Accuracy of computerized automatic identification of cephalometric landmarks

Comparison of Three Commercially Available, AI-Driven Cephalometric Analysis Tools in Orthodontics

Background: Cephalometric analysis (CA) is an indispensable diagnostic tool in orthodontics for treatment planning and outcome assessment. Manual CA is time-consuming and prone to variability. Methods: This study aims to compare the accuracy and repeatability of CA results among three commercial AI-driven programs: CephX, WebCeph, and AudaxCeph. This study involved a retrospective analysis of lateral cephalograms from a single orthodontic center. Automated CA was performed using the AI programs, focusing on common parameters defined by Downs, Ricketts, and Steiner. Repeatability was tested through 50 randomly reanalyzed cases by each software. Statistical analyses included intraclass correlation coefficients (ICC3) for agreement and the Friedman test for concordance. Results: One hundred twenty-four cephalograms were analyzed. High agreement between the AI systems was noted for most parameters (ICC3 > 0.9). Notable differences were found in the measurements of angle convexity and the occlusal plane, where discrepancies suggested different methodologies among the programs. Some analyses presented high variability in the results, indicating errors. Repeatability analysis revealed perfect agreement within each program. Conclusions: AI-driven cephalometric analysis tools demonstrate a high potential for reliable and efficient orthodontic assessments, with substantial agreement in repeated analyses. Despite this, the observed discrepancies and high variability in part of analyses underscore the need for standardization across AI platforms and the critical evaluation of automated results by clinicians, particularly in parameters with significant treatment implications.

Accuracy of and dental students' preferences toward manual and digital cephalometric landmark identification: A randomized cross-over study

OBJECTIVE

To evaluate dental students' perceptions of manual and digital cephalometric landmark identification methods based on their preferences, difficulty level, and procedure time required to provide insights into the future of dental education, considering incorporating digital technology in dental schools.

MATERIALS AND METHODS

Fifty-five second-year dental students were randomly divided into two groups: (1) group A, students who performed manual landmark identification first, followed by digital method; and (2) group B, students who performed digital method first, followed by manual method. The duration of the procedure was recorded. Subsequently, all students completed a questionnaire regarding the difficulty they experienced using a visual analog scale and their preferences. Landmark identification accuracy was measured.

RESULTS

Digital landmark identification was preferred by 93% of students. The mean procedure time for digital method was significantly lower than that of manual method (13.00 ± 5.60 vs. 9.70 ± 4.60; p = 0.002). Group B completed manual and digital methods in a shorter time than group A. Group A experienced less difficulty with manual procedure than group B. However, statistically significant differences were not observed in the difficulty level of digital technique. A statistically significant difference in the mean accuracy was shown in favor of the manual method. However, this difference is clinically insignificant (p = 0.001).

CONCLUSIONS

Students considered digital method to be effective for learning and preferred it over manual method. Furthermore, digital landmark identification demonstrated better performance and was faster than manual method, suggesting that this must be incorporated in undergraduate dental education.

Landmark annotation through feature combinations: a comparative study on cephalometric images with in-depth analysis of model's explainability

OBJECTIVES

The objectives of this study are to explore and evaluate the automation of anatomical landmark localization in cephalometric images using machine learning techniques, with a focus on feature extraction and combinations, contextual analysis, and model interpretability through Shapley Additive exPlanations (SHAP) values.

METHODS

We conducted extensive experimentation on a private dataset of 300 lateral cephalograms to thoroughly study the annotation results obtained using pixel feature descriptors including raw pixel, gradient magnitude, gradient direction, and histogram-oriented gradient (HOG) values. The study includes evaluation and comparison of these feature descriptions calculated at different contexts namely local, pyramid, and global. The feature descriptor obtained using individual combinations is used to discern between landmark and nonlandmark pixels using classification method. Additionally, this study addresses the opacity of LGBM ensemble tree models across landmarks, introducing SHAP values to enhance interpretability.

RESULTS

The performance of feature combinations was assessed using metrics like mean radial error, standard deviation, success detection rate (SDR) (2 mm), and test time. Remarkably, among all the combinations explored, both the HOG and gradient direction operations demonstrated significant performance across all context combinations. At the contextual level, the global texture outperformed the others, although it came with the trade-off of increased test time. The HOG in the local context emerged as the top performer with an SDR of 75.84% compared to others.

CONCLUSIONS

The presented analysis enhances the understanding of the significance of different features and their combinations in the realm of landmark annotation but also paves the way for further exploration of landmark-specific feature combination methods, facilitated by explainability.

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.

Reproducibility of linear and angular cephalometric measurements obtained by an artificial-intelligence assisted software (WebCeph) in comparison with digital software (AutoCEPH) and manual tracing method

INTRODUCTION

It has been suggested that human errors during manual tracing of linear/angular cephalometric parameters can be eliminated by using computer-aided analysis. The landmarks, however, are located manually and the computer system completes the analysis. With the advent of Artificial Intelligence in the field of Dentistry, automatic location of the landmarks has become a promising tool in digital Orthodontics.

METHODS

Fifty pretreatment lateral cephalograms obtained from the Orthodontic department of SRM dental college (India) were used. Analysis were done by the same investigator using the following methods: WebCeph™, AutoCEPH© for Windows or manual tracing. Landmark identification was carried out automatically by Artificial Intelligence in WebCeph™ and with a mouse driven cursor in AutoCEPH©, and manually using acetate sheet and 0.3-mm pencil, ruler and a protractor. The mean differences of the cephalometric parameters obtained between the three methods were calculated using ANOVA with statistical significance set at p<0.05. Intraclass correlation coefficient (ICC) was used to determine both reproducibility and agreement between linear and angular measurements obtained from the three methods and intrarater reliability of repeated measurements. ICC value of >0.75 indicated good agreement.

RESULTS

Intraclass correlation coefficient between the three groups was >0.830, showing good level of agreement, and the value within each group was >0.950, indicating high intrarater reliability.

CONCLUSION

Artificial Intelligence assisted software showed good agreement with AutoCEPH© and manual tracing for all the cephalometric measurements.

Comparison of AudaxCeph®'s fully automated cephalometric tracing technology to a semi-automated approach by human examiners

OBJECTIVE

To compare the reliability of cephalometric landmark identification by an automated tracing software based on convolutional neural networks to human tracers.

MATERIALS AND METHODS

Sixty cephalograms were traced by two board-certified orthodontists and AudaxCeph®'s artificial intelligence software. The following thirteen landmarks were identified in each tracing: Sella, Nasion, A point, B point, Porion, Menton, Pogonion, Orbitale, Gonion, Upper Central Incisor Incisal Edge (U1 Tip), Upper Central Incisor Root Apex (U1 apex), Lower Central Incisor Incisal Edge (L1 Tip), Lower Central Incisor Root Apex (L1 apex). An x-y axis was positioned in the bottom left corner of each cephalogram, and the x- and y-coordinates for the landmarks were exported into Excel. Distributions of landmarks (X, Y, radial distance) were compared using t-tests of equivalence with a 2mm equivalence bound. These compared the AI position to the two orthodontists - and the orthodontists' reliability by comparing equivalence against each other.

RESULTS

There was no statistical difference between the orthodontists and AudaxCeph®'s automatic tracing software except for the x- and y-dimension of Porion and the y-dimension of L1 apex. The two orthodontists had good intra-examiner reliability with no statistical difference found when comparing them.

CONCLUSION

AudaxCeph®'s automated cephalometric tracing software is a good adjunctive tool to use when diagnosing and treatment planning orthodontic cases.

Recent Advanced Diagnostic Aids in Orthodontics

Everyone aspires to have a youthful appearance, complete with a beaming grin. By treating skeletal and dental malocclusions that interfere with facial aesthetics, orthodontics helps patients attain a beautiful face and a smile they will be proud of. The diagnosis of the underlying illness or condition serves as the basis for all medical, dental, and surgical operations. Dental professionals all across the world have reaped the benefits of a major scientific advance in recent years. Many measurements may now be seen and quantified more easily using computer-assisted cephalometry. The accuracy and high quality of all dental materials have been enhanced thanks to computer-aided design and computer-aided manufacture. There have been several developments in the realm of orthodontics. The growing use of technology in recent years has transformed every sector, including medicine and dentistry. From Angle's time to today's nanorobotic age, orthodontic ideas, biomaterials, and technology have evolved greatly. It has been extensively utilized for diagnosis, treatment planning, three-dimensional printing, appliance systems, digital storage, integration, and data retrieval. The technology used in orthodontics is always getting better, and this article aims to give an overview of the most recent changes.

Vertical Skeletal Changes after Extraction and Nonextraction Orthodontic Treatment

OBJECTIVES

 Due to the constant battle regarding the controversial topic of orthodontic extraction, this study aims to assess the changes in vertical dimensions of patients treated with premolar extractions compared with nonextraction orthodontic patients.

MATERIALS AND METHODS

 A sample of 60 borderline patients were recruited and divided into extraction and nonextraction groups. Eleven pretreatment cephalometric measurements were recorded using WebCeph and patients were followed-up until the completion of treatment.

STATISTICAL ANALYSIS

 Intragroup and intergroup comparisons were made using paired t-test and two-sample independent t-test, respectively. The joint significance of differences was measured using F-tests.

RESULTS

 The intragroup comparison revealed that in the extraction group, the vertical dimension was significantly increased posttreatment for four cephalometric measurements, that is, mandibular plane angle (p < 0.05), palatal plane angle (p < 0.05), Frankfort mandibular plane angle (p < 0.05), and y-axis (p < 0.05). In the comparison of the posttreatment values of both groups, the mean differences of the posttreatment values for sella nasion (SN)-gonion (Go)-gnathion (Gn) angle (p = 0.008), the total anterior (p = 0.050), and lower anterior facial heights (AFH; p = 0.011) were significantly higher. At the same time, the Jarabak ratio was significantly (p = 0.006) lower in the extraction group than in the nonextraction group.

CONCLUSION

 The increase in vertical dimension is significantly higher in the extraction group than in the nonextraction group which indicates a significant impact of orthodontic extraction on the vertical dimensions.

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.

Potential and impact of artificial intelligence algorithms in dento-maxillofacial radiology

OBJECTIVES

Novel artificial intelligence (AI) learning algorithms in dento-maxillofacial radiology (DMFR) are continuously being developed and improved using advanced convolutional neural networks. This review provides an overview of the potential and impact of AI algorithms in DMFR.

MATERIALS AND METHODS

A narrative review was conducted on the literature on AI algorithms in DMFR.

RESULTS

In the field of DMFR, AI algorithms were mainly proposed for (1) automated detection of dental caries, periapical pathologies, root fracture, periodontal/peri-implant bone loss, and maxillofacial cysts/tumors; (2) classification of mandibular third molars, skeletal malocclusion, and dental implant systems; (3) localization of cephalometric landmarks; and (4) improvement of image quality. Data insufficiency, overfitting, and the lack of interpretability are the main issues in the development and use of image-based AI algorithms. Several strategies have been suggested to address these issues, such as data augmentation, transfer learning, semi-supervised training, few-shot learning, and gradient-weighted class activation mapping.

CONCLUSIONS

Further integration of relevant AI algorithms into one fully automatic end-to-end intelligent system for possible multi-disciplinary applications is very likely to be a field of increased interest in the future.

CLINICAL RELEVANCE

This review provides dental practitioners and researchers with a comprehensive understanding of the current development, performance, issues, and prospects of image-based AI algorithms in DMFR.

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 landmark annotation in 3D surface scans of skulls: Methodological proposal and reliability study

BACKGROUND AND OBJECTIVES

Craniometric landmarks are essential in many biomedical applications, such as morphometric analysis or forensic identification. The process of locating landmarks is usually a manual and slow task, highly influenced by fatigue, skills and the experience of the practitioner. Localization errors are propagated and magnified in subsequent steps, which can result in incorrect measurements or assumptions. Thereby, standardization, reliability and reproducibility lay the foundations for the necessary accuracy in subsequent measurements or anatomical analysis. In this paper, we present an automatic method to annotate 3D surface skull models taking into account anatomical and geometrical features.

METHODS

The proposed method follows a hybrid structure where a deformable template is used to initialize the landmark positions. Then, a refinement stage is applied using prior anatomical knowledge to ensure a correct placement. Our proposal is validated over thirty 3D skull scans of male Caucasians, acquired by hand-held surface scanning, and a set of 58 craniometric landmarks. A statistical analysis was carried out to analyze the inter- and intra-observer variability of manual annotations and the automatic results, along with a visual assessment of the final results.

RESULTS

Inter-observer errors show significant differences, which are reflected in the expert consensus used as reference. The average localization error was 2.19±1.5 mm when comparing the automatic landmarks to the reference location. The subsequent visual analysis confirmed the reliability of the refinement method for most landmarks.

CONCLUSIONS

Repeated manual annotations show a high variability depending on both skills and expertise of the observer, and landmarks' location and characteristics. In contrast, the automatic method provides an accurate, robust and reproducible alternative to the tedious and error-prone task of manual landmarking.

The accuracy and reliability of WebCeph for cephalometric analysis

Objective

This study compares the accuracy and reliability of WebCeph (web-based program for cephalometric analysis) with the AutoCAD computer software.

Materials and methods

A sample of pretreatment digital lateral cephalograms of 50 orthodontic patients was analysed with WebCeph and AutoCAD software (as a standard measure). On each cephalogram, 17 landmarks and 11 measurements were marked and performed as skeletal, dental, and soft–tissue parameters. We used six angular and five linear measurements. A paired t-test was used to assess the systematic bias. The intraclass correlation coefficient (ICC) and Bland–Altman plot with linear regression analysis were used to assess the agreement between the two methods.

Results

There was adequate reproducibility for the measurements with both WebCeph and AutoCAD. The paired t-test showed statistically significant differences for five angular and two linear measurements (P < 0.05). The ICC test between WebCeph and AutoCAD revealed very good to excellent agreement for all measurements, except for the lower incisor to mandibular plane angle. The Bland–Altman plot visually showed a relatively acceptable limit of agreement for three angular and two linear measurements only, and the linear regression analysis revealed a significant proportional bias between the two methods for four angles and the upper lip-Esthetic line (U Lip-E Line). The systematic bias and level of agreement improved with the use of the semi-automatic WebCeph.

Conclusions

Different problems, such as poor landmark identification/soft tissue tracing and inconsistency of measurements, are inherent to the automatic WebCeph. The semi-automatic WebCeph can overcome some limitations of the automatic WebCeph; however, it should be used for cephalometric analysis with a great deal of caution.

Three dimensional (3D) imaging techniques in orthodontics-An update

3D imaging is a technique which develops or creates the impression of depth within an image by deploying 2D data into 3-dimensional format. To aid in quality regulating processes for industrial purposes, 3D imaging has become an extremely valuable factor. Owing to their various drawbacks, a wide range of investigative methods formulated for demonstration of facial structures and the dentition were dilapidated. Currently in medicine, the most prevalent method is perhaps 3D imaging technique renders thorough and problem specific information regarding hard and the soft tissues, such as Computerized Tomography (CT), Cone Beam Computerized Tomography (CBCT), Micro Computerized Tomography (MCT), 3D laser scanning, structured light technique, stereophotogrammetry or 3D surface imaging systems (3dMD), 3D facial morphometry (3DFM), Tuned Aperture Computed Tomography (TACT), and Magnetic Resonance Imaging (MRI). 3D imaging techniques in orthodontics plays an important role by facilitating more elaborated diagnostic information on the precise cases like patients having craniofacial anomalies. Hence, the aim of this study was to review advances in 3D imaging with in the field of orthodontics.

Artificial intelligence in orthodontics : Evaluation of a fully automated cephalometric analysis using a customized convolutional neural network

PURPOSE

The aim of this investigation was to create an automated cephalometric X‑ray analysis using a specialized artificial intelligence (AI) algorithm. We compared the accuracy of this analysis to the current gold standard (analyses performed by human experts) to evaluate precision and clinical application of such an approach in orthodontic routine.

METHODS

For training of the network, 12 experienced examiners identified 18 landmarks on a total of 1792 cephalometric X‑rays. To evaluate quality of the predictions of the AI, both AI and each examiner analyzed 12 commonly used orthodontic parameters on a basis of 50 cephalometric X‑rays that were not part of the training data for the AI. Median values of the 12 examiners for each parameter were defined as humans' gold standard and compared to the AI's predictions.

RESULTS

There were almost no statistically significant differences between humans' gold standard and the AI's predictions. Differences between the two analyses do not seem to be clinically relevant.

CONCLUSIONS

We created an AI algorithm able to analyze unknown cephalometric X‑rays at almost the same quality level as experienced human examiners (current gold standard). This study is one of the first to successfully enable implementation of AI into dentistry, in particular orthodontics, satisfying medical requirements.

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.

A new methodology for automatic detection of reference points in 3D cephalometry: A pilot study

OBJECTIVE

The aim of this study was to develop a new method for an automatic detection of reference points in 3D cephalometry to overcome the limits of 2D cephalometric analyses.

MATERIALS AND METHODS

A specific application was designed using the C++ language for automatic and manual identification of 21 (reference) points on the craniofacial structures. Our algorithm is based on the implementation of an anatomical and geometrical network adapted to the craniofacial structure. This network was constructed based on the anatomical knowledge of the 3D cephalometric (reference) points.

RESULTS

The proposed algorithm was tested on five CBCT images. The proposed approach for the automatic 3D cephalometric identification was able to detect 21 points with a mean error of 2.32mm.

DISCUSSION

In this pilot study, we propose an automated methodology for the identification of the 3D cephalometric (reference) points. A larger sample will be implemented in the future to assess the method validity and reliability.

Reliability Assessment of Orthodontic Apps for Cephalometrics

Objective

The aim of this study was to evaluate the accuracy and reliability of cephalometric measurements using iPad apps called CephNinja and SmartCeph Pro. The measurements were compared with the measurements obtained using Dolphin Imaging computer software.

Methods

Twenty digital cephalometric radiographs were randomly selected from the archives and traced using the CephNinja app, SmartCeph Pro app, and Dolphin Imaging software. Twenty-one landmarks and 16 measurements were performed in each program. The statistical analysis was conducted using the Bland-Altman analysis at a significance level of 0.05.

Results

For the CephNinja app, there were seven measurements that were in accordance with Dolphin Imaging software. For the SmartCeph Pro app, six measurements were in accordance with Dolphin software. Both apps gave better results for angular measurements than linear ones.

Conclusion

These findings indicate that, although they are user-friendly, orthodontic apps for cephalometrics are not equal with Dolphin software now and need to be developed to be more reliable for most of the measurements.

Reproducibility of measurements in tablet-assisted, PC-aided, and manual cephalometric analysis

OBJECTIVE

To assess the reproducibility of cephalometric measurements performed with software for a tablet, with a program for personal computers (PCs), and manually.

MATERIALS AND METHODS

The pretreatment lateral cephalograms of 20 patients that were acquired using the same digital cephalometer were collected. Tracings were performed with NemoCeph for Windows (Nemotec), with SmileCeph for iPad (Glace Software), and by hand. Landmark identification was carried out with a mouse-driven cursor using NemoCeph and with a stylus pen on the iPad screen using SmileCeph. Hand tracings were performed on printouts of the cephalograms, using a 0.3-mm 2H pencil and a protractor. Cephalometric landmarks and linear and angular measurements were recorded. All the tracings were done by the same investigator. To evaluate reproducibility, for each cephalometric measurement the agreement between the value derived from NemoCeph, that given by SmileCeph and that measured manually was assessed with the intraclass correlation coefficient (ICC). Agreement was rated as low for an ICC≤0.75, and an ICC>0.75 was considered indicative of good agreement. Also, differences in measurements between each software and manual tracing were statistically evaluated (P<.05).

RESULTS

All the measurements had ICC>0.8, indicative of a high agreement among the tracing methods. Relatively lower ICCs occurred for linear measurements related to the occlusal plane and to N perpendicular to the Frankfurt plane. Differences in measurements between both software programs and hand tracing were not statistically significant for any of the cephalometric parameters.

CONCLUSION

Tablet-assisted, PC-aided, and manual cephalometric tracings showed good agreement.

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.

Accuracy and repeatability of computer aided cervical vertebra landmarking in cephalogram

The accuracy and repeatability of computer aided cervical vertebra landmarking (CACVL) were investigated in cephalogram. 120 adolescents (60 boys, 60 girls) aged from 9.1 to 17.2 years old were randomly selected. Twenty-seven landmarks from the second to fifth cervical vertebrae on the lateral cephalogram were identified. In this study, the system of CACVL was developed and used to identify and calculate the landmarks by fast marching method and parabolic curve fitting. The accuracy and repeatability in CACVL group were compared with those in two manual landmarking groups [orthodontic experts (OE) group and orthodontic novices (ON) group]. The results showed that, as for the accuracy, there was no significant difference between CACVL group and OE group no matter in x-axis or y-axis (P>0.05), but there was significant difference between CACVL group and ON group, as well as OE group and ON group in both axes (P<0.05). As for the repeatability, CACVL group was more reliable than OE group and ON group in both axes. It is concluded that CACVL has the same or higher accuracy, better repeatability and less workload than manual landmarking methods. It's reliable for cervical parameters identification on the lateral cephalogram and cervical vertebral maturation prediction in orthodontic practice and research.

Reliability of four different computerized cephalometric analysis programs

The aim of this investigation was to compare the traditional method of manual cephalometric tracing with four different computerized tracing programs, where the lateral cephalograms were scanned at 300 dpi and digitized onscreen. Thirty randomly selected cephalometric radiographs were used in this study. Four programs Dolphin Imaging, Vistadent, Nemoceph, and Quick Ceph were evaluated. Three dental, 11 skeletal, and 1 soft tissue parameters were measured that consisted of 5 linear and 10 angular measurements. Statistical analysis was carried out using multivariate analysis of variance and Box's and Levene's tests. No statistically significant difference was found between manual tracing and the computerized tracing programs. The measurements obtained with the cephalometric analysis programs used in the study were reliable.

Influence of a programme of professional calibration in the variability of landmark identification using cone beam computed tomography-synthesized and conventional radiographic cephalograms

OBJECTIVES

The validity of any measurement obtained through a cephalogram largely depends on the reproducibility of the cephalometric landmarks. The purpose of this study is to evaluate the influence of a programme of professional calibration (PPC) on the variability of landmark identification comparing conventional radiographs and cone beam CT (CBCT)-synthesized cephalograms.

METHODS

5 graduate students in oral radiology identified 20 cephalometric landmarks from cephalograms generated from conventional radiographs (RADs), Ray-Sum CBCT-synthesized cephalograms (CBTs) and half-skull CBT (HSTs) from 10 patients. After a period of reinforcement on instruction and calibration with inter- and intraexaminer assessment of reproducibility (intraclass coefficient correlation scores > 0.75) for RADs, CBTs and HSTs obtained from 5 different patients, observers were asked to repeat the analysis of the first 10 patients under the same circumstances. Values in millimetres represented each landmark in a table of Cartesian co-ordinates (x- and y-axes).

RESULTS

ANOVA showed significant reduction in variability levels after the PPC, and there were no differences among the methods of image acquisition. Repeated measures ANOVA indicated that the PPC accounted for reduction in variability levels in 14 of 20 landmarks.

CONCLUSIONS

The results suggest that a PPC has more influence than the type of image acquisition on variability of landmark identification based on two-dimensional cephalometric analysis. Cephalograms obtained from RAD or CBCT can be considered equivalent for clinical and experimental applications.

A comparison of hand-tracing and cephalometric analysis computer programs with and without advanced features--accuracy and time demands

The aim of this study was to evaluate the basic and advanced features of five different cephalometric analysis computer programs. The level of measurement agreement with hand-tracing and time demands was examined. The material consisted of 30 digital lateral radiographic images. Twenty-three measurements were calculated by one operator both manually and using five different cephalometric analysis software programs. Intraclass correlation coefficient (ICC) was used to detect differences in measurement agreement between hand-tracing and basic features as well as between hand-tracing and advanced features. Coefficient of variation (CV) was used to assess intra-user error and a Student's t-test to determine time differences. Of the 23 measurements tested for each procedure, one [(Ii to NB (mm)] showed better agreement with hand-tracing when the advanced features were used, 20 showed good agreement with hand-tracing for both basic and advanced features, while two (AB on FOP and Ii to A/Pog) showed poor intra-user reproducibility. Hand-tracing took a significantly longer time (P < 0.001) than both the basic and advanced features. The advanced features took a significantly longer time (P < 0.001) than the basic features. Both basic and advanced features showed good measurement agreement with the hand-tracing technique. The use of the basic features minimizes the time requirements for analysis. A computerized tracing technique, which consists of either basic or advanced feature, can be regarded as less time consuming and equally reliable to hand-tracing as far as cephalometric measurements are concerned.

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.

Reliability of traditional cephalometric landmarks as seen in three-dimensional analysis in maxillary expansion treatments

OBJECTIVE

To evaluate intra-examiner and inter-examiner reliability of 3D CBCT-generated landmarks previously used in traditional 2D cephalometry.

MATERIALS AND METHODS

Twenty-four CBCTs NewTom 3G (Aperio Services, Verona, Italy) were randomly selected from patients participating in a clinical trial involving maxillary expansion treatments. The principal investigator located the landmarks five times, and four other investigators located the same landmarks once. Intra-examiner and inter-examiner reliability values were determined using intraclass correlation coefficients (ICCs). To assist in interpretation of the clinical significance of landmark identification differences, average mean differences for x, y, and z landmark coordinates were determined from the repeated assessments. Landmarks then were separated into groups with respect to the region they represented and then were compared via repeated measures ANOVA and multiple comparisons via Bonferroni corrected alpha.

RESULTS

Intra-examiner and inter-examiner reliability for x, y, and z coordinates for all landmarks were acceptable, all being greater than 0.80. Most of the mean measurement differences obtained from trials within the principal investigator in all three axes were less than 1.5 mm. Inter-examiner mean measurement differences generally were larger than the intra-examiner differences.

CONCLUSIONS

Based on this, the best landmarks for use in verifying expansion treatment results are Ekm, buccal surface, and apexes of upper molars, upper premolars and upper canines, and buccal surfaces of lower molars and lower canines. Foramen Spinosum, ELSA, Auditory External Meatus, and Dorsum Foramen Magnum demonstrated adequate reliability for determining a standardized reference system.

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.

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.

Evaluation of speed, repeatability, and reproducibility of digital radiography with manual versus computer-assisted cephalometric analyses

The aims of this study were to evaluate intra-examiner repeatability and inter-examiner reproducibility of landmarks using two cephalometric analysing techniques, manual and computerized, and to compare these for speed. One hundred lateral cephalometric radiographs were randomly selected and 11 angular and six linear parameters were traced and measured by two examiners using the manual method and Dolphin Image Software 9.0 on each radiograph. A Student's t-test for paired and independent samples was used to compare the mean values of intra- and inter-examiner differences. Intraclass correlation coefficients (ICC) were calculated to determine intra- and inter-examiner correlation (r value). Both operators were generally consistent in the repeated measurements; however, for one examiner, the differences for Na is perpendicular to A (P < 0.001), Na is perpendicular Pog, and U1-NA (P < 0.01) distance measurements were found to be statistically significant. Intra-examiner repeatability of landmarks both with the manual and Dolphin techniques showed high correlation coefficients. While inter-examiner reproducibility of landmarks was unacceptable, measurement errors with the manual technique were generally comparable with the Dolphin technique. The mean tracing times of the two operators for a single tracing was 2 minutes 41 seconds for Dolphin and 6 minutes 51 seconds for manual tracings. Computer-assisted cephalometric analysis does not increase intra- and inter-examiner reliability but can result in time saving.

Accounting for measurement error: a critical but often overlooked process

AIMS

Due to instrument imprecision and human inconsistencies, measurements are not free of error. Technical error of measurement (TEM) is the variability encountered between dimensions when the same specimens are measured at multiple sessions. A goal of a data collection regimen is to minimise TEM. The few studies that actually quantify TEM, regardless of discipline, report that it is substantial and can affect results and inferences. This paper reviews some statistical approaches for identifying and controlling TEM. Statistically, TEM is part of the residual ('unexplained') variance in a statistical test, so accounting for TEM, which requires repeated measurements, enhances the chances of finding a statistically significant difference if one exists.

METHODS

The aim of this paper was to review and discuss common statistical designs relating to types of error and statistical approaches to error accountability. This paper addresses issues of landmark location, validity, technical and systematic error, analysis of variance, scaled measures and correlation coefficients in order to guide the reader towards correct identification of true experimental differences.

CONCLUSIONS

Researchers commonly infer characteristics about populations from comparatively restricted study samples. Most inferences are statistical and, aside from concerns about adequate accounting for known sources of variation with the research design, an important source of variability is measurement error. Variability in locating landmarks that define variables is obvious in odontometrics, cephalometrics and anthropometry, but the same concerns about measurement accuracy and precision extend to all disciplines. With increasing accessibility to computer-assisted methods of data collection, the ease of incorporating repeated measures into statistical designs has improved. Accounting for this technical source of variation increases the chance of finding biologically true differences when they exist.

Comparison of hand-traced and computerized cephalograms: landmark identification, measurement, and superimposition accuracy

INTRODUCTION

The purposes of this study were (1) to investigate the variations of landmark identification between film and digital cephalometric tracings, (2) to compare the ability of Quick Ceph 2000 (Quick Ceph Systems, Inc, San Diego, Calif) to measure the linear and angular measurements with the hand-traced method, and (3) to compare Quick Ceph 2000 superimpositions to the hand-traced method of superimpositions that are currently accepted by the American Board of Orthodontics (ABO).

MATERIALS

We used 30 sets of serial cephalometric radiographs of growing patients from 1 orthodontic office. Fiduciary x- and y-axes were drawn in pencil on the T1 radiographs in the regions of the cranial base, the maxilla, and the mandible. The fiduciary lines were transferred to the digital and film serial cephalograms by regionally superimposing the tracings as described in the ABO Phase III examination handbook. A Mann-Whitney test was done to compare the median and Delta of the T1 and T2 values for each measurement acquired by hand and by Quick Ceph.

RESULTS AND CONCLUSIONS

There was no difference in the identification of cephalometric landmarks made manually vs digitally with Quick Ceph 2000. There was no difference in acquiring consistent cephalometric values for the measurements required by the ABO for the Phase III clinical examination manually vs digitally by using Quick Ceph 2000. There was no difference in the regional superimpositions of the mandible, the maxilla, and the cranial base, manually vs digitally with Quick Ceph 2000.

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.

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.

Landmark identification on direct digital versus film-based cephalometric radiographs: a human skull study

The purpose of this study was to investigate differences in landmark identification on vertically scanned, direct digital and conventional (18 x 24 cm) cephalometric radiographs. Eight observers, all orthodontists or postgraduate orthodontic students, recorded 6 landmarks twice on 3 digital and 3 conventional cephalograms obtained from 3 human skulls in a standardized fashion. Digital images were displayed on a 15.1-in TFT monitor in 3:1 mode (20 x 26 cm). Recordings were transferred into standardized coordinate systems and evaluated separately for each coordinate. After correcting for magnification, precision was assessed with Maloney-Rastogi tests, and intraobserver and interobserver reproducibility was calculated from squared differences. Effective magnification was larger for the digital images (x, 13%; y, 12%). Significantly different (P <.05) precision was found for nasion (N), posterior nasal spine (PNS), sella (S), supraspinale (A), and orbitale (Or), but average differences were entirely below 1 mm. Interobserver and intraobserver reproducibility did not differ significantly between the 2 image modes. Squared differences were largest for PNS and Or in both modalities. Our results indicate comparable errors in landmark recording for both evaluated machines. However, these results must be considered in the context of the specific display conditions for digital images, because no general standard exists for this purpose.