Comparative Evaluation of Artificial Intelligence and Endoscopists' Accuracy in Endoscopic Ultrasound for Identifying Normal Anatomical Structures: A Multi-Institutional Cross-Sectional Study.

Autor: Robles-Medranda, C., Baquerizo-Burgos, J., Puga-Tejada, M., Raijman, I., Berzin, T., Nebel, J., Iglesias-Garcia, J., Kunda, R., Del Valle, R., Alcivar-Vasquez, J., Mendez, J. C., Chilan-Pincay, A., Sanchez-Cepeda, M., Lara, G., Oregel-Aguilar, V., Boston, I., Pattni, C., Egas-Izquierdo, M., Cunto, D., Arevalo-Mora, M.
Předmět:
Zdroj: Endoscopy; 2024 Supplement 2, Vol. 56, pS174-S174, 1p
Abstrakt: This article discusses a study that compares the diagnostic accuracy of an artificial intelligence (AI) model with that of expert and nonexpert gastrointestinal endoscopists in identifying normal anatomical structures during endoscopic ultrasound (EUS) procedures. EUS is a valuable tool in gastroenterology, but its accuracy can vary depending on the operator. The AI model, called AIWorks-EUS, achieved higher diagnostic accuracy than both expert and nonexpert endoscopists in identifying normal anatomical structures. The study suggests that this AI model could provide assistance during live procedures and may have potential applications in EUS training programs. [Extracted from the article]
Databáze: Complementary Index