Reliability of the artificial intelligence (ChatGPT) in the management of tibial plateau fracture
Published: 5 March 2025| Version 1 | DOI: 10.17632/46ff95x62f.1
Contributors:
, , , , , , Francisco Rivas-Ruiz, Description
The database is designed to analyze the concordance in the radiographic interpretation of tibial plateau fractures among three groups of evaluators: orthopedic and trauma surgeons (3), residents in training (3), and an artificial intelligence model (ChatGPT 4.0). The objective is to compare diagnostic accuracy and fracture classification in relation to a gold standard based on advanced imaging studies such as computed tomography (CT).
Files
Steps to reproduce
The anonymized database contains the classification of fracture presence and severity (grades 1-6) according to the gold standard (CT), along with the assessments from three medical residents and three attending physicians.
Institutions
- Hospital Costa del SolAndalucía, Marbella
Categories
Artificial Intelligence, Fracture, Clinical Decision Support System, Traumatology