Can the large language model <scp>ChatGPT</scp> ‐4omni predict outcomes in adult patients with status epilepticus?
ChatGPT-4o unreliably predicts outcomes in patients with status epilepticus. Clinicians should refrain from using ChatGPT-4o for prognostication in these patients.
Authors
- Sebastian Berger (ORCID: https://orcid.org/0009-0006-0654-6479)
- Pascale Grzonka (ORCID: https://orcid.org/0000-0002-6723-320X)
- Sira M. Baumann (ORCID: https://orcid.org/0000-0002-4762-1408)
- Caroline E. Gebhard
- Raoul Sutter (ORCID: https://orcid.org/0000-0002-6575-356X)
- Paulina S. C. Kliem (ORCID: https://orcid.org/0000-0003-3521-5880)
- Armon Arpagaus (ORCID: https://orcid.org/0009-0007-2065-6698)
- Simon B. Egli
- Urs Fisch (ORCID: https://orcid.org/0000-0003-1557-9062)
- Simon Adrian Amacher (ORCID: https://orcid.org/0000-0001-6530-1240)
- Sabina Hunziker
Institutions
- Brigham and Women's Hospital (US)
- Harvard University (US)
- University of Freiburg (DE)
- University of Basel (CH)
- University Hospital of Basel (CH)
Publication Details
- Journal
- Epilepsia
- Published
- 2024-12-26
- DOI
- https://doi.org/10.1111/epi.18215
- Citations
- 6
- Primary Topic
- Epilepsy research and treatment
- Type
- article
- Field-Weighted Citation Impact
- 2.31
Funders
- Universität Basel
- Mach-Gaensslen Foundation of Canada
- Universitätsspital Basel
- Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung