Drug labels (prescribing information or package inserts) describe what a particular medicine is supposed to do, who should or should not take it, how to use it, and specific safety concerns. The US Food and Drug Administration (FDA) regulates the content and format of these labels. One of the major aspects of drug information are safety concerns in the form of Adverse Drug Reactions (ADRs). The purpose of this FDA-sponsored TAC track is to evaluate information extraction (IE) performance on adverse reactions found in drug labels. The FDA is highly interested in automatic extraction of ADRs from drug labels for many purposes. Two possible applications enabled by this task are: (1) comparing the ADRs present in labels from different manufacturers for the same drug, and (2) performing post-marketing safety analysis (pharmacovigilance) by identifying new ADRs not currently present in the labels.
A large set of labels will be provided to participants, including a training dataset annotated with ADRs and detailed supporting annotations (severity, negation, factors, etc.). The positive ADRs are also normalized to MedDRA Preferred Terms (PT). In particular four tasks will be evaluated: (1) mention-level extraction of ADRs and related entities, (2) relations between ADRs and related entities, (3) label-level (i.e., document-level) extraction of unique, positive ADRs, and (4) label-level normalization of ADRs to MedDRA.
--- Timeline --- May 2017: Registration deadline for participants. September 2017: Participant submissions due. Early October 2017: Individual results sent to participants. 1 November 2017: Participant workshop notebook papers due. 13-14 November 2017: TAC 2017 Workshop in Gaithersburg, MD, USA. Mid October 2017: Participant workshop notebook papers due. February 2018: Final proceedings papers due.
--- Website --- https://bionlp.nlm.nih.gov/tac2017adversereactions/
--- Mailing List --- tac-adr at googlegroups.com -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: text/html Size: 2403 bytes Desc: not available URL: <https://mailman.uib.no/public/corpora/attachments/20170503/54b7e936/attachment.txt>