[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[eu_members at aclweb dot org] Washington, DC: Challenges in Natural Language Processing for Clinical Data, Second i2b2 Shared-Task and Workshop -- Call for Participation



Call for Participation

Second i2b2 Shared-Task and Workshop
Challenges in Natural Language Processing for Clinical Data
Obesity Challenge (A Shared-Task on Obesity): Who's obese and what co-morbidities do they (definitely/likely) have?

 in conjunction with AMIA Fall Symposium
Workshop: 7-8 November, 2008 in Hilton Washington and Towers, Washington, DC

Registration Deadline: 31 October, 2008

Register at: https://i2b2.org/NLP

Organizer: Informatics for Integrating Biology and the Bedside, i2b2, a National Center for Biomedical Computing

The second i2b2 Workshop on Challenges in Natural Language Processing for Clinical Data will present the top performing systems from the i2b2 shared-task on obesity.  

The obesity shared-task was a multi-class, multi-label classification task focused on obesity and its co-morbidities. The data for the shared-task consisted of discharge summaries from Partners Healthcare. All records had been fully de-identified. Obesity information and co-morbidities had been marked at a document level as present, absent, questionable, or unmentioned in the documents. For each patient, both textual judgments, i.e., what the text explicitly states about obesity and co-morbidities, and intuitive judgments, i.e., what the text implies about obesity and co-morbidities, were provided. The goal of the shared-task was to evaluate systems on their ability to recognize whether a patient is obese and what co-morbidities they exhibit.

The i2b2 workshop is now open to registration.  (https://i2b2.org/NLP)

Organizing Committee:

Ozlem Uzuner, Chair, SUNY at Albany
Middle East Technical University Northern Cyprus Campus
Peter Szolovits, MIT CSAIL
Isaac Kohane, Partners Healthcare

Questions on the challenge can be addressed to Ozlem Uzuner, i2b2nlp at albany dot edu.