3D Tooth Segmentation and Labeling Using Deep Convolutional Neural Networks

Efficient complete denture metal base design via a dental feature-driven segmentation network

BACKGROUND AND OBJECTIVE

Complete denture is a common restorative treatment in dental patients and the design of the core components (major connector and retentive mesh) of complete denture metal base (CDMB) is the basis of successful restoration. However, the automated design process of CDMB has become a challenging task primarily due to the complexity of manual interaction, low personalization, and low design accuracy.

METHODS

To solve the existing problems, we develop a computer-aided Segmentation Network-driven CDMB design framework, called CDMB-SegNet, to automatically generate personalized digital design boundaries for complete dentures of edentulous patients. Specifically, CDMB-SegNet consists of a novel upright-orientation adjustment module (UO-AM), a dental feature-driven segmentation network, and a specific boundary-optimization design module (BO-DM). UO-AM automatically identifies key points for locating spatial attitude of the three-dimensional dental model with arbitrary posture, while BO-DM can result in smoother and more personalized designs for complete denture. In addition, to achieve efficient and accurate feature extraction and segmentation of 3D edentulous models with irregular gingival tissues, the light-weight backbone network is also incorporated into CDMB-SegNet.

RESULTS

Experimental results on a large clinical dataset showed that CDMB-SegNet can achieve superior performance over the state-of-the-art methods. Quantitative evaluation (major connector/retentive mesh) showed improved Accuracy (98.54 ± 0.58 %/97.73 ± 0.92 %) and IoU (87.42 ± 5.48 %/70.42 ± 7.95 %), and reduced Maximum Symmetric Surface Distance (4.54 ± 2.06 mm/4.62 ± 1.68 mm), Average Symmetric Surface Distance (1.45 ± 0.63mm/1.28 ± 0.54 mm), Roughness Rate (6.17 ± 1.40 %/6.80 ± 1.23 %) and Vertices Number (23.22 ± 1.85/43.15 ± 2.72). Moreover, CDMB-SegNet shortened the overall design time to around 4 min, which is one tenth of the comparison methods.

CONCLUSIONS

CDMB-SegNet is the first intelligent neural network for automatic CDMB design driven by oral big data and dental features. The designed CDMB is able to couple with patient's personalized dental anatomical morphology, providing higher clinical applicability compared with the state-of-the-art methods.

Automatic removal of soft tissue from 3D dental photo scans; an important step in automating future forensic odontology identification

The potential of intraoral 3D photo scans in forensic odontology identification remains largely unexplored, even though the high degree of detail could allow automated comparison of ante mortem and post mortem dentitions. Differences in soft tissue conditions between ante- and post mortem intraoral 3D photo scans may cause ambiguous variation, burdening the potential automation of the matching process and underlining the need for limiting inclusion of soft tissue in dental comparison. The soft tissue removal must be able to handle dental arches with missing teeth, and intraoral 3D photo scans not originating from plaster models. To address these challenges, we have developed the grid-cutting method. The method is customisable, allowing fine-grained analysis using a small grid size and adaptation of how much of the soft tissues are excluded from the cropped dental scan. When tested on 66 dental scans, the grid-cutting method was able to limit the amount of soft tissue without removing any teeth in 63/66 dental scans. The remaining 3 dental scans had partly erupted third molars (wisdom teeth) which were removed by the grid-cutting method. Overall, the grid-cutting method represents an important step towards automating the matching process in forensic odontology identification using intraoral 3D photo scans.

Deep learning-based tooth segmentation methods in medical imaging: A review

Deep learning approaches for tooth segmentation employ convolutional neural networks (CNNs) or Transformers to derive tooth feature maps from extensive training datasets. Tooth segmentation serves as a critical prerequisite for clinical dental analysis and surgical procedures, enabling dentists to comprehensively assess oral conditions and subsequently diagnose pathologies. Over the past decade, deep learning has experienced significant advancements, with researchers introducing efficient models such as U-Net, Mask R-CNN, and Segmentation Transformer (SETR). Building upon these frameworks, scholars have proposed numerous enhancement and optimization modules to attain superior tooth segmentation performance. This paper discusses the deep learning methods of tooth segmentation on dental panoramic radiographs (DPRs), cone-beam computed tomography (CBCT) images, intro oral scan (IOS) models, and others. Finally, we outline performance-enhancing techniques and suggest potential avenues for ongoing research. Numerous challenges remain, including data annotation and model generalization limitations. This paper offers insights for future tooth segmentation studies, potentially facilitating broader clinical adoption.

Deep learning for virtual orthodontic bracket removal: tool establishment and application

OBJECTIVE

We aimed to develop a tool for virtual orthodontic bracket removal based on deep learning algorithms for feature extraction from bonded teeth and to demonstrate its application in a bracket position assessment scenario.

MATERIALS AND METHODS

Our segmentation network for virtual bracket removal was trained using dataset A, containing 978 bonded teeth, 20 original teeth, and 20 brackets generated by scanners. The accuracy and segmentation time of the network were tested by dataset B, which included an additional 118 bonded teeth without knowing the original tooth morphology. This tool was then applied for bracket position assessment. The clinical crown center, bracket center, and orientations of separated teeth and brackets were extracted for analyzing the linear distribution and angular deviation of bonded brackets.

RESULTS

This tool performed virtual bracket removal in 2.9 ms per tooth with accuracies of 98.93% and 97.42% (P < 0.01) in datasets A and B, respectively. The tooth surface and bracket characteristics were extracted and used to evaluate the results of manually bonded brackets by 49 orthodontists. Personal preferences for bracket angulation and bracket distribution were displayed graphically and tabularly.

CONCLUSIONS

The tool's efficiency and precision are satisfactory, and it can be operated without original tooth data. It can be used to display the bonding deviation in the bracket position assessment scenario.

CLINICAL SIGNIFICANCE

With the aid of this tool, unnecessary bracket removal can be avoided when evaluating bracket positions and modifying treatment plans. It has the potential to produce retainers and orthodontic devices prior to tooth debonding.

Endoscopic versus laparoscopic bariatric procedures: A computational biomechanical study through a patient-specific approach

BACKGROUND AND OBJECTIVES

Within the framework of computational biomechanics, finite element models of the gastric district could be seen as a potential clinical tool not only to study the effects apported by bariatric surgery, but also to compare different surgical techniques such as the new emerging Endoscopic Sleeve Gastroplasty (ESG) with respect to well-established ones (such as the Laparoscopic Sleeve Gastrectomy, LSG).

METHODS

This work realized a fully computational comparison between the outcomes obtained from 10 patient-specific stomach models, which were used to simulate ESG, and the complementary results obtained from models representing the post-LSG of the same subjects. Specifically, once the ESG was simulated, a mechanical stimulus was applied by increasing an intragastric pressure up to a maximum of 5 kPa, in order to replicate the process of food intake, as well as for post-LSG models.

RESULTS

Results revealed non negligible differences between the techniques also within the same subject. In particular, not only LSG could lead to a greater reduction in the stomach volume (about 77 % at baseline, which is strictly linked to weight loss), but also influence the gastric distension (12 % less than pre-operative models). On the contrary, if ESG would be performed, a more similar pre-operative mechanical stimulation of the gastric walls may be seen (difference of about 1 %), thus preserving the mechanosensation, but the detriment of the volume reduction (about 56 % at baseline, and even decreases with increasing pressure). Moreover, since results suggested ESG may be more influenced by the pre-operative gastric cavity than LSG, a predictive model was proposed to support the surgical planning and the estimation of the volume reduction after ESG.

CONCLUSIONS

ESG and LSG have substantial differences in their protocols and post-surgical effects. This work pointed out that variations between the two procedures may be observed also from a computational point of view, especially when including patient-specific geometries. These insights support gastric modelling as a valuable tool to evaluate, design and critically compare emerging bariatric surgical procedures, not only from empirical aspects and clinical outcomes, but also from a mechanical point of view.

A fine-grained orthodontics segmentation model for 3D intraoral scan data

With the widespread application of digital orthodontics in the diagnosis and treatment of oral diseases, more and more researchers focus on the accurate segmentation of teeth from intraoral scan data. The accuracy of the segmentation results will directly affect the follow-up diagnosis of dentists. Although the current research on tooth segmentation has achieved promising results, the 3D intraoral scan datasets they use are almost all indirect scans of plaster models, and only contain limited samples of abnormal teeth, so it is difficult to apply them to clinical scenarios under orthodontic treatment. The current issue is the lack of a unified and standardized dataset for analyzing and validating the effectiveness of tooth segmentation. In this work, we focus on deformed teeth segmentation and provide a fine-grained tooth segmentation dataset (3D-IOSSeg). The dataset consists of 3D intraoral scan data from more than 200 patients, with each sample labeled with a fine-grained mesh unit. Meanwhile, 3D-IOSSeg meticulously classified every tooth in the upper and lower jaws. In addition, we propose a fast graph convolutional network for 3D tooth segmentation named Fast-TGCN. In the model, the relationship between adjacent mesh cells is directly established by the naive adjacency matrix to better extract the local geometric features of the tooth. Extensive experiments show that Fast-TGCN can quickly and accurately segment teeth from the mouth with complex structures and outperforms other methods in various evaluation metrics. Moreover, we present the results of multiple classical tooth segmentation methods on this dataset, providing a comprehensive analysis of the field. All code and data will be available at https://github.com/MIVRC/Fast-TGCN.

Connecting the dots towards precision orthodontics

Precision orthodontics entails the use of personalized clinical, biological, social and environmental knowledge of each patient for deep individualized clinical phenotyping and diagnosis combined with the delivery of care using advanced customized devices, technologies and biologics. From its historical origins as a mechanotherapy and materials driven profession, the most recent advances in orthodontics in the past three decades have been propelled by technological innovations including volumetric and surface 3D imaging and printing, advances in software that facilitate the derivation of diagnostic details, enhanced personalization of treatment plans and fabrication of custom appliances. Still, the use of these diagnostic and therapeutic technologies is largely phenotype driven, focusing mainly on facial/skeletal morphology and tooth positions. Future advances in orthodontics will involve comprehensive understanding of an individual's biology through omics, a field of biology that involves large-scale rapid analyses of DNA, mRNA, proteins and other biological regulators from a cell, tissue or organism. Such understanding will define individual biological attributes that will impact diagnosis, treatment decisions, risk assessment and prognostics of therapy. Equally important are the advances in artificial intelligence (AI) and machine learning, and its applications in orthodontics. AI is already being used to perform validation of approaches for diagnostic purposes such as landmark identification, cephalometric tracings, diagnosis of pathologies and facial phenotyping from radiographs and/or photographs. Other areas for future discoveries and utilization of AI will include clinical decision support, precision orthodontics, payer decisions and risk prediction. The synergies between deep 3D phenotyping and advances in materials, omics and AI will propel the technological and omics era towards achieving the goal of delivering optimized and predictable precision orthodontics.

Automatic Detection of Tooth-Gingiva Trim Lines on Dental Surfaces

Detecting the tooth-gingiva trim line from a dental surface plays a critical role in dental treatment planning and aligner 3D printing. Existing methods treat this task as a segmentation problem, which is resolved with geometric deep learning based mesh segmentation techniques. However, these methods can only provide indirect results (i.e., segmented teeth) and suffer from unsatisfactory accuracy due to the incapability of making full use of high-resolution dental surfaces. To this end, we propose a two-stage geometric deep learning framework for automatically detecting tooth-gingiva trim lines from dental surfaces. Our framework consists of a trim line proposal network (TLP-Net) for predicting an initial trim line from the low-resolution dental surface as well as a trim line refinement network (TLR-Net) for refining the initial trim line with the information from the high-resolution dental surface. Specifically, our TLP-Net predicts the initial trim line by fusing the multi-scale features from a U-Net with a proposed residual multi-scale attention fusion module. Moreover, we propose feature bridge modules and a trim line loss to further improve the accuracy. The resulting trim line is then fed to our TLR-Net, which is a deep-based LDDMM model with the high-resolution dental surface as input. In addition, dense connections are incorporated into TLR-Net for improved performance. Our framework provides an automatic solution to trim line detection by making full use of raw high-resolution dental surfaces. Extensive experiments on a clinical dental surface dataset demonstrate that our TLP-Net and TLR-Net are superior trim line detection methods and outperform cutting-edge methods in both qualitative and quantitative evaluations.

An artificial intelligence study: automatic description of anatomic landmarks on panoramic radiographs in the pediatric population

BACKGROUND

Panoramic radiographs, in which anatomic landmarks can be observed, are used to detect cases closely related to pediatric dentistry. The purpose of the study is to investigate the success and reliability of the detection of maxillary and mandibular anatomic structures observed on panoramic radiographs in children using artificial intelligence.

METHODS

A total of 981 mixed images of pediatric patients for 9 different pediatric anatomic landmarks including maxillary sinus, orbita, mandibular canal, mental foramen, foramen mandible, incisura mandible, articular eminence, condylar and coronoid processes were labelled, the training was carried out using 2D convolutional neural networks (CNN) architectures, by giving 500 training epochs and Pytorch-implemented YOLO-v5 models were produced. The success rate of the AI model prediction was tested on a 10% test data set.

RESULTS

A total of 14,804 labels including maxillary sinus (1922), orbita (1944), mandibular canal (1879), mental foramen (884), foramen mandible (1885), incisura mandible (1922), articular eminence (1645), condylar (1733) and coronoid (990) processes were made. The most successful F1 Scores were obtained from orbita (1), incisura mandible (0.99), maxillary sinus (0.98), and mandibular canal (0.97). The best sensitivity values were obtained from orbita, maxillary sinus, mandibular canal, incisura mandible, and condylar process. The worst sensitivity values were obtained from mental foramen (0.92) and articular eminence (0.92).

CONCLUSIONS

The regular and standardized labelling, the relatively larger areas, and the success of the YOLO-v5 algorithm contributed to obtaining these successful results. Automatic segmentation of these structures will save time for physicians in clinical diagnosis and will increase the visibility of pathologies related to structures and the awareness of physicians.

Measurement of proximal contact of single crowns to assess interproximal relief: A pilot study

Background

It is common for dental technicians to adjust the proximal surface of adjacent teeth on casts when fabricating single crowns. However, whether the accuracy of the proximal contact is affected if this step is eliminated is unclear.

Objective

To evaluate the accuracy of the proximal contact of single crowns for mandibular first molars fabricated from four different restorative materials, without adjustment of the proximal surface of the adjacent teeth by the laboratory/dental technician.

Methods

This study was in vitro; all the clinical procedures were conducted on a dentoform. The mandibular first molar tooth on the dentoform was prepared using diamond burs and a high speed handpiece. Twenty single crowns were fabricated, five for each group (monolithic zirconia, lithium disilicate, metal ceramic, and cast gold). No proximal surface adjacent to the definitive crowns was adjusted for tight contact in the dental laboratory. Both the qualitative analyses, using dental floss and shimstock, and the quantitative analyses, using a stereo microscope, were performed to evaluate the accuracy of the proximal contact of the restoration with the adjacent teeth. In the quantitative analysis, one-way analysis of variance was used to compare mean values at a significance level of 0.05.

Results

In quantitative analysis, the differences between the proximal contact tightness of the four groups was not statistically significant (P = 0.802 for mesial contacts, P = 0.354 for distal contacts). In qualitative analysis, in most crowns, dental floss passed through the contact with tight resistance and only one film of shimstock could be inserted between the adjacent teeth and the restoration. However, one specimen from the cast gold crown had open contact.

Conclusions

Even without proximal surface adjustment of the adjacent teeth during the crown fabrication process, adequate proximal contact tightness between the restoration and adjacent teeth could be achieved.

Denoised encoder-based residual U-net for precise teeth image segmentation and damage prediction on panoramic radiographs

OBJECTIVES

This research focuses on performing teeth segmentation with panoramic radiograph images using a denoised encoder-based residual U-Net model, which enhances segmentation techniques and has the capacity to adapt to predictions with different and new data in the dataset, making the proposed model more robust and assisting in the accurate identification of damages in individual teeth.

METHODS

The effective segmentation starts with pre-processing the Tufts dataset to resize images to avoid computational complexities. Subsequently, the prediction of the defect in teeth is performed with the denoised encoder block in the residual U-Net model, in which a modified identity block is provided in the encoder section for finer segmentation on specific regions in images, and features are identified optimally. The denoised block aids in handling noisy ground truth images effectively.

RESULTS

Proposed module achieved greater values of mean dice and mean IoU with 98.90075 and 98.74147 CONCLUSIONS: Proposed AI enabled model permitted a precise approach to segment the teeth on Tuffs dental dataset in spite of the existence of densed dental filling and the kind of tooth.

CLINICAL SIGNIFICANCE

The proposed model is pivotal for improved dental diagnostics, offering precise identification of dental anomalies. This could revolutionize clinical dental settings by facilitating more accurate treatments and safer examination processes with lower radiation exposure, thus enhancing overall patient care.

Deep learning-enabled 3D multimodal fusion of cone-beam CT and intraoral mesh scans for clinically applicable tooth-bone reconstruction

High-fidelity three-dimensional (3D) models of tooth-bone structures are valuable for virtual dental treatment planning; however, they require integrating data from cone-beam computed tomography (CBCT) and intraoral scans (IOS) using methods that are either error-prone or time-consuming. Hence, this study presents Deep Dental Multimodal Fusion (DDMF), an automatic multimodal framework that reconstructs 3D tooth-bone structures using CBCT and IOS. Specifically, the DDMF framework comprises CBCT and IOS segmentation modules as well as a multimodal reconstruction module with novel pixel representation learning architectures, prior knowledge-guided losses, and geometry-based 3D fusion techniques. Experiments on real-world large-scale datasets revealed that DDMF achieved superior segmentation performance on CBCT and IOS, achieving a 0.17 mm average symmetric surface distance (ASSD) for 3D fusion with a substantial processing time reduction. Additionally, clinical applicability studies have demonstrated DDMF's potential for accurately simulating tooth-bone structures throughout the orthodontic treatment process.

Hierarchical Self-Supervised Learning for 3D Tooth Segmentation in Intra-Oral Mesh Scans

Accurately delineating individual teeth and the gingiva in the three-dimension (3D) intraoral scanned (IOS) mesh data plays a pivotal role in many digital dental applications, e.g., orthodontics. Recent research shows that deep learning based methods can achieve promising results for 3D tooth segmentation, however, most of them rely on high-quality labeled dataset which is usually of small scales as annotating IOS meshes requires intensive human efforts. In this paper, we propose a novel self-supervised learning framework, named STSNet, to boost the performance of 3D tooth segmentation leveraging on large-scale unlabeled IOS data. The framework follows two-stage training, i.e., pre-training and fine-tuning. In pre-training, three hierarchical-level, i.e., point-level, region-level, cross-level, contrastive losses are proposed for unsupervised representation learning on a set of predefined matched points from different augmented views. The pretrained segmentation backbone is further fine-tuned in a supervised manner with a small number of labeled IOS meshes. With the same amount of annotated samples, our method can achieve an mIoU of 89.88%, significantly outperforming the supervised counterparts. The performance gain becomes more remarkable when only a small amount of labeled samples are available. Furthermore, STSNet can achieve better performance with only 40% of the annotated samples as compared to the fully supervised baselines. To the best of our knowledge, we present the first attempt of unsupervised pre-training for 3D tooth segmentation, demonstrating its strong potential in reducing human efforts for annotation and verification.

Machine Learning in Dentistry: A Scoping Review

Machine learning (ML) is being increasingly employed in dental research and application. We aimed to systematically compile studies using ML in dentistry and assess their methodological quality, including the risk of bias and reporting standards. We evaluated studies employing ML in dentistry published from 1 January 2015 to 31 May 2021 on MEDLINE, IEEE Xplore, and arXiv. We assessed publication trends and the distribution of ML tasks (classification, object detection, semantic segmentation, instance segmentation, and generation) in different clinical fields. We appraised the risk of bias and adherence to reporting standards, using the QUADAS-2 and TRIPOD checklists, respectively. Out of 183 identified studies, 168 were included, focusing on various ML tasks and employing a broad range of ML models, input data, data sources, strategies to generate reference tests, and performance metrics. Classification tasks were most common. Forty-two different metrics were used to evaluate model performances, with accuracy, sensitivity, precision, and intersection-over-union being the most common. We observed considerable risk of bias and moderate adherence to reporting standards which hampers replication of results. A minimum (core) set of outcome and outcome metrics is necessary to facilitate comparisons across studies.

An Explainable Deep Learning Model to Prediction Dental Caries Using Panoramic Radiograph Images

Dental caries is the most frequent dental health issue in the general population. Dental caries can result in extreme pain or infections, lowering people's quality of life. Applying machine learning models to automatically identify dental caries can lead to earlier treatment. However, physicians frequently find the model results unsatisfactory due to a lack of explainability. Our study attempts to address this issue with an explainable deep learning model for detecting dental caries. We tested three prominent pre-trained models, EfficientNet-B0, DenseNet-121, and ResNet-50, to determine which is best for the caries detection task. These models take panoramic images as the input, producing a caries-non-caries classification result and a heat map, which visualizes areas of interest on the tooth. The model performance was evaluated using whole panoramic images of 562 subjects. All three models produced remarkably similar results. However, the ResNet-50 model exhibited a slightly better performance when compared to EfficientNet-B0 and DenseNet-121. This model obtained an accuracy of 92.00%, a sensitivity of 87.33%, and an F1-score of 91.61%. Visual inspection showed us that the heat maps were also located in the areas with caries. The proposed explainable deep learning model diagnosed dental caries with high accuracy and reliability. The heat maps help to explain the classification results by indicating a region of suspected caries on the teeth. Dentists could use these heat maps to validate the classification results and reduce misclassification.

A weakly supervised learning-based segmentation network for dental diseases

With the development of deep learning, medical image segmentation has become a promising technique for computer-aided medical diagnosis. However, the supervised training of the algorithm relies on a large amount of labeled data, and the private dataset bias generally exists in previous research, which seriously affects the algorithm's performance. In order to alleviate this problem and improve the robustness and generalization of the model, this paper proposes an end-to-end weakly supervised semantic segmentation network to learn and infer mappings. Firstly, an attention compensation mechanism (ACM) aggregating the class activation map (CAM) is designed to learn complementarily. Then the conditional random field (CRF) is introduced to prune the foreground and background regions. Finally, the obtained high-confidence regions are used as pseudo labels for the segmentation branch to train and optimize using a joint loss function. Our model achieves a Mean Intersection over Union (MIoU) score of 62.84% in the segmentation task, which is an effective improvement of 11.18% compared to the previous network for segmenting dental diseases. Moreover, we further verify that our model has higher robustness to dataset bias by improved localization mechanism (CAM). The research shows that our proposed approach improves the accuracy and robustness of dental disease identification.

Artificial Intelligence in Diagnosis of Oral Diseases: A Systematic Review

AIM

To understand the role of Artificial intelligence (AI) in oral radiology and its applications.

BACKGROUND

Over the last two decades, the field of AI has undergone phenomenal progression and expansion. Artificial intelligence applications have taken up new roles in dentistry like digitized data acquisition and machine learning and diagnostic applications.

MATERIALS AND METHODS

All research papers outlining the population, intervention, control, and outcomes (PICO) questions were searched for in PubMed, ERIC, Embase, CINAHL, database from the last 10 years on first January 2023. Two authors independently reviewed the titles and abstracts of the selected studies, and any discrepancy between the two review authors was handled by a third reviewer. Two independent investigators evaluated all the included studies for the quality assessment using the modified tool for the quality assessment of diagnostic accuracy studies (QUADAS- 2).

REVIEW RESULTS

After the removal of duplicates and screening of titles and abstracts, 18 full texts were agreed upon for further evaluation, of which 14 that met the inclusion criteria were included in this review. The application of artificial intelligence models has primarily been reported on osteoporosis diagnosis, classification/segmentation of maxillofacial cysts and/or tumors, and alveolar bone resorption. Overall study quality was deemed to be high for two (14%) studies, moderate for six (43%) studies, and low for another six (43%) studies.

CONCLUSION

The use of AI for patient diagnosis and clinical decision-making can be accomplished with relative ease, and the technology should be regarded as a reliable modality for potential future applications in oral diagnosis.

Robust Mesh Segmentation Using Feature-Aware Region Fusion

This paper introduces a simple but powerful segmentation algorithm for 3D meshes. Our algorithm consists of two stages: over-segmentation and region fusion. In the first stage, adaptive space partition is applied to perform over-segmentation, which is very efficient. In the second stage, we define a new intra-region difference, inter-region difference, and fusion condition with the help of various shape features and propose an iterative region fusion method. As the region fusion process is feature aware, our algorithm can deal with complex 3D meshes robustly. Massive qualitative and quantitative experiments also validate the advantages of the proposed algorithm.

Improving performance of deep learning models using 3.5D U-Net via majority voting for tooth segmentation on cone beam computed tomography

Deep learning allows automatic segmentation of teeth on cone beam computed tomography (CBCT). However, the segmentation performance of deep learning varies among different training strategies. Our aim was to propose a 3.5D U-Net to improve the performance of the U-Net in segmenting teeth on CBCT. This study retrospectively enrolled 24 patients who received CBCT. Five U-Nets, including 2Da U-Net, 2Dc U-Net, 2Ds U-Net, 2.5Da U-Net, 3D U-Net, were trained to segment the teeth. Four additional U-Nets, including 2.5Dv U-Net, 3.5Dv5 U-Net, 3.5Dv4 U-Net, and 3.5Dv3 U-Net, were obtained using majority voting. Mathematical morphology operations including erosion and dilation (E&D) were applied to remove diminutive noise speckles. Segmentation performance was evaluated by fourfold cross validation using Dice similarity coefficient (DSC), accuracy, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV). Kruskal-Wallis test with post hoc analysis using Bonferroni correction was used for group comparison. P < 0.05 was considered statistically significant. Performance of U-Nets significantly varies among different training strategies for teeth segmentation on CBCT (P < 0.05). The 3.5Dv5 U-Net and 2.5Dv U-Net showed DSC and PPV significantly higher than any of five originally trained U-Nets (all P < 0.05). E&D significantly improved the DSC, accuracy, specificity, and PPV (all P < 0.005). The 3.5Dv5 U-Net achieved highest DSC and accuracy among all U-Nets. The segmentation performance of the U-Net can be improved by majority voting and E&D. Overall speaking, the 3.5Dv5 U-Net achieved the best segmentation performance among all U-Nets.

Advancements in Dentistry with Artificial Intelligence: Current Clinical Applications and Future Perspectives

Artificial intelligence has been widely used in the field of dentistry in recent years. The present study highlights current advances and limitations in integrating artificial intelligence, machine learning, and deep learning in subfields of dentistry including periodontology, endodontics, orthodontics, restorative dentistry, and oral pathology. This article aims to provide a systematic review of current clinical applications of artificial intelligence within different fields of dentistry. The preferred reporting items for systematic reviews (PRISMA) statement was used as a formal guideline for data collection. Data was obtained from research studies for 2009-2022. The analysis included a total of 55 papers from Google Scholar, IEEE, PubMed, and Scopus databases. Results show that artificial intelligence has the potential to improve dental care, disease diagnosis and prognosis, treatment planning, and risk assessment. Finally, this study highlights the limitations of the analyzed studies and provides future directions to improve dental care.

Application of deep learning artificial intelligence technique to the classification of clinical orthodontic photos

BACKGROUND

Taking facial and intraoral clinical photos is one of the essential parts of orthodontic diagnosis and treatment planning. Among the diagnostic procedures, classification of the shuffled clinical photos with their orientations will be the initial step while it was not easy for a machine to classify photos with a variety of facial and dental situations. This article presents a convolutional neural networks (CNNs) deep learning technique to classify orthodontic clinical photos according to their orientations.

METHODS

To build an automated classification system, CNNs models of facial and intraoral categories were constructed, and the clinical photos that are routinely taken for orthodontic diagnosis were used to train the models with data augmentation. Prediction procedures were evaluated with separate photos whose purpose was only for prediction.

RESULTS

Overall, a 98.0% valid prediction rate resulted for both facial and intraoral photo classification. The highest prediction rate was 100% for facial lateral profile, intraoral upper, and lower photos.

CONCLUSION

An artificial intelligence system that utilizes deep learning with proper training models can successfully classify orthodontic facial and intraoral photos automatically. This technique can be used for the first step of a fully automated orthodontic diagnostic system in the future.

A Tooth Segmentation Method Based on Multiple Geometric Feature Learning

Tooth segmentation is an important aspect of virtual orthodontic systems. In some existing studies using deep learning-based tooth segmentation methods, the feature learning of point coordinate information and normal vector information is not effectively distinguished. This will lead to the feature information of these two methods not producing complementary intermingling. To address this problem, a tooth segmentation method based on multiple geometric feature learning is proposed in this paper. First, the spatial transformation (T-Net) module is used to complete the alignment of dental model mesh features. Second, a multiple geometric feature learning module is designed to encode and enhance the centroid coordinates and normal vectors of each triangular mesh to highlight the differences between geometric features of different meshes. Finally, for local to global fusion features, feature downscaling and channel optimization are accomplished layer by layer using multilayer perceptron (MLP) and efficient channel attention (ECA). The experimental results show that our algorithm achieves better accuracy and efficiency of tooth segmentation and can assist dentists in their treatment work.

Transformer-Based Deep Learning Network for Tooth Segmentation on Panoramic Radiographs

Panoramic radiographs can assist dentist to quickly evaluate patients' overall oral health status. The accurate detection and localization of tooth tissue on panoramic radiographs is the first step to identify pathology, and also plays a key role in an automatic diagnosis system. However, the evaluation of panoramic radiographs depends on the clinical experience and knowledge of dentist, while the interpretation of panoramic radiographs might lead misdiagnosis. Therefore, it is of great significance to use artificial intelligence to segment teeth on panoramic radiographs. In this study, SWin-Unet, the transformer-based Ushaped encoder-decoder architecture with skip-connections, is introduced to perform panoramic radiograph segmentation. To well evaluate the tooth segmentation performance of SWin-Unet, the PLAGH-BH dataset is introduced for the research purpose. The performance is evaluated by F1 score, mean intersection and Union (IoU) and Acc, Compared with U-Net, Link-Net and FPN baselines, SWin-Unet performs much better in PLAGH-BH tooth segmentation dataset. These results indicate that SWin-Unet is more feasible on panoramic radiograph segmentation, and is valuable for the potential clinical application.

Clinical tooth segmentation based on local enhancement

The tooth arrangements of human beings are challenging to accurately observe when relying on dentists’ naked eyes, especially for dental caries in children, which is difficult to detect. Cone-beam computer tomography (CBCT) is used as an auxiliary method to measure patients’ teeth, including children. However, subjective and irreproducible manual measurements are required during this process, which wastes much time and energy for the dentists. Therefore, a fast and accurate tooth segmentation algorithm that can replace repeated calculations and annotations in manual segmentation has tremendous clinical significance. This study proposes a local contextual enhancement model for clinical dental CBCT images. The local enhancement model, which is more suitable for dental CBCT images, is proposed based on the analysis of the existing contextual models. Then, the local enhancement model is fused into an encoder–decoder framework for dental CBCT images. At last, extensive experiments are conducted to validate our method.

A Fully Automated Method for 3D Individual Tooth Identification and Segmentation in Dental CBCT

Accurate and automatic segmentation of three-dimensional (3D) individual teeth from cone-beam computerized tomography (CBCT) images is a challenging problem because of the difficulty in separating an individual tooth from adjacent teeth and its surrounding alveolar bone. Thus, this paper proposes a fully automated method of identifying and segmenting 3D individual teeth from dental CBCT images. The proposed method addresses the aforementioned difficulty by developing a deep learning-based hierarchical multi-step model. First, it automatically generates upper and lower jaws panoramic images to overcome the computational complexity caused by high-dimensional data and the curse of dimensionality associated with limited training dataset. The obtained 2D panoramic images are then used to identify 2D individual teeth and capture loose- and tight- regions of interest (ROIs) of 3D individual teeth. Finally, accurate 3D individual tooth segmentation is achieved using both loose and tight ROIs. Experimental results showed that the proposed method achieved an F1-score of 93.35 percent for tooth identification and a Dice similarity coefficient of 94.79 percent for individual 3D tooth segmentation. The results demonstrate that the proposed method provides an effective clinical and practical framework for digital dentistry.

Analysis of Deep Learning Techniques for Dental Informatics: A Systematic Literature Review

Within the ever-growing healthcare industry, dental informatics is a burgeoning field of study. One of the major obstacles to the health care system’s transformation is obtaining knowledge and insightful data from complex, high-dimensional, and diverse sources. Modern biomedical research, for instance, has seen an increase in the use of complex, heterogeneous, poorly documented, and generally unstructured electronic health records, imaging, sensor data, and text. There were still certain restrictions even after many current techniques were used to extract more robust and useful elements from the data for analysis. New effective paradigms for building end-to-end learning models from complex data are provided by the most recent deep learning technology breakthroughs. Therefore, the current study aims to examine the most recent research on the use of deep learning techniques for dental informatics problems and recommend creating comprehensive and meaningful interpretable structures that might benefit the healthcare industry. We also draw attention to some drawbacks and the need for better technique development and provide new perspectives about this exciting new development in the field.

Semantic Segmentation of Maxillary Teeth and Palatal Rugae in Two-Dimensional Images

The superimposition of sequential radiographs of the head is commonly used to determine the amount and direction of orthodontic tooth movement. A harmless method includes the timely unlimited superimposition on the relatively stable palatal rugae, but the method is performed manually and, if automated, relies on the best fit of surfaces, not only rugal structures. In the first step, motion estimation requires segmenting and detecting the location of teeth and rugae at any time during the orthodontic intervention. Aim: to develop a process of tooth segmentation that eliminates all manual steps to achieve an autonomous system of assessment of the dentition. Methods: A dataset of 797 occlusal views from photographs of teeth was created. The photographs were manually semantically segmented and labeled. Machine learning methods were applied to identify a robust deep network architecture able to semantically segment teeth in unseen photographs. Using well-defined metrics such as accuracy, precision, and the average mean intersection over union (mIoU), four network architectures were tested: MobileUnet, AdapNet, DenseNet, and SegNet. The robustness of the trained network was additionally tested on a set of 47 image pairs of patients before and after orthodontic treatment. Results: SegNet was the most accurate network, producing 95.19% accuracy and an average mIoU value of 86.66% for the main sample and 86.2% for pre- and post-treatment images. Conclusions: Four architectural tests were developed for automated individual teeth segmentation and detection in two-dimensional photos that required no post-processing. Accuracy and robustness were best achieved with SegNet. Further research should focus on clinical applications and 3D system development.

Accuracy and efficiency of automatic tooth segmentation in digital dental models using deep learning

This study evaluates the accuracy and efficiency of automatic tooth segmentation in digital dental models using deep learning. We developed a dynamic graph convolutional neural network (DGCNN)-based algorithm for automatic tooth segmentation and classification using 516 digital dental models. We segmented 30 digital dental models using three methods for comparison: (1) automatic tooth segmentation (AS) using the DGCNN-based algorithm from LaonSetup software, (2) landmark-based tooth segmentation (LS) using OrthoAnalyzer software, and (3) tooth designation and segmentation (DS) using Autolign software. We evaluated the segmentation success rate, mesiodistal (MD) width, clinical crown height (CCH), and segmentation time. For the AS, LS, and DS, the tooth segmentation success rates were 97.26%, 97.14%, and 87.86%, respectively (p < 0.001, post-hoc; AS, LS > DS), the means of MD widths were 8.51, 8.28, and 8.63 mm, respectively (p < 0.001, post hoc; DS > AS > LS), the means of CCHs were 7.58, 7.65, and 7.52 mm, respectively (p < 0.001, post-hoc; LS > DS, AS), and the means of segmentation times were 57.73, 424.17, and 150.73 s, respectively (p < 0.001, post-hoc; AS < DS < LS). Automatic tooth segmentation of a digital dental model using deep learning showed high segmentation success rate, accuracy, and efficiency; thus, it can be used for orthodontic diagnosis and appliance fabrication.

A Few-Shot Dental Object Detection Method Based on a Priori Knowledge Transfer

With the continuous improvement in oral health awareness, people’s demand for oral health diagnosis has also increased. Dental object detection is a key step in automated dental diagnosis; however, because of the particularity of medical data, researchers usually cannot obtain sufficient medical data. Therefore, this study proposes a dental object detection method for small-size datasets based on teeth semantics, structural information feature extraction, and an a priori knowledge migration, called a segmentation, points, segmentation, and classification network (SPSC-NET). In the region of interest area extraction method, the SPSC-NET method converts the teeth X-ray image into an a priori knowledge information image, composed of the edges of the teeth and the semantic segmentation image; the network structure used to extract the a priori knowledge information is a symmetric structure, which then generates the key points of the object instance. Next, it uses the key points of the object instance (i.e., the dental semantic segmentation image and the dental edge image) to obtain the object instance image (i.e., the positioning of the teeth). Using 10 training images, the test precision and recall rate of the tooth object center point of the SPSC-NET method were between 99–100%. In the classification method, the SPSC-NET identified the single instance segmentation image generated by migrating the dental object area, the edge image, and the semantic segmentation image as a priori knowledge. Under the premise of using the same deep neural network classification model, the model classification with a priori knowledge was 20% more accurate than the ordinary classification methods. For the overall object detection performance indicators, the SPSC-NET’s average precision (AP) value was more than 92%, which is better than that of the transfer-based faster region-based convolutional neural network (Faster-RCNN) object detection model; moreover, its AP and mean intersection-over-union (mIOU) were 14.72% and 19.68% better than the transfer-based Faster-CNN model, respectively.

Mandibular Premolar Identification System based on a Deep Learning Model

OBJECTIVES

For constructing an isolated tooth identification system using deep learning, Igarashi et al. (2021) began constructing a learning model as basic research to identify the left and right mandibular first and second premolars. These teeth were chosen for analysis because they are difficult to identify from one another. The learning method itself was proven appropriate but presented low accuracy. Therefore, further improvement in the learning data should increase the accuracy of the model. The study objectives were to modify the learning data and increase the learning model accuracy for enabling the identification of isolated lower premolars.

METHODS

Static images of the occlusal surface of the premolars made from the dental plaster casts of dental students were used as the training, validation, and test data. A convolutional neural network with 32 hidden layers, AlexNet, convolutional architecture for fast feature embedding, and stochastic gradient descent was used to construct four learning models.

RESULTS

The accuracy of the identification model increased using static images of the occlusal surface of the teeth with the adjacent teeth deleted as the training and validation data; however, a learning model that could perfectly identify the teeth could not be realized.

CONCLUSIONS

Static images of the occlusal surface of the teeth with the adjacent teeth deleted should be used as both training and validation data. The ratio of the numbers of training, validation, and test data should be optimized.

Two-Stream Graph Convolutional Network for Intra-Oral Scanner Image Segmentation

Precise segmentation of teeth from intra-oral scanner images is an essential task in computer-aided orthodontic surgical planning. The state-of-the-art deep learning-based methods often simply concatenate the raw geometric attributes (i.e., coordinates and normal vectors) of mesh cells to train a single-stream network for automatic intra-oral scanner image segmentation. However, since different raw attributes reveal completely different geometric information, the naive concatenation of different raw attributes at the (low-level) input stage may bring unnecessary confusion in describing and differentiating between mesh cells, thus hampering the learning of high-level geometric representations for the segmentation task. To address this issue, we design a two-stream graph convolutional network (i.e., TSGCN), which can effectively handle inter-view confusion between different raw attributes to more effectively fuse their complementary information and learn discriminative multi-view geometric representations. Specifically, our TSGCN adopts two input-specific graph-learning streams to extract complementary high-level geometric representations from coordinates and normal vectors, respectively. Then, these single-view representations are further fused by a self-attention module to adaptively balance the contributions of different views in learning more discriminative multi-view representations for accurate and fully automatic tooth segmentation. We have evaluated our TSGCN on a real-patient dataset of dental (mesh) models acquired by 3D intraoral scanners. Experimental results show that our TSGCN significantly outperforms state-of-the-art methods in 3D tooth (surface) segmentation.

Caries Segmentation on Tooth X-ray Images with a Deep Network

OBJECTIVES

Deep learning has been a promising technology in many biomedical applications. In this study, a deep network was proposed aiming for caries segmentation on the clinically collected tooth X-ray images.

METHODS

The proposed network inherited the skip connection characteristic from the widely used U-shaped network, and creatively adopted vision Transformer, dilated convolution, and feature pyramid fusion methods to enhance the multi-scale and global feature extraction capability. It was then trained on the clinically self-collected and augmented tooth X-ray image dataset, and the dice similarity and pixel classification precision were calculated for the network's performance evaluation.

RESULTS

Experimental results revealed an average dice similarity of 0.7487 and an average pixel classification precision of 0.7443 on the test dataset, which outperformed the compared networks such as UNet, Trans-UNet, and Swin-UNet, demonstrating the remarkable improvement of the proposed network.

CONCLUSIONS

This study contributed to the automatic caries segmentation by using a deep network, and highlighted the potential clinical utility value.

Applications of artificial intelligence in dentistry: A comprehensive review

OBJECTIVE

To perform a comprehensive review of the use of artificial intelligence (AI) and machine learning (ML) in dentistry, providing the community with a broad insight on the different advances that these technologies and tools have produced, paying special attention to the area of esthetic dentistry and color research.

MATERIALS AND METHODS

The comprehensive review was conducted in MEDLINE/PubMed, Web of Science, and Scopus databases, for papers published in English language in the last 20 years.

RESULTS

Out of 3871 eligible papers, 120 were included for final appraisal. Study methodologies included deep learning (DL; n = 76), fuzzy logic (FL; n = 12), and other ML techniques (n = 32), which were mainly applied to disease identification, image segmentation, image correction, and biomimetic color analysis and modeling.

CONCLUSIONS

The insight provided by the present work has reported outstanding results in the design of high-performance decision support systems for the aforementioned areas. The future of digital dentistry goes through the design of integrated approaches providing personalized treatments to patients. In addition, esthetic dentistry can benefit from those advances by developing models allowing a complete characterization of tooth color, enhancing the accuracy of dental restorations.

CLINICAL SIGNIFICANCE

The use of AI and ML has an increasing impact on the dental profession and is complementing the development of digital technologies and tools, with a wide application in treatment planning and esthetic dentistry procedures.

Toward Clinically Applicable 3-Dimensional Tooth Segmentation via Deep Learning

Digital dentistry plays a pivotal role in dental health care. A critical step in many digital dental systems is to accurately delineate individual teeth and the gingiva in the 3-dimension intraoral scanned mesh data. However, previous state-of-the-art methods are either time-consuming or error prone, hence hindering their clinical applicability. This article presents an accurate, efficient, and fully automated deep learning model trained on a data set of 4,000 intraoral scanned data annotated by experienced human experts. On a holdout data set of 200 scans, our model achieves a per-face accuracy, average-area accuracy, and area under the receiver operating characteristic curve of 96.94%, 98.26%, and 0.9991, respectively, significantly outperforming the state-of-the-art baselines. In addition, our model takes only about 24 s to generate segmentation outputs, as opposed to >5 min by the baseline and 15 min by human experts. A clinical performance test of 500 patients with malocclusion and/or abnormal teeth shows that 96.9% of the segmentations are satisfactory for clinical applications, 2.9% automatically trigger alarms for human improvement, and only 0.2% of them need rework. Our research demonstrates the potential for deep learning to improve the efficacy and efficiency of dental treatment and digital dentistry.

Automatic Segmentation of Dental Root Canal and Merging with Crown Shape

In this paper, machine learning approaches are proposed to support dental researchers and clinicians to study the shape and position of dental crowns and roots, by implementing a Patient Specific Classification and Prediction tool that includes RootCanalSeg and DentalModelSeg algorithms and then merges the output of these tools for intraoral scanning and volumetric dental imaging. RootCanalSeg combines image processing and machine learning approaches to automatically segment the root canals of the lower and upper jaws from large datasets, providing clinical information on tooth long axis for orthodontics, endodontics, prosthodontic and restorative dentistry procedures. DentalModelSeg includes segmenting the teeth from the crown shape to provide clinical information on each individual tooth. The merging algorithm then allows users to integrate dental models for quantitative assessments. Precision in dentistry has been mainly driven by dental crown surface characteristics, but information on tooth root morphology and position is important for successful root canal preparation, pulp regeneration, planning of orthodontic movement, restorative and implant dentistry. In this paper we propose a patient specific classification and prediction of dental root canal and crown shape analysis workflow that employs image processing and machine learning methods to analyze crown surfaces, obtained by intraoral scanners, and three-dimensional volumetric images of the jaws and teeth root canals, obtained by cone beam computed tomography (CBCT).

Efficient Computer-Aided Design of Dental Inlay Restoration: A Deep Adversarial Framework

Restoring the normal masticatory function of broken teeth is a challenging task primarily due to the defect location and size of a patient's teeth. In recent years, although some representative image-to-image transformation methods (e.g. Pix2Pix) can be potentially applicable to restore the missing crown surface, most of them fail to generate dental inlay surface with realistic crown details (e.g. occlusal groove) that are critical to the restoration of defective teeth with varying shapes. In this article, we design a computer-aided Deep Adversarial-driven dental Inlay reStoration (DAIS) framework to automatically reconstruct a realistic surface for a defective tooth. Specifically, DAIS consists of a Wasserstein generative adversarial network (WGAN) with a specially designed loss measurement, and a new local-global discriminator mechanism. The local discriminator focuses on missing regions to ensure the local consistency of a generated occlusal surface, while the global discriminator aims at defective teeth and adjacent teeth to assess if it is coherent as a whole. Experimental results demonstrate that DAIS is highly efficient to deal with a large area of missing teeth in arbitrary shapes and generate realistic occlusal surface completion. Moreover, the designed watertight inlay prostheses have enough anatomical morphology, thus providing higher clinical applicability compared with more state-of-the-art methods.

Hierarchical CNN-based occlusal surface morphology analysis for classifying posterior tooth type using augmented images from 3D dental surface models

OBJECTIVE

3D Digitization of dental model is growing in popularity for dental application. Classification of tooth type from single 3D point cloud model without assist of relative position among teeth is still a challenging task.

METHODS

In this paper, 8-class posterior tooth type classification (first premolar, second premolar, first molar, second molar in maxilla and mandible respectively) was investigated by convolutional neural network (CNN)-based occlusal surface morphology analysis. 3D occlusal surface was transformed to depth image for basic CNN-based classification. Considering the logical hierarchy of tooth categories, a hierarchical classification structure was proposed to decompose 8-class classification task into two-stage cascaded classification subtasks. Image augmentations including traditional geometrical transformation and deep convolutional generative adversarial networks (DCGANs) were applied for each subnetworks and cascaded network.

RESULTS

Results indicate that combing traditional and DCGAN-based augmented images to train CNN models can improve classification performance. In the paper, we achieve overall accuracy 91.35%, macro precision 91.49%, macro-recall 91.29%, and macro-F1 0.9139 for the 8-class posterior tooth type classification, which outperform other deep learning models. Meanwhile, Grad-cam results demonstrate that CNN model trained by our augmented images will focus on smaller important region for better generality. And anatomic landmarks of cusp, fossa, and groove work as important regions for cascaded classification model.

CONCLUSION

The reported work has proved that using basic CNN to construct two-stage hierarchical structure can achieve the best classification performance of posterior tooth type in 3D model without assistance of relative position information. The proposed method has advantages of easy training, great ability to learn discriminative features from small image region.

Localization of cementoenamel junction in intraoral ultrasonographs with machine learning

OBJECTIVE

Our goal was to automatically identify the cementoenamel junction (CEJ) location in ultrasound images using deep convolution neural networks (CNNs).

METHODS

Three CNNs were evaluated using 1400 images and data augmentation. The training and validation were performed by an experienced nonclinical rater with 1000 and 200 images, respectively. Four clinical raters with different levels of experience with ultrasound tested the networks using the other 200 images. In addition to the comparison of the best approach with each rater, we also employed the simultaneous truth and performance level estimation (STAPLE) algorithm to estimate a ground truth based on all labelings by four clinical raters. The final CEJ location estimate was obtained by taking the first moment of the posterior probability computed using the STAPLE algorithm. The study also computed the machine learning-measured CEJ-alveolar bone crest distance.

RESULTS

Quantitative evaluations of the 200 images showed that the comparison of the best approach with the STAPLE-estimate yielded a mean difference (MD) of 0.26 mm, which is close to the comparison with the most experienced nonclinical rater (MD=0.25 mm) but far better than the comparison with clinical raters (MD=0.27-0.33 mm). The machine learning-measured CEJ-alveolar bone crest distances correlated strongly (R = 0.933, p < 0.001) with the manual clinical labeling and the measurements were in good agreement with the 95% Bland-Altman's lines of agreement between -0.68 and 0.57 mm.

CONCLUSIONS

The study demonstrated the feasible use of machine learning methodology to localize CEJ in ultrasound images with clinically acceptable accuracy and reliability. Likelihood-weighted ground truth by combining multiple labels by the clinical experts compared favorably with the predictions by the best deep CNN approach.

CLINICAL SIGNIFICANCE

Identification of CEJ and its distance from the alveolar bone crest play an important role in the evaluation of periodontal status. Machine learning algorithms can learn from complex features in ultrasound images and have potential to provide a reliable and accurate identification in subsecond. This will greatly assist dental practitioners to provide better point-of-care to patients and enhance the throughput of dental care.

Dental enumeration and multiple treatment detection on panoramic X-rays using deep learning

In this paper, a new powerful deep learning framework, named as DENTECT, is developed in order to instantly detect five different dental treatment approaches and simultaneously number the dentition based on the FDI notation on panoramic X-ray images. This makes DENTECT the first system that focuses on identification of multiple dental treatments; namely periapical lesion therapy, fillings, root canal treatment (RCT), surgical extraction, and conventional extraction all of which are accurately located within their corresponding borders and tooth numbers. Although DENTECT is trained on only 1005 images, the annotations supplied by experts provide satisfactory results for both treatment and enumeration detection. This framework carries out enumeration with an average precision (AP) score of 89.4% and performs treatment identification with a 59.0% AP score. Clinically, DENTECT is a practical and adoptable tool that accelerates the process of treatment planning with a level of accuracy which could compete with that of dental clinicians.

A deep-learning-based workflow to assess taxonomic affinity of hominid teeth with a test on discriminating Pongo and Homo upper molars

OBJECTIVES

Convolutional neural network (CNN) is a state-of-art deep learning (DL) method with superior performance in image classification. Here, a CNN-based workflow is proposed to discriminate hominid teeth. Our hope is that this method could help confirm otherwise questionable records of Homo from Pleistocene deposits where there is a standing risk of mis-attributing molars of Pongo to Homo.

METHODS AND MATERIALS

A two-step workflow was designed. The first step is converting the enamel-dentine junction (EDJ) into EDJ card, that is, a two-dimensional image conversion of the three-dimensional EDJ surface. In this step, researchers must carefully orient the teeth according to the cervical plane. The second step is training the CNN learner with labeled EDJ cards. A sample consisting of 53 fossil Pongo and 53 Homo (modern human and Neanderthal) was adopted to generate EDJ cards, which were then separated into training set (n = 84) and validation set (n = 22). To assess the feasibility of this workflow, a Pongo-Homo classifier was trained from the aforementioned EDJ card set, and then the classifier was used to predict the taxonomic affinities of six samples (test set) from von Koenigswald's Chinese Apothecary collection.

RESULTS

Results show that EDJ cards in validation set are classified accurately by the CNN learner. More importantly, taxonomic predictions for six specimens in test set match well with the diagnosis results deduced from multiple lines of evidence, implying the great potential of CNN method.

DISCUSSION

This workflow paves a way for future studies using CNN to address taxonomic complexity (e.g., distinguishing Pongo and Homo teeth from the Pleistocene of Asia). Further improvements include visual interpretation and extending the applicability to moderately worn teeth.

TSegNet: An efficient and accurate tooth segmentation network on 3D dental model

Automatic and accurate segmentation of dental models is a fundamental task in computer-aided dentistry. Previous methods can achieve satisfactory segmentation results on normal dental models; however, they fail to robustly handle challenging clinical cases such as dental models with missing, crowding, or misaligned teeth before orthodontic treatments. In this paper, we propose a novel end-to-end learning-based method, called TSegNet, for robust and efficient tooth segmentation on 3D scanned point cloud data of dental models. Our algorithm detects all the teeth using a distance-aware tooth centroid voting scheme in the first stage, which ensures the accurate localization of tooth objects even with irregular positions on abnormal dental models. Then, a confidence-aware cascade segmentation module in the second stage is designed to segment each individual tooth and resolve ambiguities caused by aforementioned challenging cases. We evaluated our method on a large-scale real-world dataset consisting of dental models scanned before or after orthodontic treatments. Extensive evaluations, ablation studies and comparisons demonstrate that our method can generate accurate tooth labels robustly in various challenging cases and significantly outperforms state-of-the-art approaches by 6.5% of Dice Coefficient, 3.0% of F1 score in term of accuracy, while achieving 20 times speedup of computational time.

Cardiac magnetic resonance image segmentation based on convolutional neural network

OBJECTIVE

In cardiac medical imaging, the extraction and segmentation of the part of interest is the key to the diagnosis of heart disease. Due to irregular diastole and contraction, magnetic resonance imaging (MRI) images have poorly defined boundaries, and traditional segmentation algorithms have poor performance. In this paper, a cardiac MRI segmentation technique using convolutional neural network and image saliency is suggested.

METHODS

The convolutional neural network is used for detecting target area, filter out the ribs, muscles and the other parts of the anatomy where the contrast is not clearly defined. It can also be used to extract the region of interest (ROI), and compute the contrast of the ROI in order to improve clarity of the heart tissue within the ROI. The cardiac image diagnosis is performed using the obtained saliency image and compared with the segmentation result of the region growth algorithm. Finally, the images of 85 patients were used to train and test the algorithm model. Here, 46 patients were randomly selected for training, and the remaining 39 were harnessed for further tests.

RESULTS

Segmentation accuracy of our algorithm model in ventricles, septum and the apex of the heart segment reaches 93.14%, 92.58% and 96.21% respectively, which are better than the segmentation method based on the regional growth technique.

CONCLUSIONS

The segmentation method using convolutional neural network and image saliency can meet the needs of automatic heart segmentation tasks based on cardiac MRI image sequences. The segmented image is able to assist the doctor to observe the patient's heart health more effectively. As such, our proposed technique has strong potential in clinical applications.

Applications of deep learning in dentistry

Over the last few years, translational applications of so-called artificial intelligence in the field of medicine have garnered a significant amount of interest. The present article aims to review existing dental literature that has examined deep learning, a subset of machine learning that has demonstrated the highest performance when applied to image processing and that has been tested as a formidable diagnostic support tool through its automated analysis of radiographic/photographic images. Furthermore, the article will critically evaluate the literature to describe potential methodological weaknesses of the studies and the need for further development. This review includes 28 studies that have described the applications of deep learning in various fields of dentistry. Research into the applications of deep learning in dentistry contains claims of its high accuracy. Nonetheless, many of these studies have substantial limitations and methodological issues (e.g., examiner reliability, the number of images used for training/testing, the methods used for validation) that have significantly limited the external validity of their results. Therefore, future studies that acknowledge the methodological limitations of existing literature will help to establish a better understanding of the usefulness of applying deep learning in dentistry.

Patient Specific Classification of Dental Root Canal and Crown Shape

This paper proposes machine learning approaches to support dentistry researchers in the context of integrating imaging modalities to analyze the morphology of tooth crowns and roots. One of the challenges to jointly analyze crowns and roots with precision is that two different image modalities are needed. Precision in dentistry is mainly driven by dental crown surfaces characteristics, but information on tooth root shape and position is of great value for successful root canal preparation, pulp regeneration, planning of orthodontic movement, restorative and implant dentistry. An innovative approach is to use image processing and machine learning to combine crown surfaces, obtained by intraoral scanners, with three dimensional volumetric images of the jaws and teeth root canals, obtained by cone beam computed tomography. In this paper, we propose a patient specific classification of dental root canal and crown shape analysis workflow that is widely applicable.