Privacy Research & Clinical Text Deidentification with NLM-Scrubber
We strive to discover new clinical facts to promote evidence-based clinical sciences, but such potential discoveries are locked in electronic health record systems due to privacy concerns. Unfortunately, there is no silver bullet to resolve this vexing social concern. In this presentation, Dr. Mehmet Kayaalp will deconstruct the problem to understand what makes privacy so complex. How can we tap into big health data while preserving the privacy of the patient? One technological solution is NLM-Scrubber, a clinical text de-identification tool developed at the National Library of Medicine. Dr. Kayaalp will discuss what NLM-Scrubber offers to clinical scientists, data managers, and privacy officers in academic medical settings.
Through this presentation, the audience will learn about:
- factors that make privacy a very difficult social problem,
- what constitutes protected health information,
- what the HIPAA Privacy Rule offers to protect patient privacy and to free health information for secondary use, and
- how to use NLM-Scrubber most effectively.
Presenter: Dr. Mehmet Kayaalp is a staff scientist at Lister Hill National Center for Biomedical Communications at the National Library of Medicine, National Institutes of Health. He has published dozens of peer-reviewed articles on artificial intelligence in medicine, probabilistic ontologies, natural language processing, statistical machine learning, and privacy. He received his M.D. from the University of Istanbul in 1989, his M.S. in computer science from Southern Methodist University in 1993, and his Ph.D. in intelligent systems from the University of Pittsburgh in 2003.