Abstract: Clinicians express concern they may be unaware of important information contained in voluminous scanned and other outside documents contained in EHR records. We compared use of automated methods (OCR with NLP) and human chart review to find risk factors for heritable breast and ovarian cancer within outside EHR records. We found automated methods identified most cancer risk factor information that would otherwise require clinician manual review and therefore is at risk for being missed.

Describe the new knowledge and additional skills the participant will gain after attending your presentation.: Better understanding of the risk that outside documents stored in the EHR may contain actionable clinical information unknown to the treating provider.
Familiarity with commercial tools that may be leveraged to search PDF documents for important clinical information.


Alexander Kostrinsky-Thomas (Presenter)
Pacific Northwest University of Health Sciences

Fuki Hisama, University of Washington School of Medicine
Thomas Payne, University of Washington School of Medicine