Albany, SUNY.  She is also a research affiliate at the Massachusetts Institute of Technology Computer Science and Artificial Intelligence Laboratory.  She has been organizing and chairing annual clinical shared tasks known as the i2b2 challenges since 2006.