Clinical Natural Language Processing Workshop at COLING 2016

Osaka, Japan. December 11, 2016.

Clinical text offers unique challenges that differentiate it not only from open-domain data, but from other types of text in the biomedical domain as well. Notably, clinical text contains a significant number of abbreviations, medical terms, and other clinical jargon. Clinical narratives are characterized by non-standard document structures that are often critical to overall understanding. Narrative provider notes are designed to communicate with other experts while at the same time serving as a legal record. Finally, clinical notes contain sensitive patient- specific information that raise privacy and security concerns that present special challenges for natural language systems.

We invite high-quality original submissions that develop methods to address the above challenges to the NLP in the clinical domain. We are interested in the work that specifically focuses on advancing state-of-the-art in clinical NLP, rather than merely applies existing NLP systems to downstream clinical problems (such as outcome prediction or clinical cohort selection). The submissions may include initial results from promising new methods that may spark interest from other members of the Clinical NLP community and lead to collaborative work. The following is list of topics of interest for this workshop:

  • Modeling clinical text in standard NLP tasks (tagging, chunking, parsing, entity identification, relation extraction, coreference, summarization, etc.)
  • De-identification of protected health information (PHI), surrogate PHI generation
  • Structure of clinical documents (e.g., section identification)
  • Information extraction from clinical text
  • Integration of structured and textual data for clinical tasks
  • Domain adaptation techniques for clinical data
  • Medical terminologies and ontologies
  • Annotation schemes and annotation methodologies