A modular pipeline for natural language processing-screened human abstraction of a pragmatic trial outcome from electronic health records
We present the design and source code for a scalable and efficient pipeline for measuring complex electronic health record-derived outcomes using natural language processing-screened human abstraction. This implementation is adaptable to diverse research needs, and its modular pipeline represents a practical middle ground between custom software and commercial platforms.
Authors
- Danae Dotolo (ORCID: https://orcid.org/0000-0001-5838-4834)
- Trevor Cohen
- Jessica Walsh (ORCID: https://orcid.org/0000-0001-9723-3043)
- Kasey Mallon Andrews
- Robert Y. Lee (ORCID: https://orcid.org/0000-0001-5535-287X)
- Erin K. Kross (ORCID: https://orcid.org/0000-0002-8206-3050)
- Elizabeth L. Nielsen
- Nicole Leduc
- WB Lober
- James Sibley
- Janaki O’Brien
- Kevin S Li
- Anna Ungar
Institutions
- New Mexico Department of Health (US)
- University of New Mexico (US)
- University of Washington (US)
- Cambia (AU)
- University of Washington Medical Center (US)
- Neurobehavioral Systems (US)
- Seattle Children's Hospital (US)
Publication Details
- Journal
- Clinical Trials
- Published
- 2026-02-12
- DOI
- https://doi.org/10.1177/17407745251405386
- Primary Topic
- Machine Learning in Healthcare
- Type
- article
- Field-Weighted Citation Impact
- 0.00