Abstract: Recent deep learning has shown superior performance in many artificial intelligence applications, including clinical information extraction. However, developing innovative high-performance information extraction approaches using electronic health record narratives remains a challenging natural language processing (NLP) task. This workshop aims to bring together biomedical researchers, clinicians and industry professionals to discuss methodological advances in clinical information extraction. Specifically, over 10 participant teams of NLP Challenges for Detecting Medication and Adverse Drug Events (MADE1.0) will present their novel design, methodologies as well as state-of-the-art results using over 1,000 expert-annotated EHR notes as a common ground truth gold standard.

Learning Objective 1: Advanced methods in clinical NLP and their applications in clinical knowledge discovery.

Learning Objective 2 (Optional): Overview of current progress in clinical information extraction and the MADE1.0 results

Learning Objective 3 (Optional): Presentations of selected MADE1.0 participant teams.


Hong Yu (Presenter)
University of Massachusetts Lowell

Presentation Materials: