Related Experiment Video
Updated: Mar 31, 2026
09:10
Digital Hybrid Model Preparation for Virtual Planning of Reconstructive Dentoalveolar Surgical Procedures
Published on: August 5, 2021
2.4K
Mobile Imaging-Based Machine Learning for Dental Caries, Sealants, and Fluorosis: Protocol for a Cross-Sectional
Sang Mok Park1, Semin Kwon1, Shaun G Hong1
1Weldon School of Biomedical Engineering, Purdue University, West Lafayette, IN, United States.
JMIR Research Protocols
|March 30, 2026
Summary
This study uses mobile health (mHealth) technology and AI to detect dental caries, sealants, and fluorosis in adolescents. This approach offers a scalable and affordable method for public health surveillance.
Area of Science:
- Dental Public Health
- Artificial Intelligence in Dentistry
- Mobile Health (mHealth)
Background:
- Public health surveillance of dental caries, sealants, and fluorosis is crucial but faces limitations in cost, accessibility, and scalability.
- Current assessment methods are often not feasible for large-scale or individual evaluations.
- Mobile health (mHealth) solutions for concurrent detection of these conditions remain underexplored for population-level studies.
Purpose of the Study:
- To develop and validate mHealth models using computer vision, machine learning, and deep learning for detecting caries lesions, identifying sealants, and quantifying fluorosis severity.
- To utilize smartphone and low-cost intraoral camera images for dental assessments.
- To establish standardized visual clinical examinations as the reference standard for model validation.
Main Methods:
- Utilizing a study population of approximately 1000 adolescents in Colorado with naturally high fluoride water levels.
- Employing standardized clinical dental examinations and imaging via intraoral cameras and smartphones.
- Developing supervised learning models with color correction, radiomic features, and neural network classifiers, validated against expert clinical assessments.
Main Results:
- Established standardized protocols for dental examination and imaging (intraoral and smartphone).
- Conducted pilot studies for logistics and examiner calibration, including mobile app and web-based training.
- Initiated data collection in May 2024, with data from ~300 participants as of January 2026.
Conclusions:
- Integration of computer vision and mHealth imaging enables cost-effective, scalable population-level dental health surveillance.
- This approach facilitates detection of caries, sealants, and fluorosis severity in adolescents.
- The protocol provides a methodological framework for mobile dental imaging data acquisition, processing, and analysis for epidemiological studies.
