Deep Learning Application in Dental Caries Detection Using Intraoral Photos Taken by Smartphones

Abstract

A mobile-phone-based diagnostic tool, which most of the population can easily access, could be a game changer in increasing the number of examinations of people with dental caries. This study aimed to apply a deep learning algorithm in diagnosing the stages of smooth surface caries via smartphone images. Materials and methods: A training dataset consisting of 1902 photos of the smooth surface of teeth taken with an iPhone 7 from 695 people was used. Four deep learning models, consisting of Faster Region-Based Convolutional Neural Networks (Faster R-CNNs), You Only Look Once version 3 (YOLOv3), RetinaNet, and Single-Shot Multi-Box Detector (SSD), were tested to detect initial caries lesions and cavities. The reference standard was the diagnosis of a dentist based on image examination according to the International Caries Classification and Management System (ICCMS) classification. Results: For cavitated caries, YOLOv3 and Faster R-CNN showed the highest sensitivity among the four tested models, at 87.4% and 71.4%, respectively. The sensitivity levels of these two models were only 36.9 % and 26% for visually non-cavitated (VNC). The specificity of the four models reached above 86% for cavitated caries and above 71% for VNC. Conclusion: The clinical application of YOLOv3 and Faster R-CNN models for diagnosing dental caries via smartphone images was promising. The current study provides a preliminary insight into the potential translation of AI from the laboratory to clinical practice.

Classification of Lateral Incisor Teeth for Aesthetic Dentistry Using Intelligent Techniques

“…In addition to this, the resolution of the tooth image. x-ray films Resnet18, Resnet50 [6] 74% 226 near-infrared light transillumination Faster R-CNN + FCN [7] 89.7% 87 intraoral camera YOLOv3(NSC vs. VNC) [3] 60.7% 1902 Mobile camera Faster R-CNN (NSC vs. VNC) [3] 67.8% 1902 Mobile camera RetinaNet (NSC vs. VNC) [3] 65.7% 1902 Mobile camera SSD (NSC vs. VNC) [3] 68.8% 1902 Mobile camera GoogLeNet Inception v3 CNN network [9] 82% 3000 radiographic images SVM [10] 91.9% 10 Camera Euclidean distance [10] 77.3% 10 Camera Regional growth [11] 68% 1500 x-ray deep convolutional neural network [12] 81% 1740 radiography convolutional neural network [13] 74.9% 196 x-ray Statistical analysis [14] 74% 100 photography proximal lesions [15] 27.2% 217…”

“…As a result, the grin that was put is not the same as the smile that was taken on camera when the subject was moving. [2,3]. This examination of the grin takes into account all of the parameters for shape and form.…”

“…While algorithmic quantitative evaluation are discussed in Section V. Section VI concludes the paper. [3] The findings of an experiment to identify dental caries were given by M. T. Thanh, N. V. Toan, V. T. Nhu, N. T. Tra, C. N. Giap, and D. M. Nguyen. Applying Classical Esthetic Principles to Create a Predictable Illusion of Symmetry While Using Digital Technology.…”

Customized convolutional neural network to detect dental caries from radio visio graphy(RVG) images

“…AL-ghamdi et al [27] With 96 %, proposed a neural search architecture to classify x-ray pictures into cavity, filling, and implant. Thanh et al [28] used four deep learning models to detect cavities from intraoral photographs, including Faster region-based convolutional neural networks (Faster R-CNNs), You Only Look Once version 3 (YOLOv3), RetinaNet, and Single-shot multi-box detector (SSD), with sensitivity values of 87.4 % and 71.4 % for Faster RCNN and YOLOv3 respectively. According to the results of the survey, there are certain challenges in the classification of dental diseases.…”

Visual Diagnostics of Dental Caries through Deep Learning of Non-Standardised Photographs Using a Hybrid YOLO Ensemble and Transfer Learning Model

“…The use of mobile handheld devices, such as smartphones, has seen exponential growth in emerging economies, as smartphone ownership and network connectivity has substantially increased within the rural population [ 3 ]. As such, recent biomedical research has begun utilising the full functionalities of the sensors in the said devices, to provide affordable solutions to complex problems in remote dental healthcare [ 4 , 5 , 6 , 7 ].…”