An evaluation of active shape models for the automatic identification of cephalometric landmarks

Cephalometric landmark annotation using transfer learning: Detectron2 and YOLOv8 baselines on a diverse cephalometric image dataset

BACKGROUND

Cephalometric landmark annotation is a key challenge in radiographic analysis, requiring automation due to its time-consuming process and inherent subjectivity. This study investigates the application of advanced transfer learning techniques to enhance the accuracy of anatomical landmarks in cephalometric images, which is a vital aspect of orthodontic diagnosis and treatment planning.

METHODS

We assess the suitability of transfer learning methods by employing state-of-the-art pose estimation models. The first framework is Detectron2, with two baselines featuring different ResNet backbone architectures: rcnn_R_50_FPN_3x and rcnn_R_101_FPN_3x. The second framework is YOLOv8, with three variants reflecting different network sizes: YOLOv8s-pose, YOLOv8m-pose, and YOLOv8l-pose. These pose estimation models are adopted for the landmark annotation task. The models are trained and evaluated on the DiverseCEPH19 dataset, comprising 1692 radiographic images with 19 landmarks, and their performance is analyzed across various images categories within the dataset. Additionally, the study is extended to a benchmark dataset of 400 images to investigate how dataset size impacts the performance of these frameworks.

RESULTS

Despite variations in objectives and evaluation metrics between pose estimation and landmark localization tasks, the results are promising. Detectron2's variant outperforms others with an accuracy of 85.89%, compared to 72.92% achieved by YOLOv8's variant on the DiverseCEPH19 dataset. This superior performance is also observed in the smaller benchmark dataset, where Detectron2 consistently maintains higher accuracy than YOLOv8.

CONCLUSION

The noted enhancements in annotation precision suggest the suitability of Detectron2 for deployment in applications that require high precision while taking into account factors such as model size, inference time, and resource utilization, the evidence favors YOLOv8 baselines.

Applications of artificial intelligence in dentomaxillofacial imaging: a systematic review

BACKGROUND

Artificial intelligence (AI) technology has been increasingly developed in oral and maxillofacial imaging. The aim of this systematic review was to assess the applications and performance of the developed algorithms in different dentomaxillofacial imaging modalities.

STUDY DESIGN

A systematic search of PubMed and Scopus databases was performed. The search strategy was set as a combination of the following keywords: "Artificial Intelligence," "Machine Learning," "Deep Learning," "Neural Networks," "Head and Neck Imaging," and "Maxillofacial Imaging." Full-text screening and data extraction were independently conducted by two independent reviewers; any mismatch was resolved by discussion. The risk of bias was assessed by one reviewer and validated by another.

RESULTS

The search returned a total of 3,392 articles. After careful evaluation of the titles, abstracts, and full texts, a total number of 194 articles were included. Most studies focused on AI applications for tooth and implant classification and identification, 3-dimensional cephalometric landmark detection, lesion detection (periapical, jaws, and bone), and osteoporosis detection.

CONCLUSION

Despite the AI models' limitations, they showed promising results. Further studies are needed to explore specific applications and real-world scenarios before confidently integrating these models into dental practice.

A Comparative Study of Traditional and Computer-aided Surgical Simulation Guides in Orthognathic Correction of Bimaxillary Protrusion

OBJECTIVE

To compare the efficacy of traditional occlusal guides with computer-aided surgical simulation (CASS) guides in enhancing postoperative outcomes for patients with bimaxillary protrusion.

METHODS

This retrospective study evaluated 34 patients undergoing anterior maxillary and mandibular subapical osteotomy at the Plastic Surgery Hospital of the Chinese Academy of Medical Sciences. Fourteen patients were treated using traditional occlusal guides, whereas 20 patients were treated with CASS guides (median age 28.6 years, median follow-up 259 days). Pre and postoperative cephalometric indicators were measured using cephalometric software. Data analysis was conducted using SPSS 14.0, with significant differences determined at P < 0.05.

RESULTS

All 34 patients experienced primary healing without complications. Follow-up indicated significant improvements in key cephalometric measurements in the CASS group compared with the traditional group, including mandibular position (SNB angle, P < 0.001), jaw relationship (ANB angle, P < 0.001), facial angle (FH-NPo, P = 0.002), and condyle-to-sella distance (Co-S, P = 0.024). The CASS group also showed better aesthetic outcomes, with significant reductions in overjet (P = 0.012), overbite (P = 0.001), and improved alignment of upper and lower incisors (U1-L1 angle, P = 0.031).

CONCLUSION

CASS-guided surgery offers a superior alternative to traditional methods for treating bimaxillary protrusion, providing more precise and aesthetically pleasing results. This study highlights the significant advantages of using advanced digital simulation and 3-dimensional printing technologies in orthognathic surgery.

The Implications of Artificial Intelligence in Pedodontics: A Scoping Review of Evidence-Based Literature

BACKGROUND

Artificial intelligence (AI) has emerged as a revolutionary technology with several applications across different dental fields, including pedodontics. This systematic review has the objective to catalog and explore the various uses of artificial intelligence in pediatric dentistry.

METHODS

A thorough exploration of scientific databases was carried out to identify studies addressing the usage of AI in pediatric dentistry until December 2023 in the Embase, Scopus, PubMed, and Web of Science databases by two researchers, S.L.R. and A.L.G.

RESULTS

From a pool of 1301 articles, only 64 met the predefined criteria and were considered for inclusion in this review. From the data retrieved, it was possible to provide a narrative discussion of the potential implications of AI in the specialized area of pediatric dentistry. The use of AI algorithms and machine learning techniques has shown promising results in several applications of daily dental pediatric practice, including the following: (1) assisting the diagnostic and recognizing processes of early signs of dental pathologies, (2) enhancing orthodontic diagnosis by automating cephalometric tracing and estimating growth and development, (3) assisting and educating children to develop appropriate behavior for dental hygiene.

CONCLUSION

AI holds significant potential in transforming clinical practice, improving patient outcomes, and elevating the standards of care in pediatric patients. Future directions may involve developing cloud-based platforms for data integration and sharing, leveraging large datasets for improved predictive results, and expanding AI applications for the pediatric population.

Machine Learning Algorithms for the Diagnosis of Class III Malocclusions in Children

Artificial intelligence has been applied to medical diagnosis and decision-making but it has not been used for classification of Class III malocclusions in children.

OBJECTIVE

This study aims to propose an innovative machine learning (ML)-based diagnostic model for automatically classifies dental, skeletal and functional Class III malocclusions.

METHODS

The collected data related to 46 cephalometric feature measurements from 4-14-year-old children (n = 666). The data set was divided into a training set and a test set in a 7:3 ratio. Initially, we employed the Recursive Feature Elimination (RFE) algorithm to filter the 46 input parameters, selecting 14 significant features. Subsequently, we constructed 10 ML models and trained these models using the 14 significant features from the training set through ten-fold cross-validation, and evaluated the models' average accuracy in test set. Finally, we conducted an interpretability analysis of the optimal model using the ML model interpretability tool SHapley Additive exPlanations (SHAP).

RESULTS

The top five models ranked by their area under the curve (AUC) values were: GPR (0.879), RBF SVM (0.876), QDA (0.876), Linear SVM (0.875) and L2 logistic (0.869). The DeLong test showed no statistical difference between GPR and the other models (p > 0.05). Therefore GPR was selected as the optimal model. The SHAP feature importance plot revealed that he top five features were SN-GoMe (the ratio of the length of the anterior skull base SN to that of the mandibular base GoMe), U1-NA (maxillary incisor angulation to NA plane), Overjet (the distance between two lines perpendicular to the functional occlusal plane from U1 and L), ANB (the difference between angles SNA and SNB), and AB-NPo (the angle between the AB and N-Pog line).

CONCLUSIONS

Our findings suggest that ML models based on cephalometric data could effectively assist dentists to classify dental, functional and skeletal Class III malocclusions in children. In addition, features such as SN_GoMe, U1_NA and Overjet can as important indicators for predicting the severity of Class III malocclusions.

Machine Learning Predictive Model as Clinical Decision Support System in Orthodontic Treatment Planning

Diagnosis and treatment planning forms the crux of orthodontics, which orthodontists gain with years of expertise. Machine Learning (ML), having the ability to learn by pattern recognition, can gain this expertise in a very short duration, ensuring reduced error, inter-intra clinician variability and good accuracy. Thus, the aim of this study was to construct an ML predictive model to predict a broader outline of the orthodontic diagnosis and treatment plan. The sample consisted of 700 case records of orthodontically treated patients in the past ten years. The data were split into a training and a test set. There were 33 input variables and 11 output variables. Four ML predictive model layers with seven algorithms were created. The test set was used to check the efficacy of the ML-predicted treatment plan and compared with that of the decision made by the expert orthodontists. The model showed an overall average accuracy of 84%, with the Decision Tree, Random Forest and XGB classifier algorithms showing the highest accuracy ranging from 87-93%. Yet in their infancy stages, Machine Learning models could become a valuable Clinical Decision Support System in orthodontic diagnosis and treatment planning in the future.

Automatic analysis of lateral cephalograms based on high-resolution net

INTRODUCTION

Cephalometric analysis is essential in orthodontic treatment, and it is progressing toward automatic cephalometric analysis. This study aimed to establish a cephalometric landmark detection model on the basis of a high-resolution net and improve the accuracy with high resolution.

METHODS

A total of 2000 lateral cephalograms were collected to construct a dataset, and the number of target landmarks was 51. A high-resolution network model was applied to the landmark detection task. Four models were trained by adjusting different input resolutions to choose the most suitable resolution. A test set consisting of 300 lateral cephalograms was used for evaluation. The model was evaluated from the error size and distribution of each landmark.

RESULTS

After 200 epochs of training, a landmark detection model was established. Under different resolutions of the input image, the mean model radial error decreased initially and then increased. At 680 × 920 pixels resolution, the minimum error and the highest detection success rate were obtained. The mean radial error was 1.08 ± 0.87 mm. The detection success rates of 2.0 mm, 2.5 mm, 3.0 mm, and 4.0 mm were 89.00%, 94.00%, 96.33%, and 98.67%, respectively. The mean radial errors of 22 landmarks were <1 mm, and the errors of other landmarks were <2 mm except for the pterion. The error distribution of landmarks followed a certain pattern.

CONCLUSIONS

An automatic landmark detection model based on a high-resolution net was established to recognize 51 landmarks. The model showed high detection accuracy, which provides a basis for further measurement application.

Cephalometric Analysis in Orthodontics Using Artificial Intelligence-A Comprehensive Review

Artificial intelligence (AI) is a branch of science concerned with developing programs and computers that can gather data, reason about it, and then translate it into intelligent actions. AI is a broad area that includes reasoning, typical linguistic dispensation, machine learning, and planning. In the area of medicine and dentistry, machine learning is currently the most widely used AI application. This narrative review is aimed at giving an outline of cephalometric analysis in orthodontics using AI. Latest algorithms are developing rapidly, and computational resources are increasing, resulting in increased efficiency, accuracy, and reliability. Current techniques for completely automatic identification of cephalometric landmarks have considerably improved efficiency and growth prospects for their regular use. The primary considerations for effective orthodontic treatment are an accurate diagnosis, exceptional treatment planning, and good prognosis estimation. The main objective of the AI technique is to make dentists' work more precise and accurate. AI is increasingly being used in the area of orthodontic treatment. It has been evidenced to be a time-saving and reliable tool in many ways. AI is a promising tool for facilitating cephalometric tracing in routine clinical practice and analyzing large databases for research purposes.

Accuracy of artificial intelligence-assisted landmark identification in serial lateral cephalograms of Class III patients who underwent orthodontic treatment and two-jaw orthognathic surgery

Objective

To investigate the pattern of accuracy change in artificial intelligence-assisted landmark identification (LI) using a convolutional neural network (CNN) algorithm in serial lateral cephalograms (Lat-cephs) of Class III (C-III) patients who underwent two-jaw orthognathic surgery.

Methods

A total of 3,188 Lat-cephs of C-III patients were allocated into the training and validation sets (3,004 Lat-cephs of 751 patients) and test set (184 Lat-cephs of 46 patients; subdivided into the genioplasty and non-genioplasty groups, n = 23 per group) for LI. Each C-III patient in the test set had four Lat-cephs initial (T0), pre-surgery (T1, presence of orthodontic brackets [OBs]), post-surgery (T2, presence of OBs and surgical plates and screws [S-PS]), and debonding (T3, presence of S-PS and fixed retainers [FR]). After mean errors of 20 landmarks between human gold standard and the CNN model were calculated, statistical analysis was performed.

Results

The total mean error was 1.17 mm without significant difference among the four time-points (T0, 1.20 mm; T1, 1.14 mm; T2, 1.18 mm; T3, 1.15 mm). In comparison of two time-points ([T0, T1] vs. [T2, T3]), ANS, A point, and B point showed an increase in error (p < 0.01, 0.05, 0.01, respectively), while Mx6D and Md6D showeda decrease in error (all p < 0.01). No difference in errors existed at B point, Pogonion, Menton, Md1C, and Md1R between the genioplasty and non-genioplasty groups.

Conclusions

The CNN model can be used for LI in serial Lat-cephs despite the presence of OB, S-PS, FR, genioplasty, and bone remodeling.

Evaluation of fully automated cephalometric measurements obtained from web-based artificial intelligence driven platform

Background

Artificial Intelligence has created a huge impact in different areas of dentistry. Automated cephalometric analysis is one of the major applications of artificial intelligence in the field of orthodontics. Various automated cephalometric software have been developed which utilizes artificial intelligence and claim to be reliable. The purpose of this study was to compare the linear and angular cephalometric measurements obtained from web-based fully automated Artificial Intelligence (AI) driven platform “WebCeph”™ with that from manual tracing and evaluate the validity and reliability of automated cephalometric measurements obtained from “WebCeph”™.

Methods

Thirty pre-treatment lateral cephalograms of patients were randomly selected. For manual tracing, digital images of same cephalograms were printed using compatible X-ray printer. After calibration, a total of 18 landmarks was plotted and 12 measurements (8 angular and 4 linear) were obtained using standard protocols. The digital images of each cephalogram were uploaded to “WebCeph”™ server. After image calibration, the automated cephalometric measurements obtained through AI digitization were downloaded for each image. Intraclass correlation coefficient (ICC) was used to determine agreement between the measurements obtained from two methods. ICC value < 0.75 was considered as poor to moderate agreement while an ICC value between 0.75 and 0.90 was considered as good agreement. Agreement was rated as excellent when ICC value > 0.90 was obtained.

Results

All the measurements had ICC value above 0.75. A higher ICC value > 0.9 was obtained for seven parameters i.e. ANB, FMA, IMPA/L1 to MP (°), LL to E-line, L1 to NB (mm), L1 to NB (°), S-N to Go-Gn whereas five parameters i.e. UL to E-line, U1 to NA (mm), SNA, SNB, U1 to NA (°) showed ICC value between 0.75 and 0.90.

Conclusion

A good agreement was found between the cephalometric measurements obtained from “WebCeph”™ and manual tracing.

Landmark Localization for Cephalometric Analysis using Multiscale Image Patch-based Graph Convolutional Networks

Accurate and robust cephalometric image analysis plays an essential role in orthodontic diagnosis, treatment assessment and surgical planning. This paper proposes a novel landmark localization method for cephalometric analysis using multiscale image patch-based graph convolutional networks. In detail, image patches with the same size are hierarchically sampled from the Gaussian pyramid to well preserve multiscale context information. We combine local appearance and shape information into spatialized features with an attention module to enrich node representations in graph. The spatial relationships of landmarks are built with the incorporation of three-layer graph convolutional networks, and multiple landmarks are simultaneously updated and moved toward the targets in a cascaded coarse-to-fine process. Quantitative results obtained on publicly available cephalometric X-ray images have exhibited superior performance compared with other state-of-the-art methods in terms of mean radial error and successful detection rate within various precision ranges. Our approach performs significantly better especially in the clinically accepted range of 2 mm and this makes it suitable in cephalometric analysis and orthognathic surgery.

Current applications and development of artificial intelligence for digital dental radiography

In the last few years, artificial intelligence (AI) research has been rapidly developing and emerging in the field of dental and maxillofacial radiology. Dental radiography, which is commonly used in daily practices, provides an incredibly rich resource for AI development and attracted many researchers to develop its application for various purposes. This study reviewed the applicability of AI for dental radiography from the current studies. Online searches on PubMed and IEEE Xplore databases, up to December 2020, and subsequent manual searches were performed. Then, we categorized the application of AI according to similarity of the following purposes: diagnosis of dental caries, periapical pathologies, and periodontal bone loss; cyst and tumor classification; cephalometric analysis; screening of osteoporosis; tooth recognition and forensic odontology; dental implant system recognition; and image quality enhancement. Current development of AI methodology in each aforementioned application were subsequently discussed. Although most of the reviewed studies demonstrated a great potential of AI application for dental radiography, further development is still needed before implementation in clinical routine due to several challenges and limitations, such as lack of datasets size justification and unstandardized reporting format. Considering the current limitations and challenges, future AI research in dental radiography should follow standardized reporting formats in order to align the research designs and enhance the impact of AI development globally.

Automatic localization of cephalometric landmarks based on convolutional neural network

INTRODUCTION

Cephalometry plays an important role in the diagnosis and treatment of orthodontics and orthognathic surgery. This study intends to develop an automatic landmark location system to make cephalometry more convenient.

METHODS

In this study, 512 lateral cephalograms were collected, and 37 landmarks were included. The coordinates of all landmarks in the 512 films were obtained to establish a labeled dataset: 312 were used as a training set, 100 as a validation set, and 100 as a testing set. An automatic landmark location system based on the convolutional neural network was developed. This system consisted of a global detection module and a locally modified module. The lateral cephalogram was first fed into the global module to obtain an initial estimate of the landmark's position, which was then adjusted with the locally modified module to improve accuracy. Mean radial error (MRE) and success detection rate (SDR) within the range of 1-4 mm were used to evaluate the method.

RESULTS

The MRE of our validation set was 1.127 ± 1.028 mm, and SDR of 1.0, 1.5, 2.0, 2.5, 3.0, and 4.0 mm were respectively 45.95%, 89.19%, 97.30%, 97.30%, and 97.30%. The MRE of our testing set was 1.038 ± 0.893 mm, and SDR of 1.0, 1.5, 2.0, 2.5, 3.0, and 4.0 mm were respectively 54.05%, 91.89%, 97.30%, 100%, 100%, and 100%.

CONCLUSIONS

In this study, we proposed a new automatic landmark location system on the basis of the convolutional neural network. The system could detect 37 landmarks with high accuracy. All landmarks are commonly used in clinical practice and could meet the requirements of different cephalometric analysis methods.

A fully deep learning model for the automatic identification of cephalometric landmarks

Purpose

This study aimed to propose a fully automatic landmark identification model based on a deep learning algorithm using real clinical data and to verify its accuracy considering inter-examiner variability.

Materials and Methods

In total, 950 lateral cephalometric images from Yonsei Dental Hospital were used. Two calibrated examiners manually identified the 13 most important landmarks to set as references. The proposed deep learning model has a 2-step structure—a region of interest machine and a detection machine—each consisting of 8 convolution layers, 5 pooling layers, and 2 fully connected layers. The distance errors of detection between 2 examiners were used as a clinically acceptable range for performance evaluation.

Results

The 13 landmarks were automatically detected using the proposed model. Inter-examiner agreement for all landmarks indicated excellent reliability based on the 95% confidence interval. The average clinically acceptable range for all 13 landmarks was 1.24 mm. The mean radial error between the reference values assigned by 1 expert and the proposed model was 1.84 mm, exhibiting a successful detection rate of 36.1%. The A-point, the incisal tip of the maxillary and mandibular incisors, and ANS showed lower mean radial error than the calibrated expert variability.

Conclusion

This experiment demonstrated that the proposed deep learning model can perform fully automatic identification of cephalometric landmarks and achieve better results than examiners for some landmarks. It is meaningful to consider between-examiner variability for clinical applicability when evaluating the performance of deep learning methods in cephalometric landmark identification.

[Cephalometric analysis of lateral skull X-ray images using soft computing components in the search for key points]

THE AIM OF THE STUDY

Was to investigate the efficiency of decoding teleradiological studies using an algorithm based on the use of convolutional neural networks - a simple convolutional architecture, as well as an extended U-Net architecture.

MATERIALS AND METHODS

For the experiment, a dataset was prepared by three orthodontists with over 10 years of clinical experience. Each of the orthodontists processed 100 X-ray images of the lateral projection of the head according to 27 parameters, 2700 measurements were made. The coordinates of the control points found by orthodontists in the images were compared with each other and a conclusion was made about the consistency of experts in the data obtained.

RESULTS

The results of convolutional neural network CNN were not satisfactory in 17 (62.96%) features, satisfactory in 10 (37.04%). The assessment of orthodontists resulted in non-satisfactory evaluation in 6 (22.22%), satisfactory in 8 (29.63%), good in 8 (29.63%), and excellent in 5 (18.52%) coordinates. Neural networks with U-Net architecture showed satisfactory results in 9 (33.3%) cases, good in 16 (59.3%) and excellent in 2 (7.4%) cases, with no non-satisfactory results.

CONCLUSION

The neural network of the U-Net architecture is more effective than a simple fully convolutional neural network and its results of determining anatomical reference points on two-dimensional images of the head are relatively comparable with the data obtained by medical specialists.

Accuracy of automated identification of lateral cephalometric landmarks using cascade convolutional neural networks on lateral cephalograms from nationwide multi-centres

OBJECTIVE

To investigate the accuracy of automated identification of cephalometric landmarks using the cascade convolutional neural networks (CNN) on lateral cephalograms acquired from nationwide multi-centres.

SETTINGS AND SAMPLE POPULATION

A total of 3150 lateral cephalograms were acquired from 10 university hospitals in South Korea for training.

MATERIALS AND METHODS

We evaluated the accuracy of the developed model with independent 100 lateral cephalograms as an external validation. Two orthodontists independently identified the anatomic landmarks of the test data set using the V-ceph software (version 8.0, Osstem, Seoul, Korea). The mean positions of the landmarks identified by two orthodontists were regarded as the gold standard. The performance of the CNN model was evaluated by calculating the mean absolute distance between the gold standard and the automatically detected positions. Factors associated with the detection accuracy for landmarks were analysed using the linear regression models.

RESULTS

The mean inter-examiner difference was 1.31 ± 1.13 mm. The overall automated detection error was 1.36 ± 0.98 mm. The mean detection error for each landmark ranged between 0.46 ± 0.37 mm (maxillary incisor crown tip) and 2.09 ± 1.91 mm (distal root tip of the mandibular first molar). A significant difference in the detection accuracy among cephalograms was noted according to hospital (P = .011), sensor type (P < .01), and cephalography machine model (P < .01).

CONCLUSION

The automated cephalometric landmark detection model may aid in preliminary screening for patient diagnosis and mid-treatment assessment, independent of the type of the radiography machines tested.

Comparing intra-observer variation and external variations of a fully automated cephalometric analysis with a cascade convolutional neural net

The quality of cephalometric analysis depends on the accuracy of the delineating landmarks in orthodontic and maxillofacial surgery. Due to the extensive number of landmarks, each analysis costs orthodontists considerable time per patient, leading to fatigue and inter- and intra-observer variabilities. Therefore, we proposed a fully automated cephalometry analysis with a cascade convolutional neural net (CNN). One thousand cephalometric x-ray images (2 k × 3 k) pixel were used. The dataset was split into training, validation, and test sets as 8:1:1. The 43 landmarks from each image were identified by an expert orthodontist. To evaluate intra-observer variabilities, 28 images from the dataset were randomly selected and measured again by the same orthodontist. To improve accuracy, a cascade CNN consisting of two steps was used for transfer learning. In the first step, the regions of interest (ROIs) were predicted by RetinaNet. In the second step, U-Net detected the precise landmarks in the ROIs. The average error of ROI detection alone was 1.55 ± 2.17 mm. The model with the cascade CNN showed an average error of 0.79 ± 0.91 mm (paired t-test, p = 0.0015). The orthodontist’s average error of reproducibility was 0.80 ± 0.79 mm. An accurate and fully automated cephalometric analysis was successfully developed and evaluated.

[Consistency of expert opinions on localization of the reference points for studying a soft tissue face profile in digital teleradiological images of the skull lateral projection]

OBJECTIVE

The aim of the study is to identify the consistency of expert opinions when manually identifying the reference points positions for studying the soft-tissue face profile in digital teleradiological images of the skull lateral projection.

MATERIAL AND METHODS

The study involved 11 orthodontists having a 1 to 8 year experience. They localized the reference points manually, using the software with image enhancement options. Each doctor processed 100 X-ray images of the skull lateral projection. Totally, 1100 positions were identified (11 parameters in 100 images). The average position of 11 manual localizations was taken as the baseline. Then, manually identified positions were automatically compared with the basic localization.

RESULTS

We have a good consistency of expert opinions when determining the reference points of the soft-tissue face profile. Herewith, the highest consistency of expert opinions is observed for soft-tissue reference points of the upper and lower lips (LL and EN points) with the average variation coefficients equal to 0.557726 and 0.566349 respectively; and the lowest consistency is observed for the hard tissue point Po with the average variation coefficient of 0.819904. Despite rather serious shortcomings in determining separate points (such as Or, DT, Po) in teleradiological images of the lateral skull projection, the method for determining soft-tissue profilometric reference points gives satisfactory results for more than 85.87% cases. In general, the method can be considered clinically reliable. However, the means and methods for identifying reference points require improving.

Learning-based local-to-global landmark annotation for automatic 3D cephalometry

The annotation of three-dimensional (3D) cephalometric landmarks in 3D computerized tomography (CT) has become an essential part of cephalometric analysis, which is used for diagnosis, surgical planning, and treatment evaluation. The automation of 3D landmarking with high-precision remains challenging due to the limited availability of training data and the high computational burden. This paper addresses these challenges by proposing a hierarchical deep-learning method consisting of four stages: 1) a basic landmark annotator for 3D skull pose normalization, 2) a deep-learning-based coarse-to-fine landmark annotator on the midsagittal plane, 3) a low-dimensional representation of the total number of landmarks using variational autoencoder (VAE), and 4) a local-to-global landmark annotator. The implementation of the VAE allows two-dimensional-image-based 3D morphological feature learning and similarity/dissimilarity representation learning of the concatenated vectors of cephalometric landmarks. The proposed method achieves an average 3D point-to-point error of 3.63 mm for 93 cephalometric landmarks using a small number of training CT datasets. Notably, the VAE captures variations of craniofacial structural characteristics.

The use and performance of artificial intelligence applications in dental and maxillofacial radiology: A systematic review

OBJECTIVES

To investigate the current clinical applications and diagnostic performance of artificial intelligence (AI) in dental and maxillofacial radiology (DMFR).

METHODS

Studies using applications related to DMFR to develop or implement AI models were sought by searching five electronic databases and four selected core journals in the field of DMFR. The customized assessment criteria based on QUADAS-2 were adapted for quality analysis of the studies included.

RESULTS

The initial electronic search yielded 1862 titles, and 50 studies were eventually included. Most studies focused on AI applications for an automated localization of cephalometric landmarks, diagnosis of osteoporosis, classification/segmentation of maxillofacial cysts and/or tumors, and identification of periodontitis/periapical disease. The performance of AI models varies among different algorithms.

CONCLUSION

The AI models proposed in the studies included exhibited wide clinical applications in DMFR. Nevertheless, it is still necessary to further verify the reliability and applicability of the AI models prior to transferring these models into clinical practice.

Automated identification of cephalometric landmarks: Part 1-Comparisons between the latest deep-learning methods YOLOV3 and SSD

OBJECTIVE

To compare the accuracy and computational efficiency of two of the latest deep-learning algorithms for automatic identification of cephalometric landmarks.

MATERIALS AND METHODS

A total of 1028 cephalometric radiographic images were selected as learning data that trained You-Only-Look-Once version 3 (YOLOv3) and Single Shot Multibox Detector (SSD) methods. The number of target labeling was 80 landmarks. After the deep-learning process, the algorithms were tested using a new test data set composed of 283 images. Accuracy was determined by measuring the point-to-point error and success detection rate and was visualized by drawing scattergrams. The computational time of both algorithms was also recorded.

RESULTS

The YOLOv3 algorithm outperformed SSD in accuracy for 38 of 80 landmarks. The other 42 of 80 landmarks did not show a statistically significant difference between YOLOv3 and SSD. Error plots of YOLOv3 showed not only a smaller error range but also a more isotropic tendency. The mean computational time spent per image was 0.05 seconds and 2.89 seconds for YOLOv3 and SSD, respectively. YOLOv3 showed approximately 5% higher accuracy compared with the top benchmarks in the literature.

CONCLUSIONS

Between the two latest deep-learning methods applied, YOLOv3 seemed to be more promising as a fully automated cephalometric landmark identification system for use in clinical practice.

Automatic 3D cephalometric annotation system using shadowed 2D image-based machine learning

This paper presents a new approach to automatic three-dimensional (3D) cephalometric annotation for diagnosis, surgical planning, and treatment evaluation. There has long been considerable demand for automated cephalometric landmarking, since manual landmarking requires considerable time and experience as well as objectivity and scrupulous error avoidance. Due to the inherent limitation of two-dimensional (2D) cephalometry and the 3D nature of surgical simulation, there is a trend away from current 2D to 3D cephalometry. Deep learning approaches to cephalometric landmarking seem highly promising, but there exist serious difficulties in handling high dimensional 3D CT data, dimension referring to the number of voxels. To address this issue of dimensionality, this paper proposes a shadowed 2D image-based machine learning method which uses multiple shadowed 2D images with various lighting and view directions to capture 3D geometric cues. The proposed method using VGG-net was trained and tested using 2700 shadowed 2D images and corresponding manual landmarkings. Test data evaluation shows that our method achieved an average point-to-point error of 1.5 mm for the seven major landmarks.

Rapid automated landmarking for morphometric analysis of three-dimensional facial scans

Automated phenotyping is essential for the creation of large, highly standardized datasets from anatomical imaging data. Such datasets can support large-scale studies of complex traits or clinical studies related to precision medicine or clinical trials. We have developed a method that generates three-dimensional landmark data that meet the requirements of standard geometric morphometric analyses. The method is robust and can be implemented without high-performance computing resources. We validated the method using both direct comparison to manual landmarking on the same individuals and also analyses of the variation patterns and outlier patterns in a large dataset of automated and manual landmark data. Direct comparison of manual and automated landmarks reveals that automated landmark data are less variable, but more highly integrated and reproducible. Automated data produce covariation structure that closely resembles that of manual landmarks. We further find that while our method does produce some landmarking errors, they tend to be readily detectable and can be fixed by adjusting parameters used in the registration and control-point steps. Data generated using the method described here have been successfully used to study the genomic architecture of facial shape in two different genome-wide association studies of facial shape.

Ceph-X: development and evaluation of 2D cephalometric system

Background

Cephalometric analysis and measurements of skull parameters using X-Ray images plays an important role in predicating and monitoring orthodontic treatment. Manual analysis and measurements of cephalometric is considered tedious, time consuming, and subjected to human errors. Several cephalometric systems have been developed to automate the cephalometric procedure; however, no clear insights have been reported about reliability, performance, and usability of those systems. This study utilizes some techniques to evaluate reliability, performance, and usability metric using SUS methods of the developed cephalometric system which has not been reported in previous studies.

Methods

In this study a novel system named Ceph-X is developed to computerize the manual tasks of orthodontics during cephalometric measurements. Ceph-X is developed by using image processing techniques with three main models: enhancements X-ray image model, locating landmark model, and computation model. Ceph-X was then evaluated by using X-ray images of 30 subjects (male and female) obtained from University of Malaya hospital. Three orthodontics specialists were involved in the evaluation of accuracy to avoid intra examiner error, and performance for Ceph-X, and 20 orthodontics specialists were involved in the evaluation of the usability, and user satisfaction for Ceph-X by using the SUS approach.

Results

Statistical analysis for the comparison between the manual and automatic cephalometric approaches showed that Ceph-X achieved a great accuracy approximately 96.6%, with an acceptable errors variation approximately less than 0.5 mm, and 1°. Results showed that Ceph-X increased the specialist performance, and minimized the processing time to obtain cephalometric measurements of human skull. Furthermore, SUS analysis approach showed that Ceph-X has an excellent usability user’s feedback.

Conclusions

The Ceph-X has proved its reliability, performance, and usability to be used by orthodontists for the analysis, diagnosis, and treatment of cephalometric.

Fully Automatic System for Accurate Localisation and Analysis of Cephalometric Landmarks in Lateral Cephalograms

Cephalometric tracing is a standard analysis tool for orthodontic diagnosis and treatment planning. The aim of this study was to develop and validate a fully automatic landmark annotation (FALA) system for finding cephalometric landmarks in lateral cephalograms and its application to the classification of skeletal malformations. Digital cephalograms of 400 subjects (age range: 7-76 years) were available. All cephalograms had been manually traced by two experienced orthodontists with 19 cephalometric landmarks, and eight clinical parameters had been calculated for each subject. A FALA system to locate the 19 landmarks in lateral cephalograms was developed. The system was evaluated via comparison to the manual tracings, and the automatically located landmarks were used for classification of the clinical parameters. The system achieved an average point-to-point error of 1.2 mm, and 84.7% of landmarks were located within the clinically accepted precision range of 2.0 mm. The automatic landmark localisation performance was within the inter-observer variability between two clinical experts. The automatic classification achieved an average classification accuracy of 83.4% which was comparable to an experienced orthodontist. The FALA system rapidly and accurately locates and analyses cephalometric landmarks in lateral cephalograms, and has the potential to significantly improve the clinical work flow in orthodontic treatment.

Accuracy of computerized automatic identification of cephalometric landmarks by a designed software

OBJECTIVES

The purpose of this study was to design software for localization of cephalometric landmarks and to evaluate its accuracy in finding landmarks.

METHODS

40 digital cephalometric radiographs were randomly selected. 16 landmarks which were important in most cephalometric analyses were chosen to be identified. Three expert orthodontists manually identified landmarks twice. The mean of two measurements of each landmark was defined as the baseline landmark. The computer was then able to compare the automatic system's estimate of a landmark with the baseline landmark. The software was designed using Delphi and Matlab programming languages. The techniques were template matching, edge enhancement and some accessory techniques.

RESULTS

The total mean error between manually identified and automatically identified landmarks was 2.59 mm. 12.5% of landmarks had mean errors less than 1 mm. 43.75% of landmarks had mean errors less than 2 mm. The mean errors of all landmarks except the anterior nasal spine were less than 4 mm.

CONCLUSIONS

This software had significant accuracy for localization of cephalometric landmarks and could be used in future applications. It seems that the accuracy obtained with the software which was developed in this study is better than previous automated systems that have used model-based and knowledge-based approaches.

Automatic craniofacial structure detection on cephalometric images

Anatomical structure tracing on cephalograms is a significant way to obtain cephalometric analysis. Cephalometric analysis is divided in two categories, manual and automatic approaches. The manual approach is limited in accuracy and repeatability due to differences in inter- and intra-personal marking. In this paper, we have attempted to develop and test a novel method for automatic localization of craniofacial structures based on the detected edges in the region of interest. Before edge detection of the particular region, the region was filtered by adaptive non local filter for noise removal by keeping the edge information undisturbed. According to the gray-scale feature at the different regions of the cephalograms, modified Canny edge detection algorithm for obtaining tissue contour was proposed. With the application of morphological opening and edge linking approaches, an improved bidirectional contour tracing methodology was proposed by an interactive selection of the starting edge pixels, the tracking process searches repetitively for an edge pixel at the neighborhood of previously searched edge pixel to segment images, and then craniofacial structures are obtained. The effectiveness of the algorithm is demonstrated by the preliminary experimental results obtained with the proposed method.

Automatic landmarking of cephalograms using active appearance models

There have been many attempts to further improve and automate cephalometric analysis in order to increase accuracy, reduce errors due to subjectivity, and to provide more efficient use of clinicians' time. The aim of this research was to evaluate an automated system for landmarking of cephalograms based on the use of an active appearance model (AAM) that contains a statistical model of shape and grey-level appearance of an object of interest and represents both shape and texture variations of the region covered by the model. Multi-resolution implementation was used, in which the AAM iterate to convergence at each level before projecting the current solution to the next level of the model. The AAM system was trained using 60 randomly selected, hand-annotated digital cephalograms of subjects between 7.2 and 25.6 years of age, and tested with a leave-five-out method that enabled testing not only of the accuracy of the AAM system but also the accuracy of each AAM. Differences between methods were examined using the non-parametric Wilcoxon signed rank test. An average accuracy of 1.68 mm was obtained, with 61 per cent of landmarks detected within 2 mm and 95 per cent of landmarks detected within 5 mm precision. A noticeable increase in overall precision and detection of low-contrast cephalometric landmarks was achieved compared with other automated systems. These results suggest that the AAM approach can adequately represent the average shape and texture variations of craniofacial structures on digital radiographs. As such it can successfully be implemented for automatic localization of cephalometric landmarks.

Automated cephalometry: system performance reliability using landmark-dependent criteria

OBJECTIVE

The purpose of the present study was to evaluate reliability of a system that performs automatic recognition of anatomic landmarks and adjacent structures on lateral cephalograms using landmark-dependent criteria unique to each landmark.

MATERIALS AND METHODS

To evaluate the reliability of the system, the system was used to examine 65 lateral cephalograms. The area of each system-identified anatomic structure surrounding the landmark and the position of each system-identified landmark were compared with norms using confidence ellipses with alpha = .01, which were derived from the scattergrams of 100 estimates obtained according to the method reported by Baumrind and Frantz. When the system-identified area overlapped with the norm area, anatomic structure recognition was considered successful. In addition, when the system-identified point was located within the norm area, landmark identification was considered successful. Based on these judgment criteria, success rates were calculated for all landmarks.

RESULTS

The system successfully identified all specified anatomic structures in all the images and determined the positions of the landmarks with a mean success rate of 88% (range, 77%- 100%).

CONCLUSION

With the incorporation of the rational assessment criteria provided by confidence ellipses, the proposed system was confirmed to be reliable.

Automatic cephalometric evaluation of patients suffering from sleep-disordered breathing

We address the problem of automatically analyzing lateral cephalometric images as a diagnostic tool for patients suffering from Sleep Disordered Breathing (SDB). First, multiple landmarks and anatomical structures that were previously associated with SDB are localized. Then statistical regression is applied in order to estimate the Respiratory Disturbance Index (RDI), which is the standard measure for the severity of obstructive sleep apnea. The landmark localization employs a new registration method that is based on Local Affine Frames (LAF). Multiple LAFs are sampled per image based on random selection of triplets of keypoints, and are used to register the input image to the training images. The landmarks are then projected from the training images to the query image. Following a refinement step, the tongue, velum and pharyngeal wall are localized. We collected a dataset of 70 images and compare the accuracy of the anatomical landmarks with recent publications, showing preferable performance in localizing most of the anatomical points. Furthermore, we are able to show that the location of the anatomical landmarks and structures predicts the severity of the disorder, obtaining an error of less than 7.5 RDI units for 44% of the patients.

An evaluation of cellular neural networks for the automatic identification of cephalometric landmarks on digital images

Several efforts have been made to completely automate cephalometric analysis by automatic landmark search. However, accuracy obtained was worse than manual identification in every study. The analogue-to-digital conversion of X-ray has been claimed to be the main problem. Therefore the aim of this investigation was to evaluate the accuracy of the Cellular Neural Networks approach for automatic location of cephalometric landmarks on softcopy of direct digital cephalometric X-rays. Forty-one, direct-digital lateral cephalometric radiographs were obtained by a Siemens Orthophos DS Ceph and were used in this study and 10 landmarks (N, A Point, Ba, Po, Pt, B Point, Pg, PM, UIE, LIE) were the object of automatic landmark identification. The mean errors and standard deviations from the best estimate of cephalometric points were calculated for each landmark. Differences in the mean errors of automatic and manual landmarking were compared with a 1-way analysis of variance. The analyses indicated that the differences were very small, and they were found at most within 0.59 mm. Furthermore, only few of these differences were statistically significant, but differences were so small to be in most instances clinically meaningless. Therefore the use of X-ray files with respect to scanned X-ray improved landmark accuracy of automatic detection. Investigations on softcopy of digital cephalometric X-rays, to search more landmarks in order to enable a complete automatic cephalometric analysis, are strongly encouraged.

Automatic cephalometric analysis

OBJECTIVE

To describe the techniques used for automatic landmarking of cephalograms, highlighting the strengths and weaknesses of each one and reviewing the percentage of success in locating each cephalometric point.

MATERIALS AND METHODS

The literature survey was performed by searching the Medline, the Institute of Electrical and Electronics Engineers, and the ISI Web of Science Citation Index databases. The survey covered the period from January 1966 to August 2006. Abstracts that appeared to fulfill the initial selection criteria were selected by consensus. The original articles were then retrieved. Their references were also hand-searched for possible missing articles. The search strategy resulted in 118 articles of which eight met the inclusion criteria. Many articles were rejected for different reasons; among these, the most frequent was that results of accuracy for automatic landmark recognition were presented as a percentage of success.

RESULTS

A marked difference in results was found between the included studies consisting of heterogeneity in the performance of techniques to detect the same landmark. All in all, hybrid approaches detected cephalometric points with a higher accuracy in contrast to the results for the same points obtained by the model-based, image filtering plus knowledge-based landmark search and "soft-computing" approaches.

CONCLUSIONS

The systems described in the literature are not accurate enough to allow their use for clinical purposes. Errors in landmark detection were greater than those expected with manual tracing and, therefore, the scientific evidence supporting the use of automatic landmarking is low.

Digital diagnostics: Three-dimensional modelling

Three-dimensional imaging techniques, such as computed tomograms (CT), structured light, and stereophotogrammetry, can be used to capture three-dimensional coordinate data, but comprehensive analysis is required to transform these techniques into powerful diagnostic tools. The object of this review is to highlight analytical functionality using software developed to study three-dimensional digital imaging and communications in medicine (DICOM) based digital data for diagnosis, planning of treatment, and evaluation of craniofacial changes. My specific aim was to apply three-dimensional software routines using geometric morphometrics or conventional measurements. These routines rely on robust algorithms to construct mean objects by manipulating the three-dimensional x, y, and z coordinates of all the objects' vertices. Conventional measurements and statistical tests can then be applied to the changes in the vertices, say, before and after treatment. Using graphical and geometric morphometric techniques such as finite-element analysis and principal components analysis, clinical craniofacial modelling can be used for the localisation and quantification of soft and hard tissue changes; diagnostic modelling can be undertaken for planning of treatment, and data-driven predictive modelling can be undertaken for the planning of many procedures based on the surgeon's own experience, patients, and resources. Three-dimensional modelling of digital data may therefore have added value for clinical diagnosis, and planning and assessment of treatment, including audit.

Learning active shape models for bifurcating contours

Statistical shape models are often learned from examples based on landmark correspondences between annotated examples. A method is proposed for learning such models from contours with inconsistent bifurcations and loops. Automatic segmentation of tibial and femoral contours in knee X-ray images is investigated as a step towards reliable, quantitative radiographic analysis of osteoarthritis for diagnosis and assessment of progression. Results are presented using various features, the Mahalanobis distance, distance weighted K-nearest neighbours, and two relevance vector machine-based methods as quality of fit measure.

An approach for the automatic cephalometric landmark detection using mathematical morphology and active appearance models

Cephalometric analysis of lateral radiographs of the head is an important diagnosis tool in orthodontics. Based on manually locating specific landmarks, it is a tedious, time-consuming and error prone task. In this paper, we propose an automated system based on the use of Active Appearance Models (AAMs). Special attention has been paid to clinical validation of our method since previous work in this field used few images, was tested in the training set and/or did not take into account the variability of the images. In this research, a top-hat transformation was used to correct the intensity inhomogeneity of the radiographs generating a consistent training set that overcomes the above described drawbacks. The AAM was trained using 96 hand-annotated images and tested with a leave-one-out scheme obtaining an average accuracy of 2.48mm. Results show that AAM combined with mathematical morphology is the suitable method for clinical cephalometric applications.

Automated 2-D cephalometric analysis on X-ray images by a model-based approach

Craniofacial landmark localization and anatomical structure tracing on cephalograms are two important ways to obtain the cephalometric analysis. In order to computerize them in parallel, a model-based approach is proposed to locate 262 craniofacial feature points, including 90 landmarks and 172 auxiliary points. In model training, 12 landmarks are selected as reference points and used to divide every training shape to 10 regions according to the anatomical knowledge; principle components analysis is employed to characterize the region shape variations and the statistical grey profile of every feature point. Locating feature points on an input image is a two-stage procedure. First, we identify the reference landmarks by image processing and pattern matching techniques, so that the shape partition is performed on the input image. Then, for each region, its feature points are located by a modified active shape model. All craniofacial anatomical structures can be traced out by connecting the located points with subdivision curves according to the prior knowledge. Users are permitted to modify the results interactively in many different ways. Experimental results show the advantage and reliability of the proposed method.

Image processing for craniofacial landmark identification and measurement: a review of photogrammetry and cephalometry

Facial surface anthropometry and cephalometry have been used for many years for the diagnosis of malformations, surgical planning and evaluation, and growth studies. These disciplines rely on the identification of craniofacial landmarks. Methods for 3D reconstruction of landmarks have been introduced, as have image processing algorithms for the automation of landmark extraction. This paper reviews facial surface anthropometry and cephalometry with reference to the image processing algorithms that have been applied and their effectiveness.

Computational models of oral and craniofacial development, growth, and repair

This paper illustrates how biological and clinical problems stimulate research in biomedical informatics and how such research contributes to their solution. The computational models described use techniques from Logic Programming, Machine Learning, Computer Vision, and Biomathematics. They address problems in the development, growth, and repair of oral and craniofacial tissues arising in cell biology, clinical genetics, and dentistry. At the micro-level, the dynamic interaction of cells in the oral epithelium is modeled. At the macro-level, models are constructed of either the craniofacial shape of an individual or the craniofacial shape differences within and between healthy and congenitally abnormal populations. In between, in terms of scale, there are models of normal dentition and the use of computerized expert knowledge to guide the design of dental prostheses used to restore function in partially edentulous patients.

Combining biometric and symbolic models for customized, automated prosthesis design

In a previous paper [Artif. Intell. Med. 5 (1993) 431] we described RaPiD, a knowledge-based system for designing dental prostheses. The present paper discusses how RaPiD has been extended using techniques from computer vision and logic grammars. The first employs point distribution and active shape models (ASMs) to determine dentition from images of casts of patient's jaws. This enables a design to be customized to, and visualised against, an image of a patient's dentition. The second is based on the notion of a path grammar, a form of logic grammar, to generate a path linking an ordered sequence of subcomponents. The shape of an important and complex prosthesis component can be automatically seeded in this fashion. Combining these models now substantially automates the design process, beginning with a photograph of a dental cast and ending with an annotated and validated design diagram ready to guide manufacture.