Student Theses, Dissertations and Capstones

Document Type

Thesis - NSU Access Only

Degree Name

Master of Science (M.S.) in Dentistry

Copyright Statement

All rights reserved. This publication is intended for use solely by faculty, students, and staff of Nova Southeastern University. No part of this publication may be reproduced, distributed, or transmitted in any form or by any means, now known or later developed, including but not limited to photocopying, recording, or other electronic or mechanical methods, without the prior written permission of the author or the publisher.

Department

College of Dental Medicine

First Advisor

Cristina Godoy

Date of original Performance / Presentation

January 2025

Publication Date / Copyright Date

1-1-2025

Abstract

Objectives: The objective of this study was to evaluate the performance of ChatGPT-4o mini in responding to frequently asked questions regarding orthodontic clear aligner therapy (CAT) in both English and Spanish. Additionally, the study aimed to compare the accuracy ratings assigned by U.S.-licensed orthodontists practicing in academia versus private practice settings. Methods: In this cross-sectional study, 56 questions about orthodontic CAT were generated using ChatGPT-4o mini in both English and Spanish. The model was then prompted to generate answers to each question in the corresponding language. Four orthodontists, self-reported as fluent in either English or Spanish, independently assessed the accuracy of the responses using a modified four-point scale. Statistical analysis was performed using the Wilcoxon signed-rank test to evaluate differences in accuracy between the English and Spanish responses and between orthodontists from academic versus private practice settings. Inter-rater reliability was established through a pilot assessment conducted prior to the formal grading process. Results: 56 questions were scored by each orthodontist. The mean scores of each orthodontist were as follows: English 1; 1.16 ( +/- 0.532), English 2; 1.29 ( +/- 0.456), Spanish 1; 1.13 ( +/- 0.334), Spanish 2; 1.32 ( +/- 0.664). All orthodontists had a median score of 1.00. When evaluating the results focusing on language, the mean score was 1.22 ( +/- 0.497) with a median score of 1.00. For Spanish, the mean score was 1.22 ( +/- 0.532) with a median score of 1.00. When comparing the accuracy of scores between orthodontists in their respective language, there was a significant difference found between English 1 and English 2 evaluators (p-value < 0.05) with a calculated effect size of |d| = 0.25. No significant difference was found between the Spanish-language evaluators. Conclusions: 1) ChatGPT-4o mini responses to commonly asked questions about orthodontic CAT were accurate. 2) No significant differences were observed in the accuracy of responses between languages. 3) A significant difference was found between the two English-speaking orthodontists but not between the two Spanish-speaking orthodontists.

Disciplines

Dentistry

Keywords

Aligner; Artificial; ChatGPT; Clear; Intelligence; Therapy

Download

Download Full Text (2.9 MB)

Download
To access this thesis/dissertation you must have a valid nova.edu OR mynsu.nova.edu email address and create an account for NSUWorks.

Free My Thesis

If you are the author of this work and would like to grant permission to make it openly accessible to all, please click the Free My Thesis button.

  Contact Author

Share

 
COinS