Related Experiment Video
Updated: May 5, 2026
13:19
Deep Neural Networks for Image-Based Dietary Assessment
Published on: March 13, 2021
9.0K
DentAge: Deep learning for automated age prediction using panoramic dental X-ray images
1Laboratory of Imaging Technologies, Faculty of Electrical Engineering, University of Ljubljana, Ljubljana, Slovenia.
Journal of Forensic Sciences
|September 18, 2024
Summary
A new deep-learning model, DentAge, accurately predicts age using dental X-rays, achieving a 3.12-year error. This tool shows promise for forensic science and anthropology applications.
Area of Science:
- Dentistry
- Artificial Intelligence
- Forensic Anthropology
Background:
- Accurate age estimation is vital in forensic science and anthropology.
- Dental X-rays offer a rich source of data for age prediction.
- Current methods for age estimation from dental images can be time-consuming and subjective.
Purpose of the Study:
- To develop and validate DentAge, a deep-learning model for automated age prediction.
- To assess the model's performance across various age groups and dental conditions.
- To provide a publicly accessible tool for age estimation in dentistry.
Main Methods:
- A dataset of 21,007 panoramic dental X-ray images from subjects aged 4-97 years was used.
- Transfer learning with ImageNet weights was applied, followed by fine-tuning.
- Stochastic gradient descent with momentum and mean absolute error (MAE) were employed for training.
Main Results:
- DentAge achieved a mean absolute error (MAE) of 3.12 years on the test dataset.
- Performance varied by age group, with MAEs from 1.94 years (10-20) to 13.40 years (90-100).
- Factors like prosthetics and tooth loss influenced prediction accuracy.
Conclusions:
- DentAge demonstrates significant advancement in automated age prediction using dental radiography.
- The model shows robust performance across diverse demographics and dental states.
- Public accessibility will facilitate further validation and real-world application.
