Fig 1 - uploaded by Luciano Oliveira
Content may be subject to copyright.
Contexts in source publication
Context 1
... these two categories, there are three types of dental X-rays that are used most often in dental examinations: Extraoral panoramic radiography -also called panoramic X-ray or orthopantomography; intraoral bitewing radiography -or bitewing Xray; and periapical intraoral radiography or only periapical X- rays. Figure 1 illustrates examples of these X-ray image types. Particularly, panoramic X-ray is a useful exam to complement the clinical examination in the diagnosis of dental diseases (caries or endodontic diseases). ...
Context 2
... these two categories, there are three types of dental X-rays that are used most often in dental examinations: Extraoral panoramic radiography -also called panoramic X-ray or orthopanto- mography; intraoral bitewing radiography -or bitewing X- ray; and periapical intraoral radiography or only periapical X- rays. Figure 1 illustrates examples of these X-ray image types. Particularly, panoramic X-ray is a useful exam to complement the clinical examination in the diagnosis of dental diseases (caries or endodontic diseases). ...
Similar publications
Citations
... The application of deep learning to dental reconstruction has evolved significantly over time. Initial efforts focused on more fundamental tasks, such as the accurate segmentation and identification of individual teeth within PXs [25,26]. These studies demonstrated the high accuracy that deep learning models could achieve in these tasks, paving the way for more complex applications. ...
Dental diagnosis relies on two primary imaging modalities: panoramic radiographs (PX) providing 2D oral cavity representations, and Cone-Beam Computed Tomography (CBCT) offering detailed 3D anatomical information. While PX images are cost-effective and accessible, their lack of depth information limits diagnostic accuracy. CBCT addresses this but presents drawbacks including higher costs, increased radiation exposure, and limited accessibility. Existing reconstruction models further complicate the process by requiring CBCT flattening or prior dental arch information, often unavailable clinically. We introduce ViT-NeBLa, a vision transformer-based Neural Beer-Lambert model enabling accurate 3D reconstruction directly from single PX. Our key innovations include: (1) enhancing the NeBLa framework with Vision Transformers for improved reconstruction capabilities without requiring CBCT flattening or prior dental arch information, (2) implementing a novel horseshoe-shaped point sampling strategy with non-intersecting rays that eliminates intermediate density aggregation required by existing models due to intersecting rays, reducing sampling point computations by , (3) replacing CNN-based U-Net with a hybrid ViT-CNN architecture for superior global and local feature extraction, and (4) implementing learnable hash positional encoding for better higher-dimensional representation of 3D sample points compared to existing Fourier-based dense positional encoding. Experiments demonstrate that ViT-NeBLa significantly outperforms prior state-of-the-art methods both quantitatively and qualitatively, offering a cost-effective, radiation-efficient alternative for enhanced dental diagnostics.
... Very few authors performed experiments on panoramic X-ray images, threshold-based [7,8,9], region-based [10], cluster-based [11], boundary-based [12,13,14] and one-class segmentation [15,16,17]. In the paper on deep instance segmentation of teeth on panoramic radiographs [18] the authors investigated teeth segmentation on the UFBA-UESC dental images dataset using a Mask R-CNN [19] for instance segmentation task, but all teeth were classified into a single category ignoring independent tooth recognition. Silva et al. [20] proposed different deep learning architectures PANet, HTC, ResNeSt, and Mask R-CNN, in [21], and they used Mask R-CNN with different segmentation heads pointRend [22] and FCN [23] and obtained excellent results. ...
Teeth segmentation and recognition play a vital role in a variety of dental applications and diagnostic procedures. The integration of deep learning models has facilitated the development of precise and automated segmentation methods. Although prior research has explored teeth segmentation, not many methods have successfully performed tooth segmentation and detection simultaneously. This study presents UFBA-425, a dental dataset derived from the UFBA-UESC dataset, featuring bounding box and polygon annotations for 425 panoramic dental X-rays. Additionally, this work introduces OralBBNet, an architecture featuring distinct segmentation and detection heads as U-Net and YOLOv8, respectively. OralBBNet is designed to improve the accuracy and robustness of tooth classification and segmentation on panoramic X-rays by leveraging the complementary strengths of U-Net and YOLOv8. Our approach achieved a 1-3% improvement in mean average precision (mAP) for teeth detection compared to existing techniques and a 15-20% improvement in the dice score for teeth segmentation over U-Net over various tooth categories and 2-4% improvement in the dice score when compared with other segmentation architectures. The results of this study establish a foundation for the wider implementation of object detection models in dental diagnostics.
... XX, XXXX 2025 each dental object while facilitating interactions among all objects via edges for global information modelling [18], and thus have been widely explored for labeling the threedimensional surfaces of teeth [16], [19]. Although such DLbased methods have been frequently applied to in dentistry such as tooth segmentation [13], [20], tooth numbering [11], detection of dental issues [21], [22] of instances of teeth and dental restorations, to the best of our knowledge, none of them has been developed for automated diagnosis of malocclusion issues. ...
... Alternatively, Anantharaman et al. [31] proposed a Mask R-CNN [32] based approach for the detection and segmentation of oral cold sores and canker sores. Jader et al. [20] applied Mask R-CNN for the first time on panoramic radiographic images for automatic detection and teeth instance segmentation. Although CNN-based methods have made significant progress, they still share a major drawback: limited ability to handle complex deformations and multi-scale features. ...
Malocclusion is a major challenge in orthodontics, and its complex presentation and diverse clinical manifestations make accurate localization and diagnosis particularly important. Currently, one of the major shortcomings facing the field of dental image analysis is the lack of large-scale, accurately labeled datasets dedicated to malocclusion issues, which limits the development of automated diagnostics in the field of dentistry and leads to a lack of diagnostic accuracy and efficiency in clinical practice. Therefore, in this study, we propose the Oral and Maxillofacial Natural Images (OMNI) dataset, a novel and comprehensive dental image dataset aimed at advancing the study of analyzing dental images for issues of malocclusion. Specifically, the dataset contains 4166 multi-view images with 384 participants in data collection and annotated by professional dentists. In addition, we performed a comprehensive validation of the created OMNI dataset, including three CNN-based methods, two Transformer-based methods, and one GNN-based method, and conducted automated diagnostic experiments for malocclusion issues. The experimental results show that the OMNI dataset can facilitate the automated diagnosis research of malocclusion issues and provide a new benchmark for the research in this field. Our OMNI dataset and baseline code are publicly available at https://github.com/RoundFaceJ/OMNI.
... Studies conducted recently showed that CNN models outperform traditional imaging approaches when identifying early lesions as well as secreted cavities and secondary caries in restored regions. [18] The automated features enabled by DL both improve diagnostic speed and enable doctors to make better decisions through their diagnostic system that provides second opinions. ...
Early diagnosis and carious lesion detection through artificial intelligence (AI) have transformed current standard methodologies because it generates precise results which work more efficiently and dependably. AI uses machine learning and deep learning technologies with computer-aided diagnostic systems to accomplish exceptional image evaluation of radiographic data and clinical records in dental caries detection through intraoral scans. This review discusses both prevailing challenges which limit and potential future uses of AI in dental diagnosis together with its ability to become a part of standard clinical work routines. Various researchers confirmed that AI works as a helpful tool which supports dental experts by improving diagnosis and minimizing human biases to enhance preventive care effects for patients.
... Recent advancements in smart diagnosis across various image modalities are assisting in earlier diagnosis of the alignments [3]. Machine intelligent approaches are also used in dentistry for smart diagnosis and as assistive tools for clinicians to recognize caries from X-ray images [4]. The conventional approaches are faced various challenges, which are listed as the technical gaps. ...
Accurate recognition and segmentation of dental caries in 2D panoramic X-ray images are crucial for timely diagnosis and strategic treatment planning. The current study uses a Generative Adversarial Network (GAN) model named SegAN to segment the X-ray samples to recognize the caries. The SegAN model works in an adversarial architecture in which the generator focuses on creating precise segmentation maps of caries from 2D panoramic X-ray images. On the other hand, the discriminator ensures that the output matches realistic segmentation patterns. The SegAN model efficiently handles the local and global contextual information for precise segmentation by considering Pixel-Wise and structural loss measures that assist in better segmentation of complex structures. Moreover, the SegAN model efficiently deals with noisy data and effectively handles class imbalances. Data augmentation, like histogram equalization and affine transforms, is performed on the input images for precise segmentation of the samples. The model was evaluated on both raw and preprocessed dental X-ray images using standard quantitative metrics. SegAN demonstrated superior performance compared to traditional segmentation approaches, achieving an accuracy of 98.5% and a dice coefficient of 0.936 in caries detection.
... Their model can accurately segment each tooth in panoramic X-rays. The model trained on 193 annotated images of 32 teeth, achieved 98% accuracy and 94% precision and also obtained 84% recall and 99% specificity on 1,224 images not previously encountered, [14]. ...
Panoramic radiography is a commonly used imaging technique for dental X-rays, it is used as a diagnostics tool in dentistry. The study introduced a hybrid deep learning approach for detecting and segmenting dental restorative elements from panoramic dental X-rays. By integrating the You Look Only Once (YOLO v8) model for object detection and the Segment Anything Model (SAM) for segmentation, the aim is to enhance the identification of different dental restorative elements such as dental implants, crowns, fillings, and root canals. The datasets of the study comprised 1290 dental X-ray images. The YOLO model effectively recognizes regions of interest and generates bounding boxes and then for achieving precise segmentation SAM is utilized. The results demonstrate high accuracy for classification rates of 95% for fillings, 88% for crowns, 93% for root canals, and 97% for implants and the Intersection over Union (IoU) metrics results also improve systems accuracy. The results show significant improvement in accuracy and highlight the effectiveness of the hybrid approach in refining diagnostic precision and enhancing efficiency in dental imaging.
... Furthermore, the specific jaw curvature and posture of patients may result in differences in the generated image (Choi 2011, Top 2023. The NN training of panoramic radiographs presents a unique set of difficulties due to their comprehensive nature, capturing not only the teeth but also the chin, spine, and jaws, as mentioned in (Jader et al. 2018). As a result, it becomes difficult to set just one fixed position that applies to the entire dataset for the intended Region of Interest (RoI). ...
Auto-cropping, the process of automatically adjusting the boundaries of an image to focus on the region of interest, is crucial to improving the diagnostic quality of dental panoramic radiographs. Its importance lies in its ability to standardize the size of different input images with minimal loss of information, thus ensuring consistency and improving the performance of subsequent image-processing tasks. Despite the widespread use of CNNs in many studies, research on auto-cropping for different-sized images remains limited. This study aims to explore the potential of differentiable auto-cropping in dental panoramic radiographs. A unique dataset of 20,973 dental panoramic radiographs, mostly with a resolution of 2836×1536 or close, divided into five classes by 3 dentists, was used, which is the same dataset from the previous study (Top et al. 2023). ResNet-101 model, which was the most successful network for the dataset (Top et al. 2023), was used for the evaluation. To reduce variance, the model was evaluated using 10-fold cross-validation for both non-auto-cropped and auto-cropped trainings. Data augmentation was also used to produce more accurate and robust results. For auto-cropped training, it was adjusted to be much less effective than the non-auto-cropped one. Accuracy was improved by 1.8%, from 92.7% to 94.5%, thanks to the proposed auto-crop optimization developed to reduce dataset-related issues. Its macro-average AUC was also raised from 0.989 to 0.993. The proposed auto-crop optimization can be implemented as a trainable network layer in an end-to-end CNN and can be used for other problems as well. Increasing the accuracy from 92.7% to 94.5% is a very challenging task due to diminishing returns, as there is little room for improvement. The results show the potential of the proposed differentiable auto-crop algorithm and encourages its use in different fields.
... Among them, Mask R-CNN (Mask Region-based Convolutional Neural Network) [15] has attracted much attention for its excellent high accuracy, high scalability, multitask learning ability and migratory nature. Jader et al. [16] proposed the first system capable of detecting and segmenting each tooth in a panoramic radiograph image. Pinheiro et al. [17] investigated and compared two Mask R-CNN-based schemes to improve rough segmentation boundaries. ...
... In 2022, Chandrashekar et al. [18] introduced a collaborative learning model that improves learning performance by combining an independent tooth instance segmentation model, Mask R-CNN, with a recognition model, Faster R-CNN [19]. Lee et al. [20] differed from the approach of Jader et al. [16] in that each panoramic radiograph in the dental images used in this study generated multiple independently annotated mask images based on the number of teeth included. Zhao et al. [21], based on the instance segmentation model Mask R-CNN combined with the U-Net architecture, modified the segmentation branching to improve the segmentation effect. ...
Traditional tooth image analysis methods primarily focus on feature extraction from individual images, often overlooking critical tooth shape and position information. This paper presents a novel computer-aided diagnosis method, Collaborative learning with Mask Region-based Convolutional Neural Network (Co-Mask R-CNN), designed to enhance tooth image analysis by leveraging the integration of complementary information. First, image enhancement is employed to generate an edge-enhanced tooth edge image. Then, a collaborative learning strategy combined with Mask R-CNN is introduced, where the original and edge images are input simultaneously, and a two-stream encoder extracts feature maps from complementary images. By utilizing an attention mechanism, the output features from the two branches are dynamically fused, quantifying the relative importance of the two complementary images at different spatial positions. Finally, the fused feature map is utilized for tooth instance segmentation. Extensive experiments are conducted using a proprietary dataset to evaluate the effectiveness of Co-Mask R-CNN, and the results are compared against those of an alternative segmentation network. The results demonstrate that Co-Mask R-CNN outperforms the other networks in terms of both segmentation accuracy and robustness. Consequently, this method holds considerable promise for providing medical professionals with precise tooth segmentation results, establishing a reliable foundation for subsequent tooth disease diagnosis and treatment.
... One such need is that of large-scale, fully annotated datasets to enable training of AI models in dental radiographic diagnoses [ 1 ]. Researchers in the field of computer vision recognize this challenge, and over the years, several open-sourced OPG datasets have been made available to facilitate the advancement of AI [2][3][4][5][6]. ...
With the digitization of radiographs, vast amounts of data have become accessible, enabling the curation and development of extensive datasets. Among radiographic modalities, Orthopantomograms (OPGs) are widely utilized in clinical practice. The integration of automated diagnostic processes into routine clinical practice holds great potential as an adjunct for dentists.Various OPG datasets exist, however their limitations affect the robustness of Artificial Intelligence (AI) models trained on them. This paper introduces an OPG dataset specifically designed for training AI algorithms in teeth segmentation and numbering tasks. A key feature of this dataset is its dual annotation, which allows for individual tooth segmentation by class, as well as numbering according to the Fédération Dentaire Internationale system.This dual-annotated dataset enhances the existing pool of OPG datasets and can be leveraged for further training of pre-trained algorithms or the development of new ones. Moreover, it offers researchers to carry out annotations tailored to their respective research objectives, thereby facilitating the development of AI models capable of addressing diverse diagnostic tasks.
... Therefore, this type of image is more suitable in diagnoses of teeth diseases, in order to plan root treatment [26,27], in diagnosis of gum [28,29] and jaw bone [30] diseases. In addition, it is frequently used by dentists and oral surgeons in routine practice or for non-medical purposes such as age estimation [24] or for preprocessing tasks such as teeth numbering [4], classification [31] and segmentation [32]. The techniques of NN and AI can be applied to a variety of radiological studies, such as the periapical X-ray and the CBCT. ...
... Bitewing X-rays [37,49,54,77,[84][85][86] Caries detection (posterior initial proximal caries) Accuracy Panoramic X-rays [4,11,20,[22][23][24]26,27,30,[30][31][32]43,44,57,58,[60][61][62][63]66,68,69,89, Full visualization of jaw, such as tumors, teeth included, infections, post-accident fractures, temporomandibular joint disorders Captured outside the mouth which makes them more acceptable for the patient, they cause a lower infection rate, and lower radiation exposure, they are simple to apply and require less time but they are the most challenging type due to uneven lighting, the presence of noise and low resolution. ...
... Some approaches focused on the detection of caries in a large [37] or small dataset [36], whereas other suggested a treatment plan based on caries depth [147]. Moreover, teeth segmentation seems to be an effective preprocessing step for further dental disease diagnosis in 2D images [32] or/and 3D teeth models [18,134]. The teeth segmentation aids in distinguishing the teeth from other tissues (i.e., gums and jaw bones). ...
Background: Dental care has been transformed by neural networks, introducing advanced methods for improving patient outcomes. By leveraging technological innovation, dental informatics aims to enhance treatment and diagnostic processes. Early diagnosis of dental problems is crucial, as it can substantially reduce dental disease incidence by ensuring timely and appropriate treatment. The use of artificial intelligence (AI) within dental informatics is a pivotal tool that has applications across all dental specialties. This systematic literature review aims to comprehensively summarize existing research on AI implementation in dentistry. It explores various techniques used for detecting oral features such as teeth, fillings, caries, prostheses, crowns, implants, and endodontic treatments. AI plays a vital role in the diagnosis of dental diseases by enabling precise and quick identification of issues that may be difficult to detect through traditional methods. Its ability to analyze large volumes of data enhances diagnostic accuracy and efficiency, leading to better patient outcomes. Methods: An extensive search was conducted across a number of databases, including Science Direct, PubMed (MEDLINE), arXiv.org, MDPI, Nature, Web of Science, Google Scholar, Scopus, and Wiley Online Library. Results: The studies included in this review employed a wide range of neural networks, showcasing their versatility in detecting the dental categories mentioned above. Additionally, the use of diverse datasets underscores the adaptability of these AI models to different clinical scenarios. This study highlights the compatibility, robustness, and heterogeneity among the reviewed studies. This indicates that AI technologies can be effectively integrated into current dental practices. The review also discusses potential challenges and future directions for AI in dentistry. It emphasizes the need for further research to optimize these technologies for broader clinical applications. Conclusions: By providing a detailed overview of AI’s role in dentistry, this review aims to inform practitioners and researchers about the current capabilities and future potential of AI-driven dental care, ultimately contributing to improved patient outcomes and more efficient dental practices.
