Panoptic Segmentation on Panoramic Radiographs: Deep Learning-Based Segmentation of Various Structures Including Maxillary Sinus and Mandibular Canal.

Jun-Young Cha,In-Sung Yeo,Hyung-In Yoon,Kyung-Hoe Huh,Jung-Suk Han

doi:10.3390/jcm10122577

Abstract

Panoramic radiographs, also known as orthopantomograms, are routinely used in most dental clinics. However, it has been difficult to develop an automated method that detects the various structures present in these radiographs. One of the main reasons for this is that structures of various sizes and shapes are collectively shown in the image. In order to solve this problem, the recently proposed concept of panoptic segmentation, which integrates instance segmentation and semantic segmentation, was applied to panoramic radiographs. A state-of-the-art deep neural network model designed for panoptic segmentation was trained to segment the maxillary sinus, maxilla, mandible, mandibular canal, normal teeth, treated teeth, and dental implants on panoramic radiographs. Unlike conventional semantic segmentation, each object in the tooth and implant classes was individually classified. For evaluation, the panoptic quality, segmentation quality, recognition quality, intersection over union (IoU), and instance-level IoU were calculated. The evaluation and visualization results showed that the deep learning-based artificial intelligence model can perform panoptic segmentation of images, including those of the maxillary sinus and mandibular canal, on panoramic radiographs. This automatic machine learning method might assist dental practitioners to set up treatment plans and diagnose oral and maxillofacial diseases.

Highlights

Deep convolutional neural networks, one of the major examples of deep learning, aid in image classification and in predicting the location of the region of interest (RoI) in images
Studies on applying convolutional neural networks (CNNs) to radiographs are increasing rapidly in the field of dentistry, and many of them deal with panoramic radiographs [7]
Unlike many previous studies that mainly focused only on teeth segmentation [8,12], the present study examined whether the maxillary sinus and mandibular canal can be identified on dental panoramic radiographs using a deep neural network model

Summary

Introduction

One of the major examples of deep learning, aid in image classification and in predicting the location of the region of interest (RoI) in images. Several studies have sought to apply these deep neural networks on medical images; the application of semantic segmentation has been attempted in some of these studies [1,2,3] where each pixel in the image is classified. Other studies have tried methods such as object detection [4,5] or instance segmentation [6]. The application of semantic segmentation [8], object detection [9,10,11], and instance segmentation [12] on dental panoramic radiographs (orthopantomogram) has been reported. To the best of our knowledge, the use of a machine learning method for this kind of task, including deep neural networks, has not been studied in the dental field so far

Objectives

Methods

Conclusion