Determining the medical Spanish translation capabilities of three artificial intelligence translation models for Mohs micrographic surgical instructions
Journal
Journal of the American Academy of Dermatology
ISSN
0190-9622
Publisher
Elsevier Inc.
Date Issued
2024
Author(s)
Scheinkman, Ryan
Montoya, Sofia
Náder, Maria
Ramírez, Mariana
Barbato, Kristiana
Philippe, Jean-Pierre
Vignau, Alexia
Nouri, Keyvan
Type
Resource Types::text::journal::journal article
Abstract
To the Editor: Artificial intelligence (AI) has been used to simplify medical-legal documentation.1 In order to protect patients from mistranslations, it is critical to assess the accuracy of AI translations. We attempted to assess the current translational capacities of 3 AI models for Mohs micrographic surgery documentation. The purpose of this analysis was to see if these programs had capabilities that were comparable to human medical translators and determine their capacity for future medical translation applications. In order to determine the validity of these models, preoperative and postoperative instructions from multiple sources were translated by Google Translate, Amazon Translate, and DeepL to Spanish from 3 publicly available academic center websites, specifically: the University of Mississippi Medical Center (University of Mississippi), University of Rochester, and Brigham Cancer Center.2-5 Accuracy of translation was then assessed by 3 native Spanish-speaking medical professionals and students that received C-1 levels on the Test of English as a Foreign Language demonstrating advanced English proficiency. ©The authors © Journal of the American Academy of Dermatology ©Elsevier Inc.
License
Acceso Restringido
How to cite
Scheinkman, R., Montoya, S., Náder, M., Ramírez, M., Barbato, K., Jean-Pierre, P., Vignau, A., & Nouri, K. (2025). Determining the medical Spanish translation capabilities of three artificial intelligence translation models for Mohs micrographic surgical instructions. Journal of the American Academy of Dermatology, 92(2), 349–351. https://doi.org/10.1016/j.jaad.2024.09.070