Comparison of Deep Learning Models for Cervical Vertebral Maturation Stage Classification on Lateral Cephalometric Radiographs

Artificial intelligence for orthodontic diagnosis and treatment planning: A scoping review

OBJECTIVES

To provide an overview of artificial intelligence (AI) applications in orthodontic diagnosis and treatment planning, and to evaluate whether AI improves accuracy, reliability, and time efficiency compared to expert-based manual approaches, while highlighting its current limitations.

DATA

This review follows the Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR) Checklist.

SOURCES

An electronic search was performed on PubMed, Web of Science, and Embase electronic databases. Additional studies were identified from Google Scholar and by hand searching through included studies. The search was carried out until June 2023 without restriction of language and publication year.

STUDY SELECTION

After applying the selection criteria, 71 articles were included in the review. The main research areas were classified into three domains based on the purpose of AI: diagnostics (n = 29), landmark identification (n = 20) and treatment planning (n = 22).

CONCLUSION

This scoping review shows that AI can be used in various orthodontic diagnosis and treatment planning applications, with anatomical landmark detection being the most studied domain. While AI shows potential in improving time efficiency and reducing operator variability, the accuracy and reliability have not yet consistently surpassed those of expert clinicians. At all moments, human supervision remains essential. Further advances and optimizations are necessary to strive towards automated patient-specific treatment planning.

CLINICAL SIGNIFICANCE

AI in orthodontics has shown its ability to serve as a decision-support system, thereby enhancing the efficiency of diagnostics and treatment planning within orthodontics digital workflow."

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.

An explainable machine learning estimated biological age based on morphological parameters of the spine

Accurately estimating biological age is beneficial for measuring aging and predicting risk. It is widely accepted that the prevalence of spine compression increases significantly with age. However, biological age based on vertebral morphological data is rarely reported. In this study, a total of 2,364 participants from the National Health and Nutrition Examination Survey were enrolled, and morphological parameters of the spine were collected from lateral radiographs scanned by dual energy X-ray absorptiometry. The biological age of the spine, called SpineAge, was calculated with the parameters by machine learning models. The SHapley Additive exPlanation was used for better interpreting each parameter's contribution. Besides, an Accelerated Aging Index (AAI) was defined as SpineAge minus chronological age and was used to quantify the accelerating aging degree of the spine. The results indicated that the SpineAge performed better than chronological age did in predicting 2-year and 5-year all-cause mortality. After adjusting all covariates, there was a significant association between AAI and all-cause mortality risk. Specifically, each 1-year increase in AAI was associated with a 25.9% increase in all-cause mortality risk (Hazards ratio, 1.259; 95% CI, 1.087-1.457; P < 0.001). Considering the first quartile of AAI as a reference, the mortality risks for the second, third, and fourth quartiles were 2.389 (95% CI, 1.064-5.364; P = 0.035), 5.911 (95% CI, 2.241-15.590; P < 0.001) and 22.925 (95% CI, 4.744-110.769; P < 0.001) times higher, respectively. Our study developed a novel and highly applicable biological-age predictor for predicting individualized long-term prognosis and facilitating personalized care.

Classification of cervical vertebral maturation stages with machine learning models: leveraging datasets with high inter- and intra-observer agreement

OBJECTIVES

This study aimed to assess the accuracy of machine learning (ML) models with feature selection technique in classifying cervical vertebral maturation stages (CVMS). Consensus-based datasets were used for models training and evaluation for their model generalization capabilities on unseen datasets.

METHODS

Three clinicians independently rated CVMS on 1380 lateral cephalograms, resulting in the creation of five datasets: two consensus-based datasets (Complete Agreement and Majority Voting), and three datasets based on a single rater's evaluations. Additionally, landmarks annotation of the second to fourth cervical vertebrae and patients' information underwent a feature selection process. These datasets were used to train various ML models and identify the top-performing model for each dataset. These models were subsequently tested on their generalization capabilities.

RESULTS

Features that considered significant in the consensus-based datasets were consistent with a CVMS guideline. The Support Vector Machine model on the Complete Agreement dataset achieved the highest accuracy (77.4%), followed by the Multi-Layer Perceptron model on the Majority Voting dataset (69.6%). Models from individual ratings showed lower accuracies (60.4-67.9%). The consensus-based training models also exhibited lower coefficient of variation (CV), indicating superior generalization capability compared to models from single raters.

CONCLUSION

ML models trained on consensus-based datasets for CVMS classification exhibited the highest accuracy, with significant features consistent with the original CVMS guidelines. These models also showed robust generalization capabilities, underscoring the importance of dataset quality.

Artificial Intelligence-Related Dental Research: Bibliometric and Altmetric Analysis

BACKGROUND

Recent years have witnessed an explosive surge in dental research related to artificial intelligence (AI). These applications have optimised dental workflows, demonstrating significant clinical importance. Understanding the current landscape and trends of this topic is crucial for both clinicians and researchers to utilise and advance this technology. However, a comprehensive scientometric study regarding this field had yet to be performed.

METHODS

A literature search was conducted in the Web of Science Core Collection database to identify eligible "research articles" and "reviews." Literature screening and exclusion were performed by 2 investigators. Thereafter, VOSviewer was utilised in co-occurrence analysis and CiteSpace in co-citation analysis. R package Bibliometrix was employed to automatically calculate scientific impacts, determining the core authors and journals. Altmetric data were described narratively and supplemented with Spearman correlation analysis.

RESULTS

A total of 1558 research publications were included. During the past 5 years, AI-related dental publications drastically increased in number, from 36 to 581. Diagnostics and Scientific Reports published the most articles, whereas Journal of Dental Research received the highest number of citations per article. China, the US, and South Korea emerged as the most prolific countries, whilst Germany received the highest number of citations per article (23.29). Charité Universitätsmedizin Berlin was the institution with the highest number of publications and citations per article (29.16). Altmetric Attention Score was correlated with News Mentions (P < .001), and significant associations were observed amongst Dimension Citations, Mendeley Readers, and Web of Science Citations (P < .001).

CONCLUSIONS

The publication numbers regarding AI-related dental research have been rising rapidly and may continue their upwards trend. China, the US, South Korea, and Germany had promoted the progress of AI-related dental research. Disease diagnosis, orthodontic applications, and morphology segmentation were current hotspots. Attention mechanism, explainable AI, multimodal data fusion, and AI-generated text assistants necessitate future research and exploration.

A questionnaire study regarding knowledge, attitude and usage of artificial intelligence and machine learning by the orthodontic fraternity of Northern India

Aim

The aim of the questionnaire study was to determine the knowledge, attitude, and perception of orthodontists regarding the role of artificial intelligence in dentistry in general and orthodontics specifically, and to determine the use of artificial intelligence by the orthodontist.

Methods

This cross-sectional study was done among the orthodontists of Northern India (clinicians, academicians, and postgraduates) through a web-based electronic survey using Google Forms. The study was designed to obtain information about AI and its basic usage in daily life, in dentistry, and in orthodontics from the participants. The options given were set specifically according to the Likert scale to maintain the correct format. The questionnaire was validated by one AI expert and one orthodontic expert, followed by pretesting in a smaller group of 25 orthodontists 2 weeks before circulation. A total of 100 orthodontists and postgraduate students responded to the pretested online questionnaire link for 31 questions in four sections sent via social media websites in a period of 3 months.

Results

The majority of the participants believe that AI could be useful in diagnosis and treatment planning and could revolutionize dentistry in general. 84 % of the orthodontic academicians and clinicians, including PG students, consider AI a useful tool for boosting performance and delivering quality care in orthodontics, and 72 % see AI as a partner rather than a competitor in the foreseeable future of dentistry. 90 % of the participants believe that the incorporation of AI into CBCT analysis can be a valuable addition to diagnosis and treatment planning. 86 % of total participants agree that AI can be helpful in decision-making for orthognathic surgery, and 84 % find AI useful for bone age assessment.

Conclusions

It was observed that academicians are more aware of AI terminologies and usage as compared to PG students and clinicians. There is a consensus that AI is a useful tool for diagnosis and treatment planning, boosting performance and quality care in orthodontics. In spite of these facts, 62.5 % of clinicians and 40 % of PG students are still not using AI for cephalometric analysis (p = 0.033).

An Automatic Grading System for Orthodontically Induced External Root Resorption Based on Deep Convolutional Neural Network

Orthodontically induced external root resorption (OIERR) is a common complication of orthodontic treatments. Accurate OIERR grading is crucial for clinical intervention. This study aimed to evaluate six deep convolutional neural networks (CNNs) for performing OIERR grading on tooth slices to construct an automatic grading system for OIERR. A total of 2146 tooth slices of different OIERR grades were collected and preprocessed. Six pre-trained CNNs (EfficientNet-B1, EfficientNet-B2, EfficientNet-B3, EfficientNet-B4, EfficientNet-B5, and MobileNet-V3) were trained and validated on the pre-processed images based on four different cross-validation methods. The performances of the CNNs on a test set were evaluated and compared with those of orthodontists. The gradient-weighted class activation mapping (Grad-CAM) technique was used to explore the area of maximum impact on the model decisions in the tooth slices. The six CNN models performed remarkably well in OIERR grading, with a mean accuracy of 0.92, surpassing that of the orthodontists (mean accuracy of 0.82). EfficientNet-B4 trained with fivefold cross-validation emerged as the final OIERR grading system, with a high accuracy of 0.94. Grad-CAM revealed that the apical region had the greatest effect on the OIERR grading system. The six CNNs demonstrated excellent OIERR grading and outperformed orthodontists. The proposed OIERR grading system holds potential as a reliable diagnostic support for orthodontists in clinical practice.

Accuracy of Artificial Intelligence for Cervical Vertebral Maturation Assessment-A Systematic Review

Background/Objectives: To systematically review and summarize the existing scientific evidence on the diagnostic performance of artificial intelligence (AI) in assessing cervical vertebral maturation (CVM). This review aimed to evaluate the accuracy and reliability of AI algorithms in comparison to those of experienced clinicians. Methods: Comprehensive searches were conducted across multiple databases, including PubMed, Scopus, Web of Science, and Embase, using a combination of Boolean operators and MeSH terms. The inclusion criteria were cross-sectional studies with neural network research, reporting diagnostic accuracy, and involving human subjects. Data extraction and quality assessment were performed independently by two reviewers, with a third reviewer resolving any disagreements. The Quality Assessment of Diagnostic Accuracy Studies (QUADAS)-2 tool was used for bias assessment. Results: Eighteen studies met the inclusion criteria, predominantly employing supervised learning techniques, especially convolutional neural networks (CNNs). The diagnostic accuracy of AI models for CVM assessment varied widely, ranging from 57% to 95%. The factors influencing accuracy included the type of AI model, training data, and study methods. Geographic concentration and variability in the experience of radiograph readers also impacted the results. Conclusions: AI has considerable potential for enhancing the accuracy and reliability of CVM assessments in orthodontics. However, the variability in AI performance and the limited number of high-quality studies suggest the need for further research.

Efficacy of the methods of age determination using artificial intelligence in panoramic radiographs - a systematic review

The aim of this systematic review is to analyze the literature to determine whether the methods of artificial intelligence are effective in determining age in panoramic radiographs. Searches without language and year limits were conducted in PubMed/Medline, Embase, Web of Science, and Scopus databases. Hand searches were also performed, and unpublished manuscripts were searched in specialized journals. Thirty-six articles were included in the analysis. Significant differences in terms of root mean square error and mean absolute error were found between manual methods and artificial intelligence techniques, favoring the use of artificial intelligence (p < 0.00001). Few articles compared deep learning methods with machine learning models or manual models. Although there are advantages of machine learning in data processing and deep learning in data collection and analysis, non-comparable data was a limitation of this study. More information is needed on the comparison of these techniques, with particular emphasis on time as a variable.

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.

CURRENT NEURAL NETWORKS DEMONSTRATE POTENTIAL IN AUTOMATED CERVICAL VERTEBRAL MATURATION STAGE CLASSIFICATION BASED ON LATERAL CEPHALOGRAMS

ARTICLE TITLE AND BIBLIOGRAPHIC INFORMATION

Neural networks for classification of cervical vertebrae maturation: a systematic review. Mathew R, Palatinus S, Padala S, Alshehri A, Awadh W, Bhandi S, Thomas J, Patil S. Angle Orthod. 2022 Nov 1;92(6):796-804.

SOURCE OF FUNDING

No financial support was reported.

TYPE OF STUDY/DESIGN

Systematic review.

Convolutional neural network-assisted diagnosis of midpalatal suture maturation stage in cone-beam computed tomography

OBJECTIVES

The selection of treatment for maxillary expansion is closely related to the calcification degree of the midpalatal suture. A classification method for individual assessment of the morphology of midpalatal suture in cone-beam computed tomography (CBCT) is useful for evaluating the calcification degree. Currently, convolutional neural networks (CNNs) have been introduced into the field of oral and maxillofacial imaging diagnosis. This study validated the ability of CNN models in assessing the maturation stage of the midpalatal suture.

METHODS

The existing CNN model ResNet50 was trained to locate the CBCT transverse plane which contained a complete midpalatal suture. ResNet18, ResNet50, RessNet101, Inception-v3, and Efficientnetv2-s models were trained to evaluate the midpalatal suture maturation stage. Multi-class classification metrics, accuracy, recall, precision, F1-score, and area under the curve values from the receiver operating characteristic curve were used to evaluate the performance of the models, and gradient-weighted class activation map technology was utilised to visualise five midpalatal suture maturation stages for each model.

RESULTS

Resnet50 demonstrated an accuracy of 99.74 % in identifying the transverse plane that contained the complete midpalatal suture. The highest accuracies achieved on the two-stage, three-stage, and five-stage maturation classification tests were 95.15, 88.06, and 75.37 %, all of which exceeded the average accuracy of three experienced orthodontists.

CONCLUSIONS

The CNN model can locate the plane of the midpalatal suture in CBCT images and can assist clinicians in assessing the maturation stage of the midpalatal suture to select the means of maxillary expansion.

CLINICAL SIGNIFICANCE

The application of artificial intelligence on CBCT midpalatal suture plane localisation and maturation stage evaluation enhances diagnostic and treatment efficiency and accuracy of individual assessment of midpalatal suture calcification degree. Additionally, it assists the clinical palatal expansion technique in achieving ideal results.

AI in Orthodontics: Revolutionizing Diagnostics and Treatment Planning-A Comprehensive Review

The advent of artificial intelligence (AI) in medicine has transformed various medical specialties, including orthodontics. AI has shown promising results in enhancing the accuracy of diagnoses, treatment planning, and predicting treatment outcomes. Its usage in orthodontic practices worldwide has increased with the availability of various AI applications and tools. This review explores the principles of AI, its applications in orthodontics, and its implementation in clinical practice. A comprehensive literature review was conducted, focusing on AI applications in dental diagnostics, cephalometric evaluation, skeletal age determination, temporomandibular joint (TMJ) evaluation, decision making, and patient telemonitoring. Due to study heterogeneity, no meta-analysis was possible. AI has demonstrated high efficacy in all these areas, but variations in performance and the need for manual supervision suggest caution in clinical settings. The complexity and unpredictability of AI algorithms call for cautious implementation and regular manual validation. Continuous AI learning, proper governance, and addressing privacy and ethical concerns are crucial for successful integration into orthodontic practice.

Artificial Intelligence and Its Clinical Applications in Orthodontics: A Systematic Review

This review aims to analyze different strategies that make use of artificial intelligence to enhance diagnosis, treatment planning, and monitoring in orthodontics. Orthodontics has seen significant technological advancements with the introduction of digital equipment, including cone beam computed tomography, intraoral scanners, and software coupled to these devices. The use of deep learning in software has sped up image processing processes. Deep learning is an artificial intelligence technology that trains computers to analyze data like the human brain does. Deep learning models are capable of recognizing complex patterns in photos, text, audio, and other data to generate accurate information and predictions.

MATERIALS AND METHODS

Pubmed, Scopus, and Web of Science were used to discover publications from 1 January 2013 to 18 October 2023 that matched our topic. A comparison of various artificial intelligence applications in orthodontics was generated.

RESULTS

A final number of 33 studies were included in the review for qualitative analysis.

CONCLUSIONS

These studies demonstrate the effectiveness of AI in enhancing orthodontic diagnosis, treatment planning, and assessment. A lot of articles emphasize the integration of artificial intelligence into orthodontics and its potential to revolutionize treatment monitoring, evaluation, and patient outcomes.

AggregateNet: A deep learning model for automated classification of cervical vertebrae maturation stages

OBJECTIVE

A study of supervised automated classification of the cervical vertebrae maturation (CVM) stages using deep learning (DL) network is presented. A parallel structured deep convolutional neural network (CNN) with a pre-processing layer that takes X-ray images and the age as the input is proposed.

METHODS

A total of 1018 cephalometric radiographs were labelled and classified according to the CVM stages. The images were separated according to gender for better model-fitting. The images were cropped to extract the cervical vertebrae automatically using an object detector. The resulting images and the age inputs were used to train the proposed DL model: AggregateNet with a set of tunable directional edge enhancers. After the features of the images were extracted, the age input was concatenated to the output feature vector. To have the parallel network not overfit, data augmentation was used. The performance of our CNN model was compared with other DL models, ResNet20, Xception, MobileNetV2 and custom-designed CNN model with the directional filters.

RESULTS

The proposed innovative model that uses a parallel structured network preceded with a pre-processing layer of edge enhancement filters achieved a validation accuracy of 82.35% in CVM stage classification on female subjects, 75.0% in CVM stage classification on male subjects, exceeding the accuracy achieved with the other DL models investigated. The effectiveness of the directional filters is reflected in the improved performance attained in the results. If AggregateNet is used without directional filters, the test accuracy decreases to 80.0% on female subjects and to 74.03% on male subjects.

CONCLUSION

AggregateNet together with the tunable directional edge filters is observed to produce higher accuracy than the other models that we investigated in the fully automated determination of the CVM stages.

Application of Artificial Intelligence in Orthodontics: Current State and Future Perspectives

In recent years, there has been the notable emergency of artificial intelligence (AI) as a transformative force in multiple domains, including orthodontics. This review aims to provide a comprehensive overview of the present state of AI applications in orthodontics, which can be categorized into the following domains: (1) diagnosis, including cephalometric analysis, dental analysis, facial analysis, skeletal-maturation-stage determination and upper-airway obstruction assessment; (2) treatment planning, including decision making for extractions and orthognathic surgery, and treatment outcome prediction; and (3) clinical practice, including practice guidance, remote care, and clinical documentation. We have witnessed a broadening of the application of AI in orthodontics, accompanied by advancements in its performance. Additionally, this review outlines the existing limitations within the field and offers future perspectives.

Deep convolutional neural network-the evaluation of cervical vertebrae maturation

OBJECTIVES

This study aimed to automatically determine the cervical vertebral maturation (CVM) processes on lateral cephalometric radiograph images using a proposed deep learning-based convolutional neural network (CNN) model and to test the success rate of this CNN model in detecting CVM stages using precision, recall, and F1-score.

METHODS

A total of 588 digital lateral cephalometric radiographs of patients with a chronological age between 8 and 22 years were included in this study. CVM evaluation was carried out by two dentomaxillofacial radiologists. CVM stages in the images were divided into 6 subgroups according to the growth process. A convolutional neural network (CNN) model was developed in this study. Experimental studies for the developed model were carried out in the Jupyter Notebook environment using the Python programming language, the Keras, and TensorFlow libraries.

RESULTS

As a result of the training that lasted 40 epochs, 58% training and 57% test accuracy were obtained. The model obtained results that were very close to the training on the test data. On the other hand, it was determined that the model showed the highest success in terms of precision and F1-score in the CVM Stage 1 and the highest success in the recall value in the CVM Stage 2.

CONCLUSION

The experimental results have shown that the developed model achieved moderate success and it reached a classification accuracy of 58.66% in CVM stage classification.

Evaluation of automated detection of head position on lateral cephalometric radiographs based on deep learning techniques

BACKGROUND

Lateral cephalometric radiograph (LCR) is crucial to diagnosis and treatment planning of maxillofacial diseases, but inappropriate head position, which reduces the accuracy of cephalometric measurements, can be challenging to detect for clinicians. This non-interventional retrospective study aims to develop two deep learning (DL) systems to efficiently, accurately, and instantly detect the head position on LCRs.

METHODS

LCRs from 13 centers were reviewed and a total of 3000 radiographs were collected and divided into 2400 cases (80.0 %) in the training set and 600 cases (20.0 %) in the validation set. Another 300 cases were selected independently as the test set. All the images were evaluated and landmarked by two board-certified orthodontists as references. The head position of the LCR was classified by the angle between the Frankfort Horizontal (FH) plane and the true horizontal (HOR) plane, and a value within - 3°- 3° was considered normal. The YOLOv3 model based on the traditional fixed-point method and the modified ResNet50 model featuring a non-linear mapping residual network were constructed and evaluated. Heatmap was generated to visualize the performances.

RESULTS

The modified ResNet50 model showed a superior classification accuracy of 96.0 %, higher than 93.5 % of the YOLOv3 model. The sensitivity&recall and specificity of the modified ResNet50 model were 0.959, 0.969, and those of the YOLOv3 model were 0.846, 0.916. The area under the curve (AUC) values of the modified ResNet50 and the YOLOv3 model were 0.985 ± 0.04 and 0.942 ± 0.042, respectively. Saliency maps demonstrated that the modified ResNet50 model considered the alignment of cervical vertebras, not just the periorbital and perinasal areas, as the YOLOv3 model did.

CONCLUSIONS

The modified ResNet50 model outperformed the YOLOv3 model in classifying head position on LCRs and showed promising potential in facilitating making accurate diagnoses and optimal treatment plans.

Cervical Vertebral Maturation Assessment using various Machine Learning techniques on Lateral cephalogram: A systematic literature review

Importance

For the assessment of optimum treatment timing in dentofacial orthopedics, understanding the growth process is of paramount importance. The evaluation of skeletal maturity based on study of the morphology of the cervical vertebrae has been devised to minimize radiation exposure of a patient due to hand wrist radiography. Cervical vertebral maturation assessment (CVMA) predictions have been examined in the state-of-the-art machine learning techniques in the recent past which require more attention and validation by clinicians and practitioners.

Objectives

This paper aimed to answer the question "How are machine learning techniques being employed in studies concerning cervical vertebral maturation assessment using lateral cephalograms?"

Method

A systematic search through the available literature was performed for this work based upon the Population, Intervention, Comparison and Outcome (PICO) framework.

Data sources study selection data extraction and synthesis

The searches were performed in Ovid Medline, Embase, PubMed and Cochrane Central Register of Controlled Trials (CENTRAL) and Cochrane Database of Systematic Reviews (CDSR). A search of the grey literature was also performed in Google Scholar and OpenGrey. We also did a hand-searching in the Angle Orthodontist, Journal of Orthodontics and Craniofacial Research, Progress in Orthodontics, and the American Journal of Orthodontics and Dentofacial Orthopedics. References from the included articles were also searched.

Main outcomes and measures results

A total of 25 papers which were assessed for full text, and 13 papers were included for the systematic review. The machine learning methods used were scrutinized according to their performance and comparison to human observers/experts. The accuracy of the models ranged between 60 and 90% or above, and satisfactory agreement and correlation with the human observers.

Conclusions and relevance

Machine learning models can be used for detection and classification of the cervical vertebrae maturation. In this systematic review (SR), the studies were summarized in terms of ML techniques applied, sample data, age range of sample and conventional method for CVMA, which showed that further studies with a uniform distribution of samples equally in stages of maturation and according to the gender is required for better training of the models in order to generalize the outputs for prolific use to target population.

The psc-CVM assessment system: A three-stage type system for CVM assessment based on deep learning

BACKGROUND

Many scholars have proven cervical vertebral maturation (CVM) method can predict the growth and development and assist in choosing the best time for treatment. However, assessing CVM is a complex process. The experience and seniority of the clinicians have an enormous impact on judgment. This study aims to establish a fully automated, high-accuracy CVM assessment system called the psc-CVM assessment system, based on deep learning, to provide valuable reference information for the growth period determination.

METHODS

This study used 10,200 lateral cephalograms as the data set (7111 in train set, 1544 in validation set and 1545 in test set) to train the system. The psc-CVM assessment system is designed as three parts with different roles, each operating in a specific order. 1) Position Network for locating the position of cervical vertebrae; 2) Shape Recognition Network for recognizing and extracting the shapes of cervical vertebrae; and 3) CVM Assessment Network for assessing CVM according to the shapes of cervical vertebrae. Statistical analysis was conducted to detect the performance of the system and the agreement of CVM assessment between the system and the expert panel. Heat maps were analyzed to understand better what the system had learned. The area of the third (C3), fourth (C4) cervical vertebrae and the lower edge of second (C2) cervical vertebrae were activated when the system was assessing the images.

RESULTS

The system has achieved good performance for CVM assessment with an average AUC (the area under the curve) of 0.94 and total accuracy of 70.42%, as evaluated on the test set. The Cohen's Kappa between the system and the expert panel is 0.645. The weighted Kappa between the system and the expert panel is 0.844. The overall ICC between the psc-CVM assessment system and the expert panel was 0.946. The F1 score rank for the psc-CVM assessment system was: CVS (cervical vertebral maturation stage) 6 > CVS1 > CVS4 > CVS5 > CVS3 > CVS2.

CONCLUSIONS

The results showed that the psc-CVM assessment system achieved high accuracy in CVM assessment. The system in this study was significantly consistent with expert panels in CVM assessment, indicating that the system can be used as an efficient, accurate, and stable diagnostic aid to provide a clinical aid for determining growth and developmental stages by CVM.

Influence of growth structures and fixed appliances on automated cephalometric landmark recognition with a customized convolutional neural network

BACKGROUND

One of the main uses of artificial intelligence in the field of orthodontics is automated cephalometric analysis. Aim of the present study was to evaluate whether developmental stages of a dentition, fixed orthodontic appliances or other dental appliances may affect detection of cephalometric landmarks.

METHODS

For the purposes of this study a Convolutional Neural Network (CNN) for automated detection of cephalometric landmarks was developed. The model was trained on 430 cephalometric radiographs and its performance was then tested on 460 new radiographs. The accuracy of landmark detection in patients with permanent dentition was compared with that in patients with mixed dentition. Furthermore, the influence of fixed orthodontic appliances and orthodontic brackets and/or bands was investigated only in patients with permanent dentition. A t-test was performed to evaluate the mean radial errors (MREs) against the corresponding SDs for each landmark in the two categories, of which the significance was set at p < 0.05.

RESULTS

The study showed significant differences in the recognition accuracy of the Ap-Inferior point and the Is-Superior point between patients with permanent dentition and mixed dentition, and no significant differences in the recognition process between patients without fixed orthodontic appliances and patients with orthodontic brackets and/or bands and other fixed orthodontic appliances.

CONCLUSIONS

The results indicated that growth structures and developmental stages of a dentition had an impact on the performance of the customized CNN model by dental cephalometric landmarks. Fixed orthodontic appliances such as brackets, bands, and other fixed orthodontic appliances, had no significant effect on the performance of the CNN model.

Neural networks for classification of cervical vertebrae maturation: a systematic review

OBJECTIVE

To assess the accuracy of identification and/or classification of the stage of cervical vertebrae maturity on lateral cephalograms by neural networks as compared with the ground truth determined by human observers.

MATERIALS AND METHODS

Search results from four electronic databases (PubMed [MEDLINE], Embase, Scopus, and Web of Science) were screened by two independent reviewers, and potentially relevant articles were chosen for full-text evaluation. Articles that fulfilled the inclusion criteria were selected for data extraction and methodologic assessment by the QUADAS-2 tool.

RESULTS

The search identified 425 articles across the databases, from which 8 were selected for inclusion. Most publications concerned the development of the models with different input features. Performance of the systems was evaluated against the classifications performed by human observers. The accuracy of the models on the test data ranged from 50% to more than 90%. There were concerns in all studies regarding the risk of bias in the index test and the reference standards. Studies that compared models with other algorithms in machine learning showed better results using neural networks.

CONCLUSIONS

Neural networks can detect and classify cervical vertebrae maturation stages on lateral cephalograms. However, further studies need to develop robust models using appropriate reference standards that can be generalized to external data.

iCVM: An Interpretable Deep Learning Model for CVM Assessment under Label Uncertainty

The Cervical Vertebral Maturation (CVM) method aims to determine the craniofacial skeletal maturational stage, which is crucial for orthodontic and orthopedic treatment. In this paper, we explore the potential of deep learning for automatic CVM assessment. In particular, we propose a convolutional neural network named iCVM. Based on the residual network, it is specialized for the challenges unique to the task of CVM assessment. 1) To combat overfitting due to limited data size, multiple dropout layers are utilized. 2) To address the inevitable label ambiguity between adjacent maturational stages, we introduce the concept of label distribution learning in the loss function. Besides, we attempt to analyze the regions important for the prediction of the model by using the Grad-CAM technique. The learned strategy shows surprisingly high consistency with the clinical criteria. This indicates that the decisions made by our model are well interpretable, which is critical in evaluation of growth and development in orthodontics. Moreover, to drive future research in the field, we release a new dataset named CVM-900 along with the paper. It contains the cervical part of 900 lateral cephalograms collected from orthodontic patients of different ages and genders. Experimental results show that the proposed approach achieves superior performance on CVM-900 in terms of various evaluation metrics.

Automatic Classification for Sagittal Craniofacial Patterns Based on Different Convolutional Neural Networks

(1) Background: The present study aims to evaluate and compare the model performances of different convolutional neural networks (CNNs) used for classifying sagittal skeletal patterns. (2) Methods: A total of 2432 lateral cephalometric radiographs were collected. They were labeled as Class I, Class II, and Class III patterns, according to their ANB angles and Wits values. The radiographs were randomly divided into the training, validation, and test sets in the ratio of 70%:15%:15%. Four different CNNs, namely VGG16, GoogLeNet, ResNet152, and DenseNet161, were trained, and their model performances were compared. (3) Results: The accuracy of the four CNNs was ranked as follows: DenseNet161 > ResNet152 > VGG16 > GoogLeNet. DenseNet161 had the highest accuracy, while GoogLeNet possessed the smallest model size and fastest inference speed. The CNNs showed better capabilities for identifying Class III patterns, followed by Classes II and I. Most of the samples that were misclassified by the CNNs were boundary cases. The activation area confirmed the CNNs without overfitting and indicated that artificial intelligence could recognize the compensatory dental features in the anterior region of the jaws and lips. (4) Conclusions: CNNs can quickly and effectively assist orthodontists in the diagnosis of sagittal skeletal classification patterns.

Estimating Cervical Vertebral Maturation with a Lateral Cephalogram Using the Convolutional Neural Network

Recently, the estimation of bone maturation using deep learning has been actively conducted. However, many studies have considered hand–wrist radiographs, while a few studies have focused on estimating cervical vertebral maturation (CVM) using lateral cephalograms. This study proposes the use of deep learning models for estimating CVM from lateral cephalograms. As the second, third, and fourth cervical vertebral regions (denoted as C2, C3, and C4, respectively) are considerably smaller than the whole image, we propose a stepwise segmentation-based model that focuses on the C2–C4 regions. We propose three convolutional neural network-based classification models: a one-step model with only CVM classification, a two-step model with region of interest (ROI) detection and CVM classification, and a three-step model with ROI detection, cervical segmentation, and CVM classification. Our dataset contains 600 lateral cephalogram images, comprising six classes with 100 images each. The three-step segmentation-based model produced the best accuracy (62.5%) compared to the models that were not segmentation-based.