Zeng Q, VA SLC Healthcare system; Garvin J, VA SLC Healthcare system; Brandt C, VA Connecticut Healthcare system; Luther S, James A. Haley Veterans Hospital;
Workshop Objectives:
The VA has invested hugely in electronic medical records and has achieved a nationwide system that collects medical information from all patients. Currently, the textual information in the medical records is inaccessible to all but a small number of researchers. In order to obtain the highest value from this existing system, administrators and practitioners need to be able to access the textual information they need. It is our responsibility to get the most benefit from this resource for biomedical and patient care. Clinical natural language processing (NLP) is an important part the solution. While the majority of NLP tools are still not designed for non-NLP experts to use, some new NLP tools have been developed specifically for biomedical researchers. The goal of this workshop is to introduce the state of art user friendly NLP tools to health service researchers and facilitate their exploration and utilization of the large VA textual datasets.
Activities:
We will first provide a brief introduction to the basics of clinical NLP. Secondly, we will demonstrate several user friendly NLP tools for different text processing needs including retrieval of clinical notes, cohort identification, text annotation, text categorization, and information extraction. These tools will be presented in the context of real VHA use cases. Their strengths and weaknesses will be discussed. Thirdly, we will introduce the participants to the NLP ecosystem which we have developed as part of the SLC COIN and IDEA 2.0 CREATE project. Through the ecosystem, the participants will be given a hands-on experience and invited to experiment with the user-friendly tools. Finally, we will invite the participants to describe their own NLP needs and discuss strategies to meet their needs. The NLP strategies will include the use of existing tools, development of new tools, evaluation metrics, and budget/time considerations.
Target Audience:
Health services and operational researchers and analysts who have no or limited experience with NLP, at the same time, have concrete use cases or future plans to explore the rich clinical texts in VA medical records.
Assumed Audience Familiarity with Topic:
Interest in utilizing the free text data for clinical or operational research. No prior NLP experience required.