Samira Asghari Moghaddam1, Morteza Yadekar2,

Arman Saeedi Vahdat1, Farzad Esmaeili1


1 - Postgraduate Student, Department of Oral and Maxillofacial Radiology, Faculty of Dentistry, Tabriz University of Medical Sciences.

2 - Department of Electronic Engineering, Tabriz Branch, Islamic Azad University. Tabriz, Iran.


urpose. Convolutional neural networks (CNNs) have extensive medical applications, such as the detection and diagnosis of diseases and clinical disorders. This study used CNN to detect mandibular fractures (MFs) on panoramic radiographs.

Methods. This study evaluates 275 panoramic radiographs retrieved from the archives of the Oral and Maxillofacial Radiology Department of Tabriz School of Dentistry. From all of the radiographs, 124 have MFs, and 151 have no fractures. In our methodology, first, the location of MFs was detected and marked on the radiographs by two oral and maxillofacial radiologists. Next, noise reduction was performed using the Chebyshev type II filter. To standardize the images, their primary resolution was modified and converted to 227 x 227. Next, the AlexNet CNN was used to train and classify images with and without MF. The images were randomly partitioned into training, validation, and test groups, where 60% were used for network training, 20% for validation, and 20% for final testing. The precision, recall, and F1 score were measured to assess this algorithm's efficacy for detecting MFs.

Results. The algorithm's precision, recall, and F1 score for detection of MFs are 0.968, 0.834, and 0.896, respectively.

Discussions. Mandibular fractures can be detected by panoramic radiographs and using CNN increase the accuracy of diagnoses, reach more definite diagnosis, and accelerate the reading and interpretation of radiographs.

Conclusion. The suggested algorithm successfully detects MFs on panoramic radiographs with high accuracy. Therefore, CNN-based models enhance the detection of MFs on panoramic radiographs.


Keywords: convolution neural network, mandibular fracture, panoramic radiography, image processing, deep learning.


Corresponding author: Arman Saeedi Vahdat, e-mail: Этот e-mail адрес защищен от спам-ботов, для его просмотра у Вас должен быть включен Javascript

For citation: Samira Asghari Moghaddam, Morteza Yadekar, Arman Saeedi Vahdat, Farzad Esmaeili. Automatic detection of mandibular fractures on panoramic radiographs using the convolutional neural network. REJR 2023; 13(3):5-13. DOI: 10.21569/2222-7415-2023-13-3-5-13.

Received:        14.02.23 Accepted:       08.08.23