Object detection on dental x-ray images using deep learning method

Abstract

Radiological examination has an important role in determining the diagnosis of dental problems and making decisions about the right type of treatment according to the case indications. Dental x-ray is a medical procedure for taking pictures of the inside of the mouth using radiation fluid, where the results are used diagnostically to help the dentist see the entire structure of the jaw bone and teeth, and dental problems that cannot be seen directly. Dental radiographic interpretation, which is generally performed by dentists, is a time-consuming and error-prone process due to high variations in tooth structure, low experience levels, and fatigue factors experienced by dentists. The workload of a dentist and the occurrence of misdiagnosis can be reduced by the existence of a system that can automatically interpret the x-ray results. To overcome these problems, a model will be developed to be able to detect objects in the dental panoramic x-ray images using Mask R-CNN, one of the methods in Deep Learning. Deep Learning is an artificial intelligence function that modelled the workings of human brain in processing data and creating patterns for use in decision making. With the detection of objects in panoramic x-ray image automatically, it is expected to save time, improve the quality of dental care, and also the quality of diagnosis made by dentists.

The application of artificial neural networks in the detection of mandibular fractures using panoramic radiography

“…[ 14 15 ] It has been reported that using an AI algorithm makes it possible to perform radiographic interpretation done by dentists. [ 16 ] Our aim was to develop a deep learning framework to detect and localize trauma and fractures in the mandible.…”

Deep learning architectures in dental diagnostics: a systematic comparison of techniques for accurate prediction of dental disease through x-ray imaging

“…In order to recognize objects from tooth panoramic images from x-rays, Suryani et al . (2021) created a model applying Mask R-CNN, which is a form of the DL approach.…”

“…Therefore, to increase the rate of detection, methods of imaging are constantly used (Michou et al ., 2022; Shah et al ., 2014), although dentists can fail to detect early tooth decay lesions even using dental imaging. As a result, the success of diagnosis and therapy rely on both modern technology and the reader's own abilities (Suryani et al ., 2021). In addition, evaluating the images from X-rays also increases the workload for doctors.…”

RDFNet: A Fast Caries Detection Method Incorporating Transformer Mechanism

“…Both methods demonstrated high accuracy on caries X-ray image datasets. Some studies, such as that of Suryani et al [ 2 ], have also directly used deep-learning methods for caries detection and identification, which saved time and process steps and improved the simplicity of the method.…”

“…With the advances of deep-learning methods in computer vision, many such methods have been applied to caries detection to improve the accuracy of detection and relieve the dentists' workload. For example, Suryani et al [ 2 ] used mask R-CNN to detect objects in dental panoramic X-ray images, which saved time and improved the quality of dentists' diagnoses by automatically detecting panoramic X-ray images. Majanga et al [ 3 ] proposed a deep learning-based technique for dental caries detection named blob detection.…”