Comparison of ChatGPT version 3.5 & 4 for utility in respiratory medicine education using clinical case scenarios - 24/04/24
Abstract |
Integration of ChatGPT in Respiratory medicine presents a promising avenue for enhancing clinical practice and pedagogical approaches. This study compares the performance of ChatGPT version 3.5 and 4 in respiratory medicine, emphasizing its potential in clinical decision support and medical education using clinical cases. Results indicate moderate performance highlighting limitations in handling complex case scenarios. Compared to ChatGPT 3.5, version 4 showed greater promise as a pedagogical tool, providing interactive learning experiences. While serving as a preliminary decision support tool clinically, caution is advised, stressing the need for ongoing validation. Future research should refine its clinical capabilities for optimal integration into medical education and practice.
Le texte complet de cet article est disponible en PDF.Keywords : Artificial Intelligence, ChatGPT, Clinical decision, Education, Pedagogy, Respiratory medicine
Plan
Vol 85
Article 101091- juin 2024 Retour au numéro
Article précédent
- High prevalence of circulating myositis-associated antibodies in non-COVID critical illness
- Alyssa Soskis, Mary B. Rice, Donald B. Bloch, Rachel K. Putman, Antonio Arciniegas Rubio, Katherin Zambrano Vera, Rene S. Bermea, Andrew J. Sauer, Claire O. Sinow, Max Shen, Mayra Pinilla Vera, Rebecca M. Baron, Robert W. Hallowell
- Combination of electromagnetic navigation and probe-based LASER endomicroscopy to guide non-solid nodules sampling: Results from the CELTICS 2 study
- Jules Kalifa, Gavin Plat, Aurélien Brindel, Valentin Héluain, Laurent Brouchet, Axel Rouch, Julien Mazières, Marie-Christelle Pajiep Chapda, Thomas Villeneuve, Nicolas Guibert
Bienvenue sur EM-consulte, la référence des professionnels de santé.
L’accès au texte intégral de cet article nécessite un abonnement.
Déjà abonné à cette revue ?