Automated epilepsy and seizure type phenotyping with pre-trained language models
By extracting expert-level epilepsy phenotypes from routine clinical text at scale, this approach transforms unstructured EHR data into a resource for longitudinal, population-informed epilepsy care. Automated phenotyping enables analyses of epilepsy trajectories and treatment outcomes that are not feasible with structured data alone, supporting future clinical and translational research applications.
Authors
- Kevin Xie (ORCID: https://orcid.org/0000-0003-1849-2085)
- Jacob Korzun (ORCID: https://orcid.org/0009-0006-9167-3244)
- Erin C. Conrad (ORCID: https://orcid.org/0000-0001-8910-1817)
- Daniel J. Zhou (ORCID: https://orcid.org/0009-0008-7555-7918)
- Ellie Chang
- Dan Roth
Institutions
- Translational Therapeutics (United States) (US)
- University of Pennsylvania (US)
- Philadelphia University (US)
Publication Details
- Journal
- medRxiv
- Published
- 2026-02-22
- DOI
- https://doi.org/10.64898/2026.02.11.26346003
- Citations
- 1
- Primary Topic
- Machine Learning in Healthcare
- Type
- article
- Field-Weighted Citation Impact
- 52.10
Funders
- National Institutes of Health
- National Institute of Neurological Disorders and Stroke