Pre-trained VGG16 model for forensic dental age estimation
See where this sits in the topic map →Summary AI-generated
- TL;DR
- This study explores the practical application of machine learning in forensic odontology, specifically focusing on dental age estimation using a pre-trained VGG16 model.
- Problem
- Traditional manual age estimation methods are labor-intensive, and applying the VGG16 model to analyze and classify detailed tooth development stages faced limitations due to an insufficient sample of orthopantomograms (OPGs).
- Method
- The researchers utilized a sample of 876 OPGs from individuals aged 10 to 25 from the Portuguese population. Using the third molars (teeth 38 and 48), they trained the VGG16 convolutional neural network to classify individuals into age groups based on 16, 18, and 21-year thresholds, and compared its performance against traditional manual methods by Demirjian and by Moorrees, Fanning, and Hunt.
- Results
- When evaluated using overall accuracy, recall, precision, and F-score, the VGG16 model achieved over 90% overall accuracy on cropped images containing only the third molars.
- Contributions
- The work demonstrates the feasibility of using a CNN-based approach with cropped dental images to classify individuals into key age groups.
- Limitations
- The limited sample size hindered the model's ability to accurately differentiate between the finer nuances of the various tooth development stages defined by traditional methods.
- Takeaways
- Classifying age groups based on third molar development shows high accuracy, but enhancing the model's reliability for detailed developmental stages requires a larger and more diverse dataset.
- Applications
- Legal proceedings and age estimation for individuals without proper documentation, such as those seeking asylum or unaccompanied minors.
- Topics
- Machine Learning, Forensic Odontology, Age Estimation, Computer Vision, VGG16
- For industry
- Legal and Forensic Services
- Why it matters
- Improves the objectivity and efficiency of forensic age estimation to better protect the rights of individuals lacking proper documentation.
Abstract
Abstract Background The practical employment of Machine Learning in Forensic Odontology remains underexplored, especially in the field of age estimation. Age estimation is essential in legal proceedings to protect the rights of individuals without proper documentation, whether for seeking asylum or when caring for a found child. This study aimed to utilize the VGG16 model to read, analyze, and provide classification of tooth development stages of the third molars from a sample of 876 orthopantomograms (OPGs) 10 to 25 years old (447 males and 429 females) from the Portuguese population collected from the ULS Santa Maria, University of Lisbon. The third molars 38 and 48 were used to classify individuals into age groups based on thresholds of 16, 18, and 21 years old. Age estimation was calculated manually using the methods established by Demirjian and by Moorrees, Funning, and Hunt. Furthermore, we trained the VGG16 model to read, analyze, and provide classification of the development stages, and afterwards we evaluated the VGG16 model through overall accuracy, recall, precision, and F-Score. The goal was to compare the accuracy of the traditional age estimation methods established by Demirjian, and by Moorrees, Funning, and Hunt, with a CNN-based approach. Results The VGG16 model provided excellent results for cropped images containing only the third molars (38 and 48) and was able to capture the patterns and the features of development stages, so the overall accuracy obtained was greater than 90%. However, to analyze and classify the development stages defined by Demirjian and by Moorrees, Funning, and Hunt, the VGG16 model faced some limitations due to the insufficient sample of OPGs. Conclusion The classification of age groups based on the development of third molars 38 and 48 demonstrated promising results with a high degree of accuracy. However, the limited sample size hindered the VGG16 model's ability to accurately differentiate between the various stages of tooth development. To enhance the model's accuracy and reliability, a larger and more diverse dataset is necessary to better capture the nuances of each developmental stage.