Diagnosis and treatment planning of orthodontic patients with 3-dimensional dentofacial records

Lateral cephalometric parameters among Arab skeletal classes II and III patients and applying machine learning models

BACKGROUND

The World Health Organization considers malocclusion one of the most essential oral health problems. This disease influences various aspects of patients' health and well-being. Therefore, making it easier and more accurate to understand and diagnose patients with skeletal malocclusions is necessary.

OBJECTIVES

The main aim of this research was the establishment of machine learning models to correctly classify individual Arab patients, being citizens of Israel, as skeletal class II or III. Secondary outcomes of the study included comparing cephalometric parameters between patients with skeletal class II and III and between age and gender-specific subgroups, an analysis of the correlation of various cephalometric variables, and principal component analysis in skeletal class diagnosis.

METHODS

This quantitative, observational study is based on data from the Orthodontic Center, Jatt, Israel. The experimental data consisted of the coded records of 502 Arab patients diagnosed as Class II or III according to the Calculated_ANB. This parameter was defined as the difference between the measured ANB angle and the individualized ANB of Panagiotidis and Witt. In this observational study, we focused on the primary aim, i.e., the establishment of machine learning models for the correct classification of skeletal class II and III in a group of Arab orthodontic patients. For this purpose, various ML models and input data was tested after identifying the most relevant parameters by conducting a principal component analysis. As secondary outcomes this study compared the cephalometric parameters and analyzed their correlations between skeletal class II and III as well as between gender and age specific subgroups.

RESULTS

Comparison of the two groups demonstrated significant differences between skeletal class II and class III patients. This was shown for the parameters NL-NSL angle, PFH/AFH ratio, SNA angle, SNB angle, SN-Ba angle. SN-Pg angle, and ML-NSL angle in skeletal class III patients, and for S-N (mm) in skeletal class II patients. In skeletal class II and skeletal class III patients, the results showed that the Calculated_ANB correlated well with many other cephalometric parameters. With the help of the Principal Component Analysis (PCA), it was possible to explain about 71% of the variation between the first two PCs. Finally, applying the stepwise forward Machine Learning models, it could be demonstrated that the model works only with the parameters Wits appraisal and SNB angle was able to predict the allocation of patients to either skeletal class II or III with an accuracy of 0.95, compared to a value of 0.99 when all parameters were used ("general model").

CONCLUSION

There is a significant relationship between many cephalometric parameters within the different groups of gender and age. This study highlights the high accuracy and power of Wits appraisal and the SNB angle in evaluating the classification of orthodontic malocclusion.

The accuracy of photographic soft-tissue profile analysis to determine Class II and vertical skeletal relationships in children

BACKGROUND

Lateral cephalometric analysis (LCA) is the reference standard for identifying common skeletal relationships in orthodontics, such as Cl II and hyperdivergent skeletal discrepancies, but it entails radiation exposure. Therefore, photographic soft-tissue profile analysis (PSPA) could be a useful alternative for these diagnoses, particularly for paediatric patients. This study aims to estimate the accuracy of PSPA for determining common skeletal discrepancies in children.

METHODS

Cephalometric radiographs and profile photographs of a consecutive series of 100 children (8.0-17.6 years old) made on the same day were included. The validity of PSPA was verified against comparable LCA. First, by assessing the Pearson correlation and then estimating the sensitivity, specificity, receiver operating characteristic (ROC) curves and area under the curve (AUC) in sample A (n = 50). After external validation in a new sample B (n = 50), the ROC-AUC, diagnostic odds ratio, best cut-off points and discriminative validity were assessed in the total sample. Interrater reliability was estimated using the intraclass correlation coefficient, the standard error of measurement and Bland-Altman plots.

RESULTS

The measurement properties of the PSPA angles A'N'B', Gl'-Sn-Pog, N'-Sn-Pog', and N'-Tra-Me' were valid (ROC-AUC > 0.7) and reliable (ICCs > 0.92). The angles A'N'B', Gl'-Sn-Pog', and N'-Sn-Pog', with the respective cut-off points ≥7.7', ≥12.8', and ≤163.5', were accurate values for determining Cl II discrepancy. The N'-Tra-Me'-angle (≥63') was an accurate estimate for a hyperdivergent discrepancy.

CONCLUSIONS

These validated PSPA angles could be used in clinical settings as a minimally invasive diagnostic tool to screen children suspected of having skeletal Cl II and hyperdivergent discrepancies.

3D Technology Used for Precision in Orthodontics

One of the most crucial technologies used by orthodontists to assess and document the dimensions of craniofacial features is imaging. Orthodontists frequently employ two-dimensional (2D) imaging methods, although 2D imaging cannot localize or determine the depth of structures. Early in the 1990s, three-dimensional (3D) imaging was invented, and it has since become a crucial part of dentistry, especially in orthodontics. One of the newest and most important breakthroughs in dentistry is 3D technology. Clinicians have been able to significantly improve patient care while also shortening the time spent on treatment planning due to these technologies, which include intra-oral scanning, 3D imaging, computed-axial tomography (CAT) scan, cone-beam computed tomography (CBCT), computer-aided design/computer-aided manufacturing (CAD/CAM), and 3D software. 3D models of maxillary and mandibular arches can take the place of conventional plaster casts and their limits for planning treatments, appliance production, and estimated treatment results as part of this continuous progress. Digital orthodontics procedures have become more popular in the recent past. The development of "personalized" orthodontic appliances makes use of technology. These technologies' overall improvement can increase clinicians' productivity and efficiency by simplifying traditional methods that are seen to be particularly laborious. The objectives of this review are to provide an overall description of the 3D technology nowadays and to assess its orthodontic applications.

Application of Three-Dimensional Digital Technology in Orthodontics: The State of the Art

Three-dimensional technologies are one of the most recent and relevant advancements in the field of Dentistry. These systems, including intraoral scans, 3D imaging exams (CAT scan, CBCT and MRI), CAD/CAM 3D printing devices and 3D computer software, have enabled clinicians to greatly improve patient care along with reducing treatment planning time. The present descriptive study aims to explore possible applications of 3D technologies during the diagnosis, treatment plan, case monitoring and result assessment in orthodontics. The overall upgrade provided by these technologies can improve the clinicians’ workflow and effectiveness by simplifying conventional techniques considered to be especially arduous.

Assessment of automatic cephalometric landmark identification using artificial intelligence

OBJECTIVE

To compare the accuracy of cephalometric landmark identification between artificial intelligence (AI) deep learning convolutional neural networks (CNN) You Only Look Once, Version 3 (YOLOv3) algorithm and the manually traced (MT) group.

SETTING AND SAMPLE POPULATION

The American Association of Orthodontists Federation (AAOF) Legacy Denver collection was used to obtain 110 cephalometric images for this study.

MATERIALS AND METHODS

Lateral cephalograms were digitized and traced by a calibrated senior orthodontic resident using Dolphin Imaging. The same images were uploaded to AI software Ceppro DDH Inc The Cartesian system of coordinates with Sella as the reference landmark was used to extract x- and y-coordinates for 16 cephalometric points: Nasion (Na), A point, B point, Menton (Me), Gonion (Go), Upper incisor tip, Lower incisor tip, Upper incisor apex, Lower incisor apex, Anterior Nasal Spine (ANS), Posterior Nasal Spine (PNS), Pogonion (Pg), Pterigomaxillary fissure point (Pt), Basion (Ba), Articulare (Art) and Orbitale (Or). The mean distances were assessed relative to the reference value of 2 mm. Student paired t-tests at significance level of P < .05 were used to compare the mean differences in each of the x- and y-components. SPSS (IBM-vs. 27.0) software was used for the data analysis.

RESULTS

There was no statistical difference for 12 out of 16 points when analysing absolute differences between MT and AI groups.

CONCLUSION

AI may increase efficiency without compromising accuracy with cephalometric tracings in routine clinical practice and in research settings.

Orthodontic radiology: development of a clinical practice guideline

OBJECTIVES

Radiographs are considered essential in orthodontics. However, their diagnostic value and indications for use are still uncertain, while exposure to radiation carries health risks. This study aimed to report on the development of a clinical practice guideline on orthodontic radiology.

METHODS

A Guideline Development Taskforce was set up. The GRADE methodology was used for development and the RIGHT Statement for reporting of the guideline. We systematically reviewed articles to address the main clinical question: how different types of radiographs contribute to orthodontic diagnosis, treatment planning and post-treatment outcome evaluation. After a literature search and data extraction, we formulated conclusions and assessed the strength of the evidence according to the GRADE method. Both literature conclusions and the most important considerations, such as patient preferences, organizational matters and expert opinions were taken into account to finally issue recommendations.

RESULTS

7 clinical questions focused on orthopantomograms, lateral cephalograms, hand-wrist radiographs, peri-apical radiographs, bitewings, antero-occlusal radiographs, and cone-beam computer tomographic imaging. The literature search lead to 484 unique studies, of which 17 were included in the analysis. The strength of evidence of the conclusions was graded low or very low. We formulated considerations and took them into account when issuing the 13 clinical recommendations to address the clinical questions.

CONCLUSIONS

There was a considerable lack of scientific evidence on this topic. Nonetheless, this guideline provides clinicians with a tool for decision-making regarding radiographic records while enhancing patient radiation protection. More research of higher quality is recommended for a future update.

Can modifying shielding, field of view, and exposure settings make the effective dose of a cone-beam computed tomography comparable to traditional radiographs used for orthodontic diagnosis?

OBJECTIVES

To analyze the effect of changes in exposure settings, field of view (FOV), and shielding on radiation to an adult and child phantom from cone-beam computed tomography (CBCT) imaging compared to panoramic and lateral cephalometric radiographs.

MATERIALS AND METHODS

The effective dose to an adult and child anthropomorphic phantom by the CS 9300 using various scan protocols was recorded. Absorbed radiation was measured with optically stimulated luminescence dosimeters and effective dose calculated using 2007 International Commission on Radiological Protection tissue weighting factors. Scan protocols included different FOVs, voxel sizes, and standard versus low-dose protocols. Radiation shielding was used when it did not interfere with FOV. Panoramic and lateral cephalometric radiographs were taken with the Orthophos SL.

RESULTS

Even with shielding, smaller FOVs, and increased voxel sizes, the effective dose of standard CBCT scans was higher than panoramic and lateral cephalometric radiographs. A shielded limited FOV standard scan combined with a lateral cephalometric radiograph resulted in a lower dose (P < .001) than a full FOV standard scan. Low-dose shielded scans resulted in significant dose reductions to the adult (P < .05) and child (P < .001) phantoms compared to the respective panoramic and lateral cephalometric radiographs combined. Image quality analysis was not possible with radiation equivalent phantoms.

CONCLUSIONS

Unlike standard CBCTs, shielded low-dose CBCT protocols in the CS 9300 have lower effective doses than conventional radiographs for adult and child phantoms. If high resolution and cranial base visualization are necessary, combining a shielded LFOV standard exposure with a cephalometric radiograph is recommended.

Surface-based 3-dimensional cephalometry: An objective analysis of cranio-mandibular morphology

INTRODUCTION

The purpose of this study was to apply a previously reported homologous model for quantifying and visualizing the three-dimensional (3D) cranio-mandibular morphology.

METHODS

Twenty-eight 3D cone-beam computed tomography (CBCT) images of Japanese patients (7 to 13 years of age) showing skeletal Class 1 malocclusions were used in this study. Wire mesh fitting was conducted for each patient based on the assignment of landmarks to each 3D surface (homologous modeling). Cranial and mandibular images were generated using surface rendering of the CBCT images. The mean and standard deviation of each point on the wire mesh were calculated as the normative mean. Two operators conducted the process twice for seven patients, with a one-week interval between fittings. The intra- and inter-examiner reliability were determined using the 95% confidence interval minimal detectable change (MDC₉₅) for random error and intra-class correlation coefficients (ICCs).

RESULTS

The intra- and inter-examiner reliability of the mesh fitting method were almost perfect. Normative ranges of the cranial and mandibular surfaces of Japanese were determined.

CONCLUSIONS

A clinical method that enables practitioners to quantify and visualize the hard tissues of a patient's face in three dimensions with almost perfect reliability is presented. This method allows practitioners to evaluate how patients' skeletal characteristics differ from normative means in three dimensions.

Integration of digital maxillary dental casts with 3D facial images in orthodontic patients

OBJECTIVE

To evaluate three-dimensional (3D) accuracy and reliability of nonradiographic dentofacial images integrated with a two-step method.

METHODS

3D facial images, cone-beam computed tomography (CBCT) images and digital maxillary dental casts were obtained from 20 pre-orthodontic subjects. Digital dental casts were integrated into 3D facial images using a two-step method based on the anterior tooth area. 3D coordinate values of five dental landmarks were identified in both dentofacial images and CBCT images. The accuracy of the integration method was assessed with paired t-tests between dentofacial images and CBCT-based reference standards. Intraclass correlation coefficients (ICCs) were assessed for the reliability of dentofacial images and CBCT-based images. Analysis of variance and Kruskal-Wallis tests evaluated the accuracy of the method in different dimensions.

RESULTS

There was no statistical difference between dentofacial images and CBCT reference standards in both translational and rotational dimensions (P > .05). Translational mean absolute errors for full dentitions were within 0.42 mm and ICCs were over 0.998 in x, y, and z directions. Rotational mean absolute errors for full dentitions were within 0.92° and ICCs over 0.734 in pitch, yaw, and roll orientations. Integration errors were significantly greater in the first molar, z-translation, and pitch rotation (P < .05).

CONCLUSIONS

Integrating 3D dentofacial images with the two-step method is precise and acceptable for clinical diagnostics and scientific purposes. Errors were greater in the molar region, z-translation, and pitch rotation.

Deep Convolutional Neural Networks Based Analysis of Cephalometric Radiographs for Differential Diagnosis of Orthognathic Surgery Indications

The aim of this study was to evaluate the deep convolutional neural networks (DCNNs) based on analysis of cephalometric radiographs for the differential diagnosis of the indications of orthognathic surgery. Among the DCNNs, Modified-Alexnet, MobileNet, and Resnet50 were used, and the accuracy of the models was evaluated by performing 4-fold cross validation. Additionally, gradient-weighted class activation mapping (Grad-CAM) was used to perform visualized interpretation to determine which region affected the DCNNs’ class classification. The prediction accuracy of the models was 96.4% for Modified-Alexnet, 95.4% for MobileNet, and 95.6% for Resnet50. According to the Grad-CAM analysis, the most influential regions for the DCNNs’ class classification were the maxillary and mandibular teeth, mandible, and mandibular symphysis. This study suggests that DCNNs-based analysis of cephalometric radiograph images can be successfully applied for differential diagnosis of the indications of orthognathic surgery.

The relationship between 3D dentofacial photogrammetry measurements and traditional cephalometric measurements

OBJECTIVES

To determine the relationship between traditional cephalometric measurements and corresponding nonradiographic three-dimensional (3D) photogrammetry measurements.

MATERIALS AND METHODS

This was a cross-sectional study of 20 orthodontic patients (10 male and 10 female) who received lateral cephalometric radiographs and 3D dentofacial photogrammetric records with each subject serving as his or her own control for a total sample size of 40 images (20 per method). A 3D analysis that resembled a traditional cephalometric analysis was established using the eyes and natural head orientation as substitutes for the cranial base. Pearson correlation coefficients and multivariable linear regression plots were calculated to evaluate the relationship between the photogrammetry measurements and the cephalometric measurements.

RESULTS

The ANB angle, mandibular plane angle, lower anterior face height, upper incisor angle to SN, upper incisor angle to NA, and all measurements of lower incisor position and inclination had strong positive Pearson correlation coefficients with the corresponding 3D photogrammetry measurements ( P < .004). Statistically significant regression plots demonstrated that cephalometric relationships between the jaws and incisor orientation can be predicted from corresponding 3D photogrammetry measurements.

CONCLUSIONS

3D photogrammetry measurements relating the jaws to each other and incisor orientation has a strong positive correlation with corresponding traditional cephalometric measurements and can serve as cephalometric predictors. Capturing the eyes using 3D photogrammetry can obviate the need to expose the cranial base and allow limiting the radiographic field to the area of interest.

Two-dimensional versus three-dimensional Frӓnkel Manoeuvre: a reproducibility study

Background/Objectives

To assess intra- and inter-observer reproducibility of the evaluation of the Frӓnkel Manoeuvre (FM) on three-dimensional (3D) scans and to compare it to the assessment on two-dimensional (2D) pictures.

Materials/Methods

This study included 44 individuals with a skeletal Class II malocclusion [9–16-years old mean age ± standard deviation: 13.5 ± 2.01 years]. All patients had a full Class II molar relationship, overjet ≥6 mm and point A, Nasion, point B angle (ANB) ≥5 degrees. During the FM, each patient was invited to bite in centric occlusion and then to posture the mandible forward to reach a Class I molar relationship. The FM was recorded both by a normal camera and a 3D scanner (3dMD system). Six examiners divided into two groups according to their orthodontic clinical experience (&lt;5 and &gt;10 years), and one gold standard in the evaluation of FM, commented twice (every 15 days) on both 2D photographs (T0 and T2) and 3D scans (T1 and T3). The intra-observer agreement and the inter-observer agreement compared to the gold standard were evaluated by computing the Cohen’s K.

Results

The agreement between observations for each examiner ranged from 0.36 to 1 on 2D pictures (T0 versus T2), and from 0.22 to 0.69 on 3D scans (T1 versus T3). The overall agreement was 0.63 (95 per cent CI = 0.35–0.91) in 2D analysis and 0.5 (95 per cent CI = 0.35–0.64) in 3D analysis.

Conclusions/Implications

The FM was less reproducible when performed on 3D records than on 2D pictures.

Quantifying faces three-dimensionally in orthodontic practice

PURPOSE

The purpose of this study was to formulate and demonstrate a method for quantifying and visualizing the three-dimensional (3-D) configuration of the soft tissues of the face at rest to facilitate a quantitative and instantaneous understanding of a patient's static facial form characteristics.

MATERIALS AND METHODS

3-D facial images of 200 Japanese adult volunteers at rest were recorded using a 3-D photogrammetric system. For each participant, a wire mesh fitting was conducted based on the assignment of landmarks to each 3-D facial image. This method generated 6,017 points on the wire mesh (i.e., the nodes of the fitted mesh). For each point, the mean and standard deviation were calculated and used for patient evaluation. The system performance was exemplified with three orthodontic patients with skeletal Class II and III malocclusions and laterotrusion.

RESULTS

Three patients with severe skeletal Class II and III deformities and laterotrusion were evaluated with the proposed method.

CONCLUSION

A clinical method for practitioners to quantify and visualize the soft tissues of a patient's face in 3 dimensions has been presented with clinical applications and considerations. This method allows practitioners to evaluate how patients' facial characteristics differ from normative faces.

A modified Steiner's analysis that does not require radiographic exposure of the cranial base

OBJECTIVE

To correlate traditional Steiner's skeletal and dental measurements to similar measurements that use the eyes and natural head position as references instead of the cranial base.

SETTING AND SAMPLE

One hundred and fifty-two lateral cephalometric radiographs (66 female and 86 male) from the Harvard Forsyth twin sample were included in the study based on record availability.

MATERIAL AND METHODS

This was an observational study in which all cephalometric radiographs were traced and analysed using Steiner's cephalometric analysis. Each cephalogram was then altered to perform a similar analysis that uses the maximum convexity of the cornea together with natural head position as references instead of the cranial base. A Pearson product-moment correlation coefficient was measured to determine the correlation between the conventional Steiner's analysis measurements and the novel measurements relying on the eyes and natural head position.

RESULTS

Steiner's cephalometric measurements of the jaws to each other, the divergency and the orientation of the incisors had a strong positive correlation with their counterparts that relied on true horizontal and/or the eyes (P < 0.001).

CONCLUSION

The eyes and true horizontal can be used as alternatives to the cranial base when diagnosing the relationship between the jaws and the position and orientation of the teeth. Since the eyes and natural head position are identifiable without ionizing radiation, future research should focus on the use of radiographic exposures limited to the upper and lower jaws for orthodontic diagnosis and outcome assessment.

Accuracy and reliability of digital craniofacial measurements using a small-format, handheld 3D camera

OBJECTIVES

Craniofacial assessments often involve three-dimensional facial imaging using an expensive camera with 6 SLR lenses to analyse the positions and relations of anatomic landmarks. Recently, a 3D small-format, handheld camera was developed; however, the accuracy and reliability of this system are largely unknown. The purpose of this study was to evaluate the accuracy and reliability of this system.

MATERIALS & METHODS

A total of 30 sets of evaluations were completed by 2 examiners on 5 human subjects, using 3 different methods: direct callipers, 3D handheld camera and conventional tripod 3D camera images. Each evaluation included 29 anthropometric landmarks that were used as reference points for facial analysis. Two examiners marked the landmarks directly on the faces and measured linear distances using the 3 measurement methods.

RESULTS

Accuracy analysis was performed for handheld vs direct calliper vs conventional camera measurements. Each of these analyses yielded a grand mean of correlation coefficients of .98. Bias measurements revealed that the handheld and conventional camera methods yielded larger measurements than direct callipers (with a mean difference of 1.74, 1.56 mm, respectively, for rater 1 and 0.94, 1.02 mm, respectively, for rater 2). When compared to one another, both the handheld camera and the conventional camera methods yielded similar values for most measurements, with the average overall difference between these modalities of 0.03 mm for rater 1 and 0.07 mm for rater 2.

CONCLUSIONS

The 3D handheld camera showed high accuracy and reliability in comparison with traditional models, indicating that this system may provide a useful tool in craniofacial anthropometry.