Page 86 - AIH-1-3

Artificial Intelligence in Health

ORIGINAL RESEARCH ARTICLE
Dental cavity analysis, prediction, localization,

and quantification using computer vision

Mohammad Aqeel , Payam Norouzzadeh 2 , Abbas Maazallahi , Salih Tutun ,
1
3
1
Golnesa Rouie Miab , Laila Al Dehailan 5 , David Stoeckel , Eli Snir 3 , and
6
4
Bahareh Rahmani *
1
1 Computer Science, Saint Louis University, St. Louis, Missouri, United States of America
2 Professional Studies, Saint Louis University, St. Louis, Missouri, United States of America
3 Data Analytics Area, Olin Business School, Washington University in Saint Louis, St. Louis,
Missouri, United States of America
4 Pacific Dental Services, St. Louis, Missouri, United States of America
5 Department of Restorative Dental Sciences, College of Dentistry, Imam Abdulrahman Bin Faisal
University, Dammam, Saudi Arabia
6 Department of Dentistry, Saint Louis University, St. Louis, Missouri, United States of America

Abstract

Dental health assessment is a critical component of overall well-being, and
advancements in computer vision and deep learning have opened new avenues for
automating and enhancing this process. In this study, we present a comprehensive
approach to dental cavity analysis, spanning localization, quantification, and
visualization. Our methodology leveraged a diverse dataset of colored dental images
that had been meticulously augmented and annotated. The You Only Look Once
*Corresponding author:
Bahareh Rahmani model was employed for precise dental cavity localization, providing bounding box
(brahmani@slu.edu) predictions. Remarkably, these results were obtained based on images from standard
device cameras. Subsequently, we introduced the use of the segment anything
Citation: Aqeel M, Norouzzadeh P,
Maazallahi A, et al. Dental cavity model segmentation model, known for its zero-shot generalization capabilities, to
analysis, prediction, localization, focus on the exact areas of dental cavities. This approach enhanced the granularity
and quantification using computer of our analysis, providing dental professionals with detailed visualizations for precise
vision. Artif Intell Health.
2024;1(3):80-88. diagnosis. During the quantification phase, we extracted cavity areas from bounding
doi: 10.36922/aih.3184 box coordinates, enabling accurate measurement of cavity sizes. The model achieved
Received: March 15, 2024 a notable mean average precision of 0.732, an accuracy of 0.789, and a recall of 0.701.
Moreover, the model converged quickly, with most metrics achieving near-optimal
Accepted: May 14, 2024
results after 100 iterations. This quantitative data augments traditional diagnosis
Published Online: July 24, 2024 methods, facilitating more informed treatment decisions.
Copyright: © 2024 Author(s).
This is an Open-Access article
distributed under the terms of the Keywords: You only look once; Segment anything model; Segmentation model; Dental
Creative Commons Attribution cavity
License, permitting distribution,
and reproduction in any medium,
provided the original work is
properly cited.
1. Introduction
Publisher’s Note: AccScience
Publishing remains neutral with Many may not be aware that our oral well-being can provide insights into our overall
regard to jurisdictional claims in
published maps and institutional health. The truth is that problems in our mouth can potentially impact the rest of our
affiliations. body. Similar to other parts of our body, our mouths harbor mostly harmless bacteria.

Volume 1 Issue 3 (2024) 80 doi: 10.36922/aih.3184