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 analyze the accuracy of ChatGPT4o-mini’s answers to commonly asked patient questions on traditional orthodontics and compare the results when the questions are prompted in different languages (English and Spanish) and compare the grading results of the orthodontist evaluators (Private practice and Academia). Methods: In this cross-sectional content analysis, 53 questions related to traditional orthodontic treatment were posed to ChatGPT4o-mini separately in English and Spanish. The generated answers were recorded and then evaluated by four orthodontists, with two Spanish fluent orthodontists assigned to evaluate the Spanish results, and two English fluent orthodontists assigned to evaluate the English results. Each language group consisted of one orthodontist selected from private practice and one orthodontist selected from academia. Each question’s answer was graded on a modified four-point scale as written: 1: Objectively true; 2: Selected facts; 3: Minimal facts; 4: False, and then recorded. Prior to the grading process, a pilot was conducted in the form of a meeting between the four orthodontists to establish a unified perception of the grading system and provide a sample question with an example of each answer scale. Following grading, a Wilcoxon test was used to evaluate if a statistically significant difference existed in score by language and if a statistically significant difference existed between the grading of the orthodontists by professional background. Results: ChatGPT4o-mini exhibited a high accuracy in answering commonly asked patient questions on traditional orthodontic treatment. For the English results, the Mean (SD) = 1.23 (0.522) with the Median = 1.00. For the Spanish results, the Mean (SD) = 1.14 (0.424) with a Median = 1.00. No statistically significant difference was found between the scores of the two languages (p = 0.177). A statistically significant difference was found between the grading of the English orthodontists (p = 0.004). No statistically significant difference was found between the grading of the Spanish orthodontists (p = 0.135). Conclusions: ChatGPT4o-mini exhibited a high accuracy in responses to frequently asked patient questions on traditional orthodontic treatment for both English and Spanish. There was no significant difference between the accuracy of ChatGPT4o-mini’s responses in English versus Spanish. There was not a definitive result if there exists a difference between orthodontists’ grading based on professional background due to contradictory findings.

Disciplines

Dentistry

Keywords

Artificial Intelligence; ChatGPT; Orthodontics; Patient Questions

Download

Download Full Text (414 KB)

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