Artificial intelligence (ChatGPT 4.0) vs. Human expertise for epileptic seizure and epilepsy diagnosis and classification in Adults: An exploratory study
Epilepsy & Behavior · 2025
DOI: 10.1016/j.yebeh.2025.110364 ↗Kurz gefasst
Fachartikel in der Zeitschrift Epilepsy & Behavior, 2025. Verfasst von Francesco Brigo und 7 weiteren Autor:innen. Beteiligte Einrichtung: Innovation, Research and Teaching Service, Teaching Hospital of the Paracelsus Medical Private University (PMU), Azienda Sanitaria dell'Alto Adige, B…. Hauptthemen laut MEDLINE: Artificial Intelligence, Epilepsy, Seizures, Clinical Competence. Die Arbeit wurde bisher 15-mal zitiert (Crossref) und nennt 16 Literaturangaben.
Abstract (Originalsprache)
AIMS: Artificial intelligence (AI) tools like ChatGPT hold promise for enhancing diagnostic accuracy and efficiency in clinical practice. This exploratory study evaluates ChatGPT's performance in diagnosing and classifying epileptic seizures, epilepsy, and underlying etiologies in adult patients compared to epileptologists and neurologists. METHODS: A prospective simulation study assessed 37 clinical vignettes based on real adult patient cases. ChatGPT was 'trained' using official ILAE documents on epilepsy diagnosis and classification. Diagnoses and classifications by ChatGPT, two epileptologists, and two neurologists were compared against a reference standard set by a senior epileptologist. Diagnostic accuracy was evaluated using sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). Cohen's kappa (κ) was calculated to assess agreement. RESULTS: ChatGPT demonstrated high sensitivity (≥96.9 %) in identifying epileptic seizures and diagnosing epilepsy, ensuring no cases were missed. However, its specificity was lower, particularly for distinguishing acute symptomatic from unprovoked seizures (33.3 %) and diagnosing epilepsy (26.7 %), leading to frequent false positives. ChatGPT excelled in diagnosing epileptic syndromes (κ = 1.00) and structural etiologies (accuracy = 90.0 %) but struggled with ambiguous cases such as unknown seizure onset (accuracy = 12.5 %) and rare etiologies. Human experts consistently outperformed ChatGPT with near-perfect accuracy and higher κ values. CONCLUSION: ChatGPT shows potential as a supplementary diagnostic tool but requires human oversight due to reduced specificity and limitations in nuanced clinical judgment. Further development with diverse datasets and targeted training is necessary to improve AI performance. Integrating AI with expert clinicians can optimize diagnostic workflows in epilepsy care.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Innovation, Research and Teaching Service, Teaching Hospital of the Paracelsus Medical Private University (PMU), Azienda Sanitaria dell'Alto Adige, Bolzano-Bozen, Italy. Electronic address: francesco.brigo@sabes.it.
- Neurology and Stroke Unit, ASST Sette Laghi, Varese, Italy.
- Epilepsy Center, Neurology Unit, Department of Translational Medicine, University of Piemonte Orientale and Azienda Ospedaliero-Universitaria "Maggiore della Carità", Novara, Italy.
- Hospital of Gorizia, Department of Neurology, ASUGI, Gorizia, Italy.
- Innovation, Research and Teaching Service, Teaching Hospital of the Paracelsus Medical Private University (PMU), Azienda Sanitaria dell'Alto Adige, Bolzano-Bozen, Italy.
- Department of Internal Medicine, Intermediate Care Unit, Hospital Alto Vicentino (AULSS-7), Santorso, Italy.
Zitieren
APA
Brigo, F., Broggi, S., Strigaro, G., Olivo, S., Tommasini, V., Massar, M., Turcato, G., & Zaboli, A. (2025). Artificial intelligence (ChatGPT 4.0) vs. Human expertise for epileptic seizure and epilepsy diagnosis and classification in Adults: An exploratory study. Epilepsy & Behavior. https://doi.org/10.1016/j.yebeh.2025.110364
Vancouver
Brigo F, Broggi S, Strigaro G, Olivo S, Tommasini V, Massar M, et al. Artificial intelligence (ChatGPT 4.0) vs. Human expertise for epileptic seizure and epilepsy diagnosis and classification in Adults: An exploratory study. Epilepsy & Behavior. 2025. doi:10.1016/j.yebeh.2025.110364
BibTeX
@article{Brigo2025,
author = {Brigo, Francesco and Broggi, Serena and Strigaro, Gionata and Olivo, Sasha and Tommasini, Valentina and Massar, Magdalena and Turcato, Gianni and Zaboli, Arian},
title = {Artificial intelligence (ChatGPT 4.0) vs. Human expertise for epileptic seizure and epilepsy diagnosis and classification in Adults: An exploratory study},
journal = {Epilepsy & Behavior},
year = {2025},
doi = {10.1016/j.yebeh.2025.110364}
}
EUVIMED