One of the major barriers to leveraging EHR data for clinical and translational science is the tremendous unstructured or semi-structured clinical narratives remain unparsed and underused. As demonstrated by large scale efforts such as ACT (Accrual of patients for Clinical Trials), eMERGE, and PCORnet, using EHR data for research rests on the capabilities of a robust data and informatics infrastructure that allows the structuring of clinical narratives and supports the extraction of clinical information for downstream applications. Current successful NLP use cases often require a strong informatics team (with NLP experts) to work with clinicians to supply their domain knowledge and build customized NLP engines iteratively. This requires a close collaboration between NLP experts and clinicians, not scalable and infeasible at institutions with limited informatics support. Additionally, the usability, portability, and generalizability of NLP systems are still limited, partially due to the lack of access to EHRs across institutions to diversify the training and testing datasets for NLP systems. The limited availability of EHR data limits the training related to improve the workforce competence in clinical NLP. In this talk, I will discuss our efforts in addressing the above challenges through privacy-preserving computing, deep learning, and open science approaches to improve the scalability, utility, portability, and generalizability of NLP systems.
After participating in this activity, the learner should be better able to:
- Understand the roles of NLP in accelerating clinical and translational sciences
- Be familiar with challenges faced by the research community in adopting NLP solutions
- Be informed with the efforts in addressing the challenges in adopting NLP solutions
Hongfang Liu, PhD
Professor of Biomedical Informatics
Mayo Clinic College of Medicine.
Hongfang Liu, PhD, is a professor of biomedical informatics in the Mayo Clinic College of Medicine, and is a consultant in the Department of Health Sciences Research at Mayo Clinic. As a researcher, she is leading Mayo Clinic's clinical natural language processing (NLP) program with the mission of providing support to access clinical information stored in unstructured text for research and practice. Administratively, Dr. Liu serves as the section head for Medical Informatics in the Division of Biomedical Statistics and Informatics.
Dr. Liu's primary research interest is in biomedical NLP and data normalization. She has been developing a suite of open-source NLP systems for accessing clinical information, such as medications or findings from clinical notes. Additionally, she has been conducting collaborative research in the past decade in utilizing existing knowledge bases for high-throughput omics profiling data analysis and functional interpretation.
Dr. Liu's work in informatics has resulted in informatics systems that unlock clinical information stored in clinical narratives. Her work accelerates the pace of knowledge discovery, implementation and delivery for improved health care.