3D Cephalometric Landmarks Research Articles

A semi-supervised learning approach for automated 3D cephalometric landmark identification using computed tomography.

Identification of 3D cephalometric landmarks that serve as proxy to the shape of human skull is the fundamental step in cephalometric analysis. Since manual landmarking from 3D computed tomography (CT) images is a cumbersome task even for the trained experts, automatic 3D landmark detection system is in a great need. Recently, automatic landmarking of 2D cephalograms using deep learning (DL) has achieved great success, but 3D landmarking for more than 80 landmarks has not yet reached a satisfactory level, because of the factors hindering machine learning such as the high dimensionality of the input data and limited amount of training data due to the ethical restrictions on the use of medical data. This paper presents a semi-supervised DL method for 3D landmarking that takes advantage of anonymized landmark dataset with paired CT data being removed. The proposed method first detects a small number of easy-to-find reference landmarks, then uses them to provide a rough estimation of the all landmarks by utilizing the low dimensional representation learned by variational autoencoder (VAE). The anonymized landmark dataset is used for training the VAE. Finally, coarse-to-fine detection is applied to the small bounding box provided by rough estimation, using separate strategies suitable for the mandible and the cranium. For mandibular landmarks, patch-based 3D CNN is applied to the segmented image of the mandible (separated from the maxilla), in order to capture 3D morphological features of mandible associated with the landmarks. We detect 6 landmarks around the condyle all at once rather than one by one, because they are closely related to each other. For cranial landmarks, we again use the VAE-based latent representation for more accurate annotation. In our experiment, the proposed method achieved a mean detection error of 2.88 mm for 90 landmarks using only 15 paired training data.

Automatic 3-Dimensional Cephalometric Landmarking via Deep Learning

The increasing use of 3-dimensional (3D) imaging by orthodontists and maxillofacial surgeons to assess complex dentofacial deformities and plan orthognathic surgeries implies a critical need for 3D cephalometric analysis. Although promising methods were suggested to localize 3D landmarks automatically, concerns about robustness and generalizability restrain their clinical use. Consequently, highly trained operators remain needed to perform manual landmarking. In this retrospective diagnostic study, we aimed to train and evaluate a deep learning (DL) pipeline based on SpatialConfiguration-Net for automatic localization of 3D cephalometric landmarks on computed tomography (CT) scans. A retrospective sample of consecutive presurgical CT scans was randomly distributed between a training/validation set (n = 160) and a test set (n = 38). The reference data consisted of 33 landmarks, manually localized once by 1 operator(n = 178) or twice by 3 operators (n = 20, test set only). After inference on the test set, 1 CT scan showed “very low” confidence level predictions; we excluded it from the overall analysis but still assessed and discussed the corresponding results. The model performance was evaluated by comparing the predictions with the reference data; the outcome set included localization accuracy, cephalometric measurements, and comparison to manual landmarking reproducibility. On the hold-out test set, the mean localization error was 1.0 ± 1.3 mm, while success detection rates for 2.0, 2.5, and 3.0 mm were 90.4%, 93.6%, and 95.4%, respectively. Mean errors were −0.3 ± 1.3° and −0.1 ± 0.7 mm for angular and linear measurements, respectively. When compared to manual reproducibility, the measurements were within the Bland–Altman 95% limits of agreement for 91.9% and 71.8% of skeletal and dentoalveolar variables, respectively. To conclude, while our DL method still requires improvement, it provided highly accurate 3D landmark localization on a challenging test set, with a reliability for skeletal evaluation on par with what clinicians obtain.

Reliability of cephalometric landmark identification on three-dimensional computed tomographic images

Our aims were to evaluate the reliability of three-dimensional (3D) cephalometric landmark identification in 3D images, and to propose an improved protocol for determining these landmarks. Computed tomographic (CT) images of 13 landmarks were obtained. One that did not show any artifacts, asymmetry in maxillofacial structures, or bony defects, was selected. Two orthodontic practitioners identified 3D cephalometric landmarks 10 times at one-week intervals. The distances of 26 landmarks were measured on the basis of three reference planes (coronal, horizontal, and sagittal). Ten mean (SD) measurements from each examiner were calculated, and the maximum and minimum values and the difference from the 10 measurements of each one were measured at a 95% confidence interval. Interexaminer differences for the three planes were found in the upper right first molar, point A, both gonions, left orbitale, and both porions. The lower right first molar, foramen magnum, gnathion, nasion, and pogonion showed interexaminer differences in two planes. Menton, basion, posterior nasal spine, upper and lower left first molar, and right mental foramen showed interexaminer differences in only one plane. With reference to intraexaminer differences, poor repeatability was observed for gonion, orbitale, condylion, and porion. Reliable 3D landmarks are the meeting point of sutures, distinct structures at converging planes, landmarks positioned in the midline, distinct anatomical structures such as the mental foramen, and teeth using multiplanar views.

S12-12 SESSION 12: FACIOCRANIOSYNOSTOSIS – PART II MONOBLOC MINUS LEFORT II: THE ROLE OF DIFFERENTIAL DISTRACTION FOR TREATMENT OF THE APERTS CRANIOFACIAL PHENOTYPE

Introduction: Apert patients often require differential advancement of the forehead, orbits, and midface to correct fronto-orbital retrusion, lengthen the midface and nose, and close the anterior open bite. While traditionally done in stages, the Monobloc + Le Fort II distraction addresses these differential movements in a single operation. Advancement of the forehead and midface is accomplished with the monobloc, while the nose and central midface is inferiorly advanced and rotated using the Le Fort II distraction. The purpose of this study is to evaluate the skeletal 3D cephalometric changes utilizing this technique in the management of Aperts syndrome. Methods: A retrospective review was performed of the senior surgeons experience utilizing this technique in patients with Aperts syndrome. Pre and post distraction 3D CT cephalometrics were analyzed and compared to a historical published cohort of Aperts patients. Vertical and axial facial ratios were calculated before and after distraction, and these ratios were also compared between cohorts. Results: 3 patients with Aperts syndrome underwent a Monobloc + lefort II differential distraction at an average age of 8 years, and with an average length of follow up of 30 months. The following 3D cephalometric landmarks and angles were measured: Glabella, nasion, incisor midline, ANS, sella, lateral inferior orbit, SNA, SNB, occlusal plane, and maxillary mandibular incisor relationship (see chart). Average Vertical Facial Movements ANS to N: 16.3mm N to Lateral inferior orbit (LIO): 3.81mm; Average Axial Facial Movements and Ratios ANS to Sella: 31.1mm Sella to LIO: 9.05mm Sella to LIO/S to ANS: 0.290 Comparison to historical control in the literature: Lefort II distraction with zygomatic repositioning. Axial facial movement ratio: Sella to LIO/S to ANS Monobloc/Lefort II mean: 0.292 Lefort II/Zygoma Repositioning: 0.276 P = 0.43 (NS) Vertical Facial Movements Ratio: N to LIO/N to ANS Monobloc/Lefort II: 0.241 Lefort II/Zygoma Repositioning: 0.203 Conclusion: Differential distraction with the Monobloc + Lefort II allows for resolution of exorbitism, improvement in negative canthal tilt, lengthening of the nose, and closure of the anterior open bite in patients with Apert syndrome. The increased distraction and rotation of the central midface results in more normal facial ratios that are not observed with traditional monobloc distraction.

In vivo reliability of 3D cephalometric landmark determination on magnetic resonance imaging: a feasibility study.

3D cephalometric analysis performed on cone-beam or multi-slice computed tomography (CBCT, MSCT) has superior diagnostic value compared to 2D cephalometry based on radiographs. However, this comes at the expense of increased radiation risks for the predominantly young patients. Magnetic resonance imaging (MRI) has the potential to overcome this diagnostic dilemma but has not been established for 3D cephalometry so far. Since landmark reliability forms the basis for 3D cephalometry, we evaluated the in vivo reliability of established 3D landmarks using MRI. Sixteen orthodontic patients underwent MRI at 3Tesla using a 0.5 mm 3D sequence. On each MRI scan, 44 cephalometric landmarks were determined. Image analysis was performed twice by two independent observers. Intra- and inter-rater agreement was assessed by mean measurement errors and intraclass correlation coefficients (ICCs). Measurement errors were calculated as Euclidean distances and distances for x-, y-, and z-coordinates. Overall, MRI-based 3D cephalometric landmarks revealed a high reliability comparable to prior in vivo studies using CBCT/MSCT. Intra- and inter-rater ICCs were consistently higher than 0.9. Intra-rater comparisons showed mean measurement differences (ranges) of 0.87mm (0.41-1.63) for rater I and 0.94mm (0.49-1.28) for rater II. Average inter-rater difference was 1.10mm (0.52-2.29). Distinct differences in reliability between the various landmarks were observed, corresponding well with the landmarks' specific shapes. The present study demonstrates that MRI enables reliable determination of 3D cephalometric landmarks in vivo. This study emphasizes the potential of MRI to perform treatment planning and monitoring in orthodontics as well as oral and maxillofacial surgery without radiation exposure.

Reliability of different three-dimensional cephalometric landmarks in cone-beam computed tomography: A systematic review

Conventional two-dimensional (2D) cephalometric radiography is an integral part of orthodontic patient diagnosis and treatment planning. One must be mindful of its limitations as it indeed is a 2D representation of a vaster three-dimensional (3D) object. Issues with projection errors, landmark identification, and measurement inaccuracies impose significant limitations, which may now be overcome with the advent of cone-beam computed tomography (CBCT). A systematic review of the reliability of different 3D cephalometric landmarks in CBCT imaging was conducted. Electronic database searches were administered until October 2017 using PubMed, MEDLINE via OvidSP, EBMR and EMBASE via OvidSP, Scopus, and Web of Science. Google Scholar was used as an adjunctive search tool. Thirteen articles considering CBCT scans of human subjects from preexisting data sets were selected and reviewed. Most of the studies had methodological limitations and were of moderate quality. Because of their heterogeneity, key data from each could not be combined and were reported qualitatively. Overall, in 3D, midsagittal plane landmarks demonstrated greater reliability compared with bilateral landmarks. A minimum number of dental landmarks were reported, although most were recommended for use. Further research is required to evaluate the reliability of 3D cephalometric landmarks when evaluating 3D craniofacial complexes.

The facial effects of tooth wear rehabilitation as measured by 3D stereophotogrammetry

ObjectiveEvaluate the effect of a full rehabilitation, including an increase of vertical dimension of occlusion (VDO), in patients with severe tooth wear on changes in facial appearance. MethodsPatients with severe tooth wear were restoratively treated using direct composite restorations, including an increase of the VDO. Before and one month after treatment 3D stereophotographs were taken and facial changes were measured with the use of 3D cephalometric landmarks. Cast models were used to determine the Tooth Wear Index (TWI) and the VDO-increase. The Orofacial Esthetic Scale-questionnaire was used to evaluate subjective happiness with appearance. Paired Student’s t-tests were used to evaluate the changes after full rehabilitation. The relation between the facial changes, VDO-increment and TWI-score were analyzed using multiple regression models (p < 0.05). Results44 Patients (35 men, 9 women, mean age: 38.2 ± 7.1 y) participated in this study. Mean TWI-score at baseline was 3.3 ± 0.5 and the mean increase in VDO was 1.6 ± 0.9 mm. Lower face height increased by 1.9 mm (p ≤ 0.001). Patients reported a significant improvement in orofacial appearance: OES-summary score of 34.3 ± 11.2 before and 59.9 ± 6.8 one month after treatment (p < 0.001). The increase of VDO resulted in an increase in Lower Face Height (p ≤ 0.002), whereas the TWI-score at baseline had no effect on the facial changes. The applied increase in VDO and TWI-score were not related to changes in OES-scores. ConclusionsA full rehabilitation with an increase in VDO resulted in objective differences in facial height as measured with 3D stereophotogrammetry. Clinical significanceIn patients with severe tooth wear, receiving a full rehabilitation including an increase in vertical dimension of occlusion, an objectively change will occur in the Lower facial height, but this effect will be subjectively judged as a positive change.

Automatic localization of three-dimensional cephalometric landmarks on CBCT images by extracting symmetry features of the skull

To propose an algorithm for automatic localization of 3D cephalometric landmarks on CBCT data, those are useful for both cephalometric and upper airway volumetric analysis. 20 landmarks were targeted for automatic detection, of which 12 landmarks exist on the mid-sagittal plane. Automatic detection of mid-sagittal plane from the volume is a challenging task. Mid-sagittal plane is detected by extraction of statistical parameters of the symmetrical features of the skull. The mid-sagittal plane is partitioned into four quadrants based on the boundary definitions extracted from the human anatomy. Template matching algorithm is applied on the mid-sagittal plane to identify the region of interest ROI, further the edge features are extracted, to form contours in the individual regions. The landmarks are automatically localized by using the extracted knowledge of anatomical definitions of the landmarks. The overall mean error for detection of 20 landmarks was 1.88 mm with a standard deviation of 1.10 mm.The cephalometric landmarks on CBCT data were detected automaticallywith in the mean error less than 2mm.

Enhanced Surgical Outcomes in Patients With Skeletal Class III Facial Asymmetry by 3-Dimensional Surgical Simulation

With the advance of image fusion techniques, the creation of 3-dimensional (3D) virtual head and 3D surgical simulations has provided previews of surgical procedures. The aim of this study was to investigate the surgical outcomes in patients receiving orthognathic surgery (OGS) with the guidance of 3D computer-assisted surgical simulation. The study included 34 consecutive patients (15 men and 19 women; age, 18.1 to 33.0yr) with skeletal Class III facial asymmetry who underwent bimaxillary OGS. One-week postoperative cone-beam computed tomographic craniofacial images (Ta) were constructed and superimposed on preoperative simulated virtual images (Ts) at the cranial base and surfaces of the frontal and periorbital regions. The 3D cephalometric landmarks were measured relative to 3 reference planes. The outcomes among different experience levels of surgeons also were compared. Although the mean values between Ta and Ts were small, statistical differences were observed in the center of maxillary and mandibular incisors and the B point relative to the midline and in the maxillary first molar in vertical distances, sagittal dentoskeletal dimensions, and pitch angles. The root mean square deviations (RMSDL) of measurement variables relative to center landmark accuracy were 1.5 and less than 2mm at the maxilla and mandible, respectively. RMSDL greater than 2mm was located at the maxillary first molar in the vertical distance and in the sagittal dimension at the anterior nasal spine and B point. Variables related to centering the midline structures were highly inter-related. The roll angle deviation was associated with centering the midline landmarks. The yaw angle deviation was not associated with midline correction. With guidance from 3D surgical simulation, surgeons could achieve similar outcomes to correct facial asymmetry regardless of their years of practice.

Three-dimensional volumetric changes in the upper airway after maxillomandibular advancement in obstructive sleep apnoea patients and the impact on quality of life

Obstructive sleep apnoea (OSA) is a sleep-related breathing disorder, characterized by repetitive airway obstructions, causing disruptive snoring and daytime sleepiness. Maxillomandibular advancement (MMA), which enlarges the upper airway, is a therapeutic surgical approach. However, no study has performed an upper airway sub-region analysis using validated three-dimensional (3D) anatomical and technical limits on cone beam computed tomography (CBCT). Hence, this prospective, observational trial was performed to evaluate 3D volumetric changes in the upper airway according to validated 3D cephalometric landmarks, before and after MMA, for all patients with a polysomnography diagnosis of OSA (apnoea–hypopnoea index (AHI) ≥5). The secondary objective was to evaluate the impact of MMA on the AHI and in a subjective manner with the Epworth Sleepiness Scale (ESS) and OSA questionnaire. Eleven consecutive OSA patients were included. A significant volume increase in the oropharynx (P=0.002) and hypopharynx (P=0.02) was observed, in contrast to a non-significant volume reduction in the nasopharynx (P >0.05). The median AHI (P=0.03) and ESS score (P=0.004) decreased significantly as a result of surgery. In conclusion, MMA significantly enlarges the airway volume of the oropharynx and hypopharynx and is associated with improved quality of life.

Experimental and clinical assessment of three-dimensional cephalometry: A systematic review

ObjectivesThis study provides a systematic review of the current scientific literature on three-dimensional (3D) cephalometry. The null hypothesis was that 3D cephalometry is an accurate and reproducible diagnostic technique. To examine this hypothesis, the following three research questions were proposed: 1) What is the accuracy of 3D cephalometric measurements compared to in vitro measurements? 2) What is the intra- and inter-observer reliability of the selection of 3D cephalometric landmarks? 3) What is the reproducibility of the linear and angular measurements? MethodsA comprehensive database search was performed, using Medline, the Cochrane Central Register of Controlled Trials, Web of Science and Google Scholar. The titles and abstracts obtained from the search were screened and evaluated by two observers according to the inclusion and exclusion criteria. ResultsThe evaluation process yielded 21 articles. A high level of agreement (<1 mm) between the in vitro measurements and those obtained from 3D cephalometry was observed and some landmarks provided highly reproducible results. However, the linear (0.04–7.49 mm) and angular (0.99–9.30°) measurements differed greatly. ConclusionsThe null hypothesis was rejected. This study indicates critical points regarding 3D cephalometry and provides guidance for future research in this field.

Vertical facial height and its correlation with facial width and depth

IntroductionThe aim of the present study was to evaluate how vertical facial height correlates with mandibular plane angle, facial width and depth from a three dimensional (3D) viewing angle.MethodsIn this study 3D cephalometric landmarks were identified and measurements from 43 randomly selected cone beam computed tomography (CBCT) images of dry skulls from the Weisbach collection of Vienna Natural History Museum were analyzed. Pearson correlation coefficients of facial height measurements and mandibular plane angle and the correlation coefficients of height-width and height-depth were calculated, respectively.ResultsThe mandibular plane angle (MP-SN) significantly correlated with ramus height (Co-Go) and posterior facial height (PFH) but not with anterior lower face height (ALFH) or anterior total face height (ATFH). The ALFH and ATFH showed significant correlation with anterior cranial base length (S-N), whereas PFH showed significant correlation with the mandible (S-B) and maxilla (S-A) anteroposterior position.ConclusionsHigh or low mandibular plane angle might not necessarily be accompanied by long or short anterior face height, respectively. The PFH rather than AFH is assumed to play a key role in the vertical facial type whereas AFH seems to undergo relatively intrinsic growth.

Reproducibility of 3D cephalometric landmarks on cone-beam and low-dose computed tomography

Reproducibility of 3D cephalometric landmarks on cone-beam and low-dose computed tomography