NLP PreSymposium 2017: AMIA NLP-WG PreSymposium 2017 Washington Hilton Hotel Washington, DC, United States, November 4, 2017 |
Conference website | https://ohnlp.github.io/NLP_WG_Pre-symposium-2017/NLP-WG.html |
Submission link | https://easychair.org/conferences/?conf=amianlpwg2017 |
Abstract registration deadline | July 31, 2017 |
Submission deadline | July 31, 2017 |
The application of Natural Language Processing (NLP) in the general English domain has seen enormous progress over the past decades, progress that was largely enabled by availability of tools and resources that could be shared, reused, and improved in multiple projects and collaborations. Applying NLP to the textual content of patient electronic health records (i.e., clinical text) is limited by strict patient privacy and confidentiality laws and regulations. These limitations render access and sharing of resources (e.g., annotated text corpora) and tools based on this clinical text (e.g., trained machine learning algorithms) very difficult. As a consequence, the impact of NLP on clinical and translational research is limited. Despite these difficulties, several research teams have succeeded in creating and then sharing resources based on clinical text. Our general objective with this pre-symposium is to enhance the awareness of these resources and tools within the biomedical and clinical NLP communities and improve the reusability, portability, and interoperability of these tools and resources.
The pre-symposium will be divided into three 2-hour sessions with a lunch break dedicated to a poster session. The following are details:
Session 1 – Graduate Student Consortium
The purpose of the graduate student consortium is to provide opportunities for direct interactions between students and researchers in the biomedical and clinical NLP field, so that students can 1) refine their research focus; 2) discuss specific questions about study design, algorithm development, or evaluation plan; 3) receive constructive feedback and suggestions about their dissertation work; and 4) establish possible collaborations. Informal feedback from students at previous years’ doctoral consortiums was very positive; they appreciated the help offered in these four topics. A call for submissions will invite graduate students to submit applications for a podium presentation of their graduate research work (in the biomedical and clinical NLP fields).
Session 2 – NLP Highlight
A selection of existing and significant NLP related efforts will be highlighted in this session. A call for submissions will invite researchers to submit their significant research effort including those published, in press or under development projects between July 1, 2016 and August 31, 2017. The presenter should identify themselves as the corresponding author during the submission process, and accepted presenters are required to make the presentation themselves.
All highlight submissions will be evaluated according to the following criteria: 1) relevance, interest, and value of the topic to NLP-WG; 2) impact of the paper(s) on informatics/medicine/biology; 3) presentability of the work to a large, diverse audience; 4) quality of oral presentations by the submitter (if known); and 5) submissions that permit the presentation of related interesting unpublished new results will be viewed favorably. These "soft" criteria attempt to capture the underlying goal of the Highlight session, namely the presentation of exciting and thought-provoking efforts in advancing NLP in clinical and biomedical domain.
Session 3 – NLP tooling “Codeathon”
Ease of use is a very important aspect greatly influencing the adoption and ultimately the success of any tool. In the case of NLP, while many of the currently available tools are feature rich, their usability remains a problem as can be seen by the numerous questions on the forums regarding installation of the tool itself and usage of modules. Therefore, this session aims to improve the field of NLP tooling by bringing together the community of tool developers and users to provide hands-on experience of using the tools to solve specific tasks and explore areas of improvement in usability.
The specific objectives of the ‘codeathon’ include:
- Installation of the NLP tool(s).
- Familiar with NLP processing pipelines.
- Explore the use of the tools to develop baseline systems for specific tasks, e.g., adverse event detection or family history information extraction.
We will adopt a ‘codeathon’ format for this session. The following process will be used:
- We will send out a Call for Proposals (CFP) to participate as a provider or a user. Providers are typically expected to be teams with mature NLP tools. The providers will propose a task in the submission and will agree to provide directions to accomplish the task. The symposium committee will review the proposal to approve the usefulness of the proposed task for the community, and on approving will include the task. The tasks will be grouped under tracks based on the degree of expertise required and the theme of the task.
- Groups will be formed by the submission to implement actual NLP systems for specific tasks in one month period where providers will serve as mentors for the users to develop the system.
- In the ‘codeathon’ session, the user team will demo the system and the duos will discuss how they achieve their goals and any lessons learnt to share with the audiences.
We will make efforts to coordinate the content of our pre-symposium so that we complement but do not overlap with specific talks or panels at AMIA on similar topics, once we know which presentations have been accepted.
The allocation of time will be adjusted based on the number of attendees and the number of submissions received. We also plan to invite poster submissions so a larger group of participants can present their research work.
Submission Guidelines
All submissions have a page limit of 2 pages using AMIA template (https://www.amia.org/sites/default/files/AMIA2017-Submission-Template.docx).
Poster: All researchers in the biomedical and clinical NLP fields are invited to submit applications for a poster of their research. Please follow AMIA poster format.
Graduate Student Consortium: Graduate students are invited to submit applications for a podium presentation of their graduate research work (in the biomedical and clinical NLP fields). The submission is suggested to include the following sections:
- Aims and Objectives - State the main objective(s) of your project.
- Justification for the Research Topic - Explain the motivations for your project.
- Research Questions - Stating your research question is essential. This might be done in a list.
- Research Methodology - If you already have plans for your research methodology, explain them here. If you have not found an appropriate methodology yet, or wonder which one to choose, this is also the place to mention it. In this case, list the requirements your methodology should fulfill.
- Research Results to Date - You are not required to have results. But if you already have some, present them here.
- References – Any citation.
Highlight: All researchers in the biomedical and clinical NLP are invited to submit their significant paper or research projects. The submission is suggested to include the following sections:
- Overview of the paper or project – It can be an abstract or project overview.
- Justification of Highlight – Explain the relevance, interest, and value of the topic to AMIA NLP-WG and the impact of the paper(s) or efforts on informatics/medicine/biology.
Codeathon: For participating as a Provider, please provide a two-page document, detailing thetargeted NLP tasks and directions to accomplish each task. The following sections are suggested:
- Aims and Objectives - State the main objective(s) of your participation in the Hackathon.
- Description of your NLP Application - Describe your tool briefly including its features, its advantages, disadvantages, etc.
- Proposed Tasks - List of tasks to be accomplished by your users to best showcase your tool.
- Projects/Products using your tool - List projects or products that are using the NLP application. Mention if this is in a production setting.
- References - List bibliographic references in Vancouver format
Committees
- Hongfang Liu, PhD, Section of Medical Informatics, Department of Health Sciences Research, Mayo Clinic, Rochester, MN
- Rong Xu, PhD, Case Western Reserve University, Cleveland, OH
- Stephane Meystre, MD, PhD, Medical University of South Carolina, Charleston, SC
- Sivaram Arabandi, MD, MS, Ontopro, Houston, TX
- Kavishwar Wagholikar, MBBS, PhD, Massachusetts General Hospital, Boston, MA
- Dina Demner-Fushman, MD, PhD, National Library of Medicine, Bethesda, MD
- Jon Patrick, PhD, Product Development Director, Health Language Analytics, Sydney, Australia
- Guergana Savova, PhD, Boston Children’s Hospital, Harvard Medical School, Boston, MA
- Ozlem Uzuner, PhD, Department of Computer Science, University at Albany, SUNY, Albany, NY
- Chunhua Weng, PhD, Columbia University, New York, NY
- Hua Xu, PhD, The University of Texas Health Science Center at Houston, TX
- Pierre Zweigenbaum, PhD, LIMSI-CNRS, Université Paris-Saclay, Orsay, France
Venue
AMIA NLP-WG PreSymposium is part of AMIA Annual Symposium 2017 (https://www.amia.org/amia2017).
Contact
All questions about submissions should be emailed to Dr. Hongfang Liu (liu.hongfang@mayo.edu).