Adaptable cascaded registration for personalized maxilla completion and cleft defect volume estimation.

BACKGROUND: Cone-beam computed tomography (CBCT) images provide high-resolution insights into the underlying craniofacial anomaly in patients with cleft lip and palate (CLP), requiring non-negligible annotation costs to measure the cleft defect for the guidance of the clinical secondary alveolar bone graft procedures. Considering the cumbersome volumetric image acquisition, there is a lack of paired CLP CBCTs and normal CBCTs for learning-based anatomical structure restoration models. Nowadays, the registration-based method relieves the annotation burden, though one-shot registration and the regular mask are limited to handling fine-grained shape variations and harmony between restored bony tissues and the defected maxilla. PURPOSE: This study aimed to design and evaluate a novel method for deformable partial registration of the CLP CBCTs and normal CBCTs, enabling personalized maxilla completion and cleft defect volume prediction from CLP CBCTs. METHODS: We proposed an adaptable deep registration framework for personalized maxilla completion and cleft defect volume prediction from CLP CBCTs. The key ingredient was a cascaded partial registration to exploit the maxillary morphology prior and attribute transfer. Cascaded registration with coarse-to-fine registration fields handled morphological variations of cleft defects and fine-grained maxillary restoration. We designed an adaptable cleft defect mask and volumetric Boolean operators for reliable voxel filling of the defected maxilla. A total of 36 clinically obtained CLP CBCTs were used to train and validate the proposed model, among which 22 CLP CBCTs were used to generate a training dataset with 440 synthetic CBCTs by B-spline deformation-based data augmentation and the remaining for testing. The proposed model was evaluated on maxilla completion and cleft defect volume prediction from clinically obtained unilateral and bilateral CLP CBCTs. RESULTS: Extensive experiments demonstrated the effectiveness of the adaptable cleft defect mask and the cascaded partial registration on maxilla completion and cleft defect volume prediction. The proposed method achieved state-of-the-art performances with the Dice similarity coefficient of 0.90 ± $\pm$ 0.02 on the restored maxilla and 0.84 ± $\pm$ 0.04 on the estimated cleft defect, respectively. The average Hausdorff distance between the estimated cleft defect and the manually annotated ground truth was 0.30 ± $\pm$ 0.08 mm. The relative volume error of the cleft defect was 0.09 ± $0.09\pm$ 0.08. The proposed model allowed for the prediction of cleft defect maps that were in line with the ground truth in the challenging unilateral and bilateral CLP CBCTs. CONCLUSIONS: The results suggest that the proposed adaptable deep registration model enables patient-specific maxilla completion and automatic annotation of cleft defects, relieving tedious voxel-wise annotation and image acquisition burdens.

Natural reference structures for three-dimensional maxillary regional superimposition in growing patients.

BACKGROUND: Assessment of growth-related or treatment-related changes in the maxilla requires a reliable method of superimposition. Such methods are well established for two-dimensional (2D) cephalometric images but not yet for three-dimensions (3D). The aims of this study were to identify natural reference structures (NRS) for the maxilla in growing patients in 3D, opportunistically using orthodontic mini-screws as reference; and to test the applicability of the proposed NRS for maxillary superimposition by assessing the concordance of this approach with Björk's 'stable reference structures' in lateral projection. METHODS: The stability of the mini-screws was tested on longitudinal pairs of pre- and post-orthodontic cone-beam computed tomography (CBCT) images by measuring the distance changes between screws. After verifying the stability of the mini-screws, rigid registration was performed for aligning the stable mini-screws. Then, non-rigid registration was used to establish the dense voxel-correspondence among CBCT images and calculate the displacement of each voxel belonging to the maxilla relative to the mini-screws. The displacement vectors were transformed to a standardized maxillary template to categorize the stability of the internal structures statistically. Those voxels that displaced less relative to the mini-screws were considered as the natural reference structures (NRS) for the maxilla. Test samples included another dataset of longitudinal CBCT scans. They were used to evaluate the applicability of the proposed NRS for maxillary superimposition. We assessed whether aligning the maxilla with proposed NRS is in concordance with the maxillary internal reference structures superimposition in the traditional 2D lateral view as suggested by Björk. This was quantitively assessed by comparing the mean sagittal and vertical tooth movements for both superimposition methods. RESULTS: The stability of the mini-screws was tested on 10 pairs of pre- and post-orthodontic cone-beam computed tomography (CBCT) images (T1: 12.9 ± 0.8 yrs, T2: 14.8 ± 0.7 yrs). Both the loaded and the unloaded mini-screws were shown to be stable during orthodontic treatment, which indicates that they can be used as reference points. By analyzing the deformation map of the maxilla, we confirmed that the infraorbital rims, maxilla around the piriform foramen, the infrazygomatic crest and the hard palate (palatal vault more than  1 cm distal to incisor foramen except the palatal suture) were stable during growth. Another dataset of longitudinal CBCT scans (T1: 12.2 ± 0.63 yrs, T2: 15.2 ± 0.96 yrs) was used to assess the concordance of this approach with Björk's 'stable reference structures'. The movement of the maxillary first molar and central incisor showed no statistically significant difference when superimposing the test images with the proposed NRS or with the classic Björk maxillary superimposition in the lateral view. CONCLUSIONS: The infraorbital rims, maxilla around the piriform foramen, the infrazygomatic crest and the hard palate (palatal vault more than 1 cm posterior to incisal foramen except the palatal suture) were identified as stable regions in the maxilla. These stable structures can be used for maxillary superimposition in 3D and generate comparable results to Björk superimposition in the lateral view.

Bi-Graph Reasoning for Masticatory Muscle Segmentation From Cone-Beam Computed Tomography.

Automated segmentation of masticatory muscles is a challenging task considering ambiguous soft tissue attachments and image artifacts of low-radiation cone-beam computed tomography (CBCT) images. In this paper, we propose a bi-graph reasoning model (BGR) for the simultaneous detection and segmentation of multi-category masticatory muscles from CBCTs. The BGR exploits the local and long-range interdependencies of regions of interest and category-specific prior knowledge of masticatory muscles by reasoning on the category graph and the region graph. The category graph of the learnable muscle prior knowledge handles high-level dependencies of muscle categories, enhancing the feature representation with noise-agnostic category knowledge. The region graph models both local and global dependencies of the candidate muscle regions of interest. The proposed BGR accommodates the high-level dependencies and enhances the region features in the presence of entangled soft tissue and image artifacts. We evaluated the proposed approach by segmenting masticatory muscles on clinically acquired CBCTs. Extensive experimental results show that the BGR effectively segments masticatory muscles with state-of-the-art accuracy.

Emergence of a new designated clade 16 with significant antigenic drift in hemagglutinin gene of H9N2 subtype avian influenza virus in eastern China.

H9N2 avian influenza viruses (AIVs) pose an increasing threat to the poultry industry worldwide and have pandemic potential. Vaccination has been principal prevention strategy to control H9N2 in China since 1998, but vaccine effectiveness is persistently challenged by the emergence of the genetic and/or antigenic variants. Here, we analysed the genetic and antigenic characteristics of H9N2 viruses in China, including 70 HA sequences of H9N2 isolates from poultry, 7358 from online databases during 2010-2020, and 15 from the early reference strains. Bayesian analyses based on hemagglutinin (HA) gene revealed that a new designated clade16 emerged in April 2012, and was prevalent and co-circulated with clade 15 since 2013 in China. Clade 16 viruses exhibited decreased cross-reactivity with those from clade 15. Antigenic Cartography analyses showed represent strains were classified into three antigenic groups named as Group1, Group2 and Group3, and most of the strains in Group 3 (15/17, 88.2%) were from Clade 16 while most of the strains in Group2 (26/29, 89.7%) were from Clade 15. The mean distance between Group 3 and Group 2 was 4.079 (95%CI 3.605-4.554), revealing that major switches to antigenic properties were observed over the emergence of clade 16. Genetic analysis indicated that 11 coevolving amino acid substitutions primarily at antigenic sites were associated with the antigenic differences between clade 15 and clade 16. These data highlight complexities of the genetic evolution and provide a framework for the genetic basis and antigenic characterization of emerging clade 16 of H9N2 subtype avian influenza virus.

Dense correspondence of deformable volumetric images via deep spectral embedding and descriptor learning.

Deformable image correspondence plays an essential role in a variety of medical image analysis tasks. Most existing deep learning-based registration and correspondence techniques exploit metric space alignments in the spatial domain and learn a nonlinear voxel-wise mapping function between volumetric images and displacement fields, agnostic to intrinsic structure correspondence. When confronted with high-frequency perturbations of patients' poses and anatomical structural variations, they relied on prior rigid and affine transformations, as well as additional segmentation masks and landmark annotations for reliable registration. This paper presents a data-driven spectral mapping-based correspondence framework to handle the intrinsic correspondence of anatomical structures. At the core of our approach lies a deep convolutional framework that approximates spectral bases and optimizes volumetric descriptors. The multi-path graph convolutional network-based spectral embedding approximation module relieves the computationally expensive eigendecomposition-based embedding of volumetric images. The deep descriptor learning module surpasses the prior hand-crafted descriptors and the descriptor selection. We showcase the efficacy of the core modules, i.e., the spectral embedding approximation and descriptor learning, for volumetric image correspondence and the atlas-based registration on two volumetric image datasets. The proposed method achieves comparable correspondence accuracy with the state-of-the-art deep registration models, resilient to pose and shape perturbations.

Automated assessment of mandibular shape asymmetry in 3-dimensions.

INTRODUCTION: This study aimed to develop an automatic pipeline for analyzing mandibular shape asymmetry in 3-dimensions. METHODS: Forty patients with skeletal Class I pattern and 80 patients with skeletal Class III pattern were used. The mandible was automatically segmented from the cone-beam computed tomography images using a U-net deep learning network. A total of 17,415 uniformly sampled quasi-landmarks were automatically identified on the mandibular surface via a template mapping technique. After alignment with the robust Procrustes superimposition, the pointwise surface-to-surface distance between original and reflected mandibles was visualized in a color-coded map, indicating the location of asymmetry. The degree of overall mandibular asymmetry and the asymmetry of subskeletal units were scored using the root-mean-squared-error between the left and right sides. These asymmetry parameters were compared between the skeletal Class I and skeletal Class III groups. RESULTS: The mandible shape was significantly more asymmetrical in patients with skeletal Class III pattern with positional asymmetry. The condyles were identified as the most asymmetric region in all groups, followed by the coronoid process and the ramus. CONCLUSIONS: This automated approach to quantify mandibular shape asymmetry will facilitate high-throughput image processing for big data analysis. The spatially-dense landmarks allow for evaluating mandibular asymmetry over the entire surface, which overcomes the information loss inherent in conventional linear distance or angular measurements. Precise quantification of the asymmetry can provide important information for individualized diagnosis and treatment planning in orthodontics and orthognathic surgery.

Deep Volumetric Descriptor Learning for Dense Correspondence of Cone-Beam Computed Tomography via Spectral Maps.

The deep neural network has achieved great success in 3D volumetric correspondence. These methods infer the dense displacement or velocity fields directly from the extracted volumetric features without addressing the intrinsic structure correspondence, being prone to shape and pose variations. On the other hand, the spectral maps address the intrinsic structure matching in the low dimensional embedding space, remain less involved in volumetric image correspondence. This paper presents an unsupervised deep volumetric descriptor learning neural network via the low dimensional spectral maps to address the dense volumetric correspondence. The neural network is optimized by a novel criterion on descriptor alignments in the spectral domain regarding the supervoxel graph. Aside from the deep convolved multi-scale features, we explicitly address the supervoxel-wise spatial and cross-channel dependencies to enrich deep descriptors. The dense volumetric correspondence is formulated as the low-dimensional spectral mapping. The proposed approach has been applied to both synthetic and clinically obtained cone-beam computed tomography images to establish dense supervoxel-wise and up-scaled voxel-wise correspondences. Extensive series of experimental results demonstrate the contribution of the proposed approach in volumetric descriptor extraction and consistent correspondence, facilitating attribute transfer for segmentation and landmark location. The proposed approach performs favorably against the state-of-the-art volumetric descriptors and the deep registration models, being resilient to pose or shape variations and independent of the prior transformations.

Emergence of a novel reassortant avian influenza virus (H10N3) in Eastern China with high pathogenicity and respiratory droplet transmissibility to mammals.

Decades have passed since the first discovery of H10-subtype avian influenza virus (AIV) in chickens in 1949, and it has been detected in many species including mammals such as minks, pigs, seals and humans. Cases of human infections with H10N8 viruses identified in China in 2013 have raised widespread attention. Two novel reassortant H10N3 viruses were isolated from chickens in December 2019 in eastern China during routine surveillance for AIVs. The internal genes of these viruses were derived from genotype S (G57) H9N2 and were consistent with H5N6, H7N9 and H10N8, which cause fatal infections in humans. Their viral pathogenicity and transmissibility were further studied in different animal models. The two H10N3 isolates had low pathogenicity in chickens and were transmitted between chickens via direct contact. These viruses were highly pathogenic in mice and could be transmitted between guinea pigs via direct contact and respiratory droplets. More importantly, these viruses can bind to both human-type SAα-2,6-Gal receptors and avian-type SAα-2,3-Gal receptors. Asymptomatic shedding in chickens and good adaptability to mammals of these H10N3 isolates would make it easier to transmit to humans and pose a threat to public health.

Unsupervised random forest for affinity estimation.

This paper presents an unsupervised clustering random-forest-based metric for affinity estimation in large and high-dimensional data. The criterion used for node splitting during forest construction can handle rank-deficiency when measuring cluster compactness. The binary forest-based metric is extended to continuous metrics by exploiting both the common traversal path and the smallest shared parent node. The proposed forest-based metric efficiently estimates affinity by passing down data pairs in the forest using a limited number of decision trees. A pseudo-leaf-splitting (PLS) algorithm is introduced to account for spatial relationships, which regularizes affinity measures and overcomes inconsistent leaf assign-ments. The random-forest-based metric with PLS facilitates the establishment of consistent and point-wise correspondences. The proposed method has been applied to automatic phrase recognition using color and depth videos and point-wise correspondence. Extensive experiments demonstrate the effectiveness of the proposed method in affinity estimation in a comparison with the state-of-the-art.

Deformable registration of lateral cephalogram and cone-beam computed tomography image.

PURPOSE: This study aimed to design and evaluate a novel method for the registration of 2D lateral cephalograms and 3D craniofacial cone-beam computed tomography (CBCT) images, providing patient-specific 3D structures from a 2D lateral cephalogram without additional radiation exposure. METHODS: We developed a cross-modal deformable registration model based on a deep convolutional neural network. Our approach took advantage of a low-dimensional deformation field encoding and an iterative feedback scheme to infer coarse-to-fine volumetric deformations. In particular, we constructed a statistical subspace of deformation fields and parameterized the nonlinear mapping function from an image pair, consisting of the target 2D lateral cephalogram and the reference volumetric CBCT, to a latent encoding of the deformation field. Instead of the one-shot registration by the learned mapping function, a feedback scheme was introduced to progressively update the reference volumetric image and to infer coarse-to-fine deformations fields, accounting for the shape variations of anatomical structures. A total of 220 clinically obtained CBCTs were used to train and validate the proposed model, among which 120 CBCTs were used to generate a training dataset with 24k paired synthetic lateral cephalograms and CBCTs. The proposed approach was evaluated on the deformable 2D-3D registration of clinically obtained lateral cephalograms and CBCTs from growing and adult orthodontic patients. RESULTS: Strong structural consistencies were observed between the deformed CBCT and the target lateral cephalogram in all criteria. The proposed method achieved state-of-the-art performances with the mean contour deviation of 0.41 ± 0.12 mm on the anterior cranial base, 0.48 ± 0.17 mm on the mandible, and 0.35 ± 0.08 mm on the maxilla, respectively. The mean surface mesh ranged from 0.78 to 0.97 mm on various craniofacial structures, and the LREs ranged from 0.83 to 1.24 mm on the growing datasets regarding 14 landmarks. The proposed iterative feedback scheme handled the structural details and improved the registration. The resultant deformed volumetric image was consistent with the target lateral cephalogram in both 2D projective planes and 3D volumetric space regarding the multicategory craniofacial structures. CONCLUSIONS: The results suggest that the deep learning-based 2D-3D registration model enables the deformable alignment of 2D lateral cephalograms and CBCTs and estimates patient-specific 3D craniofacial structures.

Differential microRNA Expression in Newcastle Disease Virus-Infected HeLa Cells and Its Role in Regulating Virus Replication.

As an oncolytic virus, Newcastle disease virus (NDV) can specifically kill tumor cells and has been tested as an attractive oncolytic agent for cancer virotherapy. Virus infection can trigger the changes of the cellular microRNA (miRNA) expression profile, which can greatly influence viral replication and pathogenesis. However, the interplay between NDV replication and cellular miRNA expression in tumor cells is still largely unknown. In the present study, we compared the profiles of cellular miRNAs in uninfected and NDV-infected HeLa cells by small RNA deep sequencing. Here we report that NDV infection in HeLa cells significantly changed the levels of 40 miRNAs at 6 h post-infection (hpi) and 62 miRNAs at 12 hpi. Among 23 highly differentially expressed miRNAs, NDV infection greatly promoted the levels of 3 miRNAs and suppressed the levels of 20 miRNAs at both time points. These 23 miRNAs are predicted to target various genes involved in virus replication and antiviral immunity such as ErbB, Jak-STAT, NF-kB and RIG-I-like receptor. Verification of deep sequencing results by quantitative RT-PCR showed that 9 out of 10 randomly selected miRNAs chosen from this 23-miRNA pool were consistent with deep sequencing data, including 6 down-regulated and 3 up-regulated. Further functional research revealed that hsa-miR-4521, a constituent in this 23-miRNA pool, inhibited NDV replication in HeLa cells. Moreover, dual-luciferase and gene expression array uncovered that the member A of family with sequence similarity 129 (FAM129A) was directly targeted by hsa-miR-4521 and positively regulated NDV replication in HeLa cells, indicating that hsa-miR-4521 may regulate NDV replication via interaction with FAM129A. To our knowledge, this is the first report of the dynamic cellular miRNA expression profile in tumor cells after NDV infection and may provide a valuable basis for further investigation on the roles of miRNAs in NDV-mediated oncolysis.

Thickness change of masseter muscles and the surrounding soft tissues in female patients during orthodontic treatment: a retrospective study.

BACKGROUND: Facial esthetics is a major concern of orthodontic patients. This study aims to evaluate orthodontic treatment-related thickness changes of the masseter muscles and surrounding soft tissues and the potential factors that would influence these changes during orthodontic treatment in female adults. METHODS: Forty-two female adult patients were included in this retrospective study and were divided into extraction (n = 22) and nonextraction (n = 20) groups. Pretreatment and posttreatment cone-beam computed tomography (CBCT) images were superimposed and reconstructed. The thickness changes of the masseter area of facial soft tissue (MAS), masseter muscles (MM) and surrounding fat tissue (FT) were measured. Pretreatment age, treatment duration, sagittal relationship (ANB), and vertical relationship (Frankfort-mandibular plane angle, FMA)-related MAS, MM and FT changes were compared between extraction and nonextraction groups. Spearman's correlation coefficient was calculated between the above variables. Regression analysis was conducted to confirm the causal relations of the variables. RESULTS: The thickness of MAS and MM significantly decreased in both groups, with larger decreases (> 1 mm) in the extraction group. There were strong correlations (r > 0.7) between the thickness decrease in MAS and MM in both groups and moderate correlations (r > 0.4) between MAS and FT in the nonextraction group. A significantly greater decrease of MAS and MM were found to be moderately correlated with a smaller FMA (r > 0.4) in the extraction group. Scatter plots and regression analysis confirmed these correlations. CONCLUSIONS: Masseter muscles and the surrounding soft tissue exhibited a significant decrease in thickness during orthodontic treatment in female adults. Low-angle patients experienced a greater decrease in soft tissue thickness in the masseter area in the extraction case. But the thickness changes were clinically very small in most patients.

Gga-miR-19b-3p Inhibits Newcastle Disease Virus Replication by Suppressing Inflammatory Response via Targeting RNF11 and ZMYND11.

Newcastle disease (ND), an acute and highly contagious avian disease caused by virulent Newcastle disease virus (NDV), often results in severe economic losses worldwide every year. Although it is clear that microRNAs (miRNAs) are implicated in modulating innate immune response to invading microbial pathogens, their role in host defense against NDV infection remains largely unknown. Our prior study indicates that gga-miR-19b-3p is up-regulated in NDV-infected DF-1 cells (a chicken embryo fibroblast cell line) and functions to suppress NDV replication. Here we report that overexpression of gga-miR-19b-3p promoted the production of NDV-induced inflammatory cytokines and suppressed NDV replication, whereas inhibition of endogenous gga-miR-19b-3p expression had an opposite effect. Dual-luciferase and gene expression array analyses revealed that gga-miR-19b-3p directly targets the mRNAs of ring finger protein 11 (RNF11) and zinc-finger protein, MYND-type containing 11 (ZMYND11), two negative regulators of nuclear factor kappa B (NF-κB) signaling, in DF-1 cells. RNF11 and ZMYND11 silencing by small interfering RNA (siRNA) induced NF-κB activity and inflammatory cytokine production, and suppressed NDV replication; whereas ectopic expression of these two proteins exhibited an opposite effect. Our study provides evidence that gga-miR-19b-3p activates NF-κB signaling by targeting RNF11 and ZMYND11, and that enhanced inflammatory cytokine production is likely responsible for the suppression of NDV replication.

MicroRNA Expression Profiling in Newcastle Disease Virus-Infected DF-1 Cells by Deep Sequencing.

Newcastle disease virus (NDV), causative agent of Newcastle disease (ND), is one of the most devastating pathogens for poultry industry worldwide. MicroRNAs (miRNAs) are non-coding RNAs that regulate gene expression by regulating mRNA translation efficiency or mRNA abundance through binding to mRNA directly. Accumulating evidence has revealed that cellular miRNAs can also affect virus replication by controlling host-virus interaction. To identify miRNA expression profile and explore the roles of miRNA during NDV replication, in this study, small RNA deep sequencing was performed of non-inoculated DF-1 cells (chicken embryo fibroblast cell line) and JS 5/05-infected cells collected at 6 and 12 h post infection (hereafter called mock' NDV-6 h, and NDV-12 h groups respectively). A total of 73 miRNAs of NDV-6 h group and 64miRNAs of NDV-12 h group were significantly differentially expressed (SDE) when compared with those in mock group. Meanwhile, 50 SDE miRNAs, including 48 up- and 2 down-regulated, showed the same expression patterns in NDV-6 h and NDV-12 h groups. qRT-PCR validation of 15 selected miRNAs' expression patterns was consistent with deep sequencing. To investigate the role of these SDE miRNAs in NDV replication, miRNA mimics and inhibitors were transfected into DF-1 cells followed by NDV infection. The results revealed that gga-miR-451 and gga-miR-199-5p promoted NDV replication while gga-miR-19b-3p and gga-miR-29a-3p inhibited NDV replication. Further function research demonstrated gga-miR-451 suppressed NDV-induced inflammatory response via targeting YWHAZ (tyrosine3-monooxygenase/tryptophan5-monooxygenase activation protein zeta). Overall, our study presented a global miRNA expression profile in DF-1 cells in response to NDV infection and verified the roles of some SDE miRNAs in NDV replication which will underpin further studies of miRNAs' roles between the host and the virus.

Spatially Consistent Supervoxel Correspondences of Cone-Beam Computed Tomography Images.

Establishing dense correspondences of cone-beam computed tomography (CBCT) images is a crucial step for the attribute transfer and morphological variation assessment in clinical orthodontics. In this paper, a novel method, unsupervised spatially consistent clustering forest, is proposed to tackle the challenges for automatic supervoxel-wise correspondences of CBCT images. A complexity analysis of the proposed method with respect to the clustering hypotheses is provided with a data-dependent learning guarantee. The learning bound considers both the sequential tree traversals determined by questions stored in branch nodes and the clustering compactness of leaf nodes. A novel tree-pruning algorithm, guided by the learning bound, is also proposed to remove locally inconsistent leaf nodes. The resulting forest yields spatially consistent affinity estimations, thanks to the pruning penalizing trees with inconsistent leaf assignments and the combinational contextual feature channels used to learn the forest. A forest-based metric is utilized to derive the pairwise affinities and dense correspondences of CBCT images. The proposed method has been applied to the label propagation of clinically captured CBCT images. In the experiments, the method outperforms variants of both supervised and unsupervised forest-based methods and state-of-the-art label-propagation methods, achieving the mean dice similarity coefficients of 0.92, 0.89, 0.94, and 0.93 for the mandible, the maxilla, the zygoma arch, and the teeth data, respectively.

Superimposition of Cone-Beam Computed Tomography Images by Joint Embedding.

OBJECTIVE: The superimposition of cone-beam computed tomography (CBCT) images is an essential step to evaluate shape variations of pre and postorthodontic operations due to pose variations and the bony growth. The aim of this paper is to present and discuss the latest accomplishments in voxel-based craniofacial CBCT superimpositions along with structure discriminations. METHODS: We propose a CBCT superimposition method based on joint embedding of subsets extracted from CBCT images. The subset is defined at local extremes of the first-order difference of Gaussian-smoothed volume images to reduce the data involved in the computation. A rotation-invariant integral operator is proposed as the context-aware textural descriptor of subsets. We cope with subset correspondences by joint embedding with matching identifications in manifolds, which take into account the structure of subsets as a whole to avoid mapping ambiguities. Once given subset correspondences, the rigid transformations, as well as the superimposition of volume images, are obtained. Our system allows users to specify the structure-of-interest based on a semisupervised label propagation technique. RESULTS: The performance of the proposed method is evaluated on ten pairs of pre and postoperative CBCT images of adult patients and ten pairs of growing patients, respectively. The experiments demonstrate that the craniofacial CBCT superimposition can be performed effectively, and outperform state of the arts. CONCLUSION: The integration of sparse subsets with context-aware spherical intensity integral descriptors and correspondence establishment by joint embedding enables the reliable and efficient CBCT superimposition. SIGNIFICANCE: The potential of CBCT superimposition techniques discussed in this paper is highlighted and related challenges are addressed.

3D exemplar-based random walks for tooth segmentation from cone-beam computed tomography images.

PURPOSE: Tooth segmentation is an essential step in acquiring patient-specific dental geometries from cone-beam computed tomography (CBCT) images. Tooth segmentation from CBCT images is still a challenging task considering the comparatively low image quality caused by the limited radiation dose, as well as structural ambiguities from intercuspation and nearby alveolar bones. The goal of this paper is to present and discuss the latest accomplishments in semisupervised tooth segmentation with adaptive 3D shape constraints. METHODS: The authors propose a 3D exemplar-based random walk method of tooth segmentation from CBCT images. The proposed method integrates semisupervised label propagation and regularization by 3D exemplar registration. To begin with, the pure random walk method is to get an initial segmentation of the teeth, which tends to be erroneous because of the structural ambiguity of CBCT images. And then, as an iterative refinement, the authors conduct a regularization by using 3D exemplar registration, as well as label propagation by random walks with soft constraints, to improve the tooth segmentation. In the first stage of the iteration, 3D exemplars with well-defined topologies are adapted to fit the tooth contours, which are obtained from the random walks based segmentation. The soft constraints on voxel labeling are defined by shape-based foreground dentine probability acquired by the exemplar registration, as well as the appearance-based probability from a support vector machine (SVM) classifier. In the second stage, the labels of the volume-of-interest (VOI) are updated by the random walks with soft constraints. The two stages are optimized iteratively. Instead of the one-shot label propagation in the VOI, an iterative refinement process can achieve a reliable tooth segmentation by virtue of exemplar-based random walks with adaptive soft constraints. RESULTS: The proposed method was applied for tooth segmentation of twenty clinically captured CBCT images. Three metrics, including the Dice similarity coefficient (DSC), the Jaccard similarity coefficient (JSC), and the mean surface deviation (MSD), were used to quantitatively analyze the segmentation of anterior teeth including incisors and canines, premolars, and molars. The segmentation of the anterior teeth achieved a DSC up to 98%, a JSC of 97%, and an MSD of 0.11 mm compared with manual segmentation. For the premolars, the average values of DSC, JSC, and MSD were 98%, 96%, and 0.12 mm, respectively. The proposed method yielded a DSC of 95%, a JSC of 89%, and an MSD of 0.26 mm for molars. Aside from the interactive definition of label priors by the user, automatic tooth segmentation can be achieved in an average of 1.18 min. CONCLUSIONS: The proposed technique enables an efficient and reliable tooth segmentation from CBCT images. This study makes it clinically practical to segment teeth from CBCT images, thus facilitating pre- and interoperative uses of dental morphologies in maxillofacial and orthodontic treatments.

Evaluation of facial attractiveness for patients with malocclusion: a machine-learning technique employing Procrustes.

OBJECTIVE: To establish an objective method for evaluating facial attractiveness from a set of orthodontic photographs. MATERIALS AND METHODS: One hundred eight malocclusion patients randomly selected from six universities in China were randomly divided into nine groups, with each group containing an equal number of patients with Class I, II, and III malocclusions. Sixty-nine expert Chinese orthodontists ranked photographs of the patients (frontal, lateral, and frontal smiling photos) before and after orthodontic treatment from "most attractive" to "least attractive" in each group. A weighted mean ranking was then calculated for each patient, based on which a three-point scale was created. Procrustes superimposition was conducted on 101 landmarks identified on the photographs. A support vector regression (SVR) function was set up according to the coordinate values of identified landmarks of each photographic set and its corresponding grading. Its predictive ability was tested for each group in turn. RESULTS: The average coincidence rate obtained for comparisons of the subjective ratings with the SVR evaluation was 71.8% according to 18 verification tests. CONCLUSIONS: Geometric morphometrics combined with SVR may be a prospective method for objective comprehensive evaluation of facial attractiveness in the near future.

Unsupervised image matching based on manifold alignment.

This paper challenges the issue of automatic matching between two image sets with similar intrinsic structures and different appearances, especially when there is no prior correspondence. An unsupervised manifold alignment framework is proposed to establish correspondence between data sets by a mapping function in the mutual embedding space. We introduce a local similarity metric based on parameterized distance curves to represent the connection of one point with the rest of the manifold. A small set of valid feature pairs can be found without manual interactions by matching the distance curve of one manifold with the curve cluster of the other manifold. To avoid potential confusions in image matching, we propose an extended affine transformation to solve the nonrigid alignment in the embedding space. The comparatively tight alignments and the structure preservation can be obtained simultaneously. The point pairs with the minimum distance after alignment are viewed as the matchings. We apply manifold alignment to image set matching problems. The correspondence between image sets of different poses, illuminations, and identities can be established effectively by our approach.

Personalized tooth shape estimation from radiograph and cast.

Three-dimensional geometric information of teeth is usually needed in pre- and postoperative diagnoses of orthodontic dentistry. The computerized tomography can provide comprehensive 3-D teeth geometries. However, there is still a discussion on computed tomography (CT) as a routine in orthodontic dentistry due to radiation dose. Moreover, the CT is useless when a dentist needs to extract 3-D structures from old archive files with only radiographs and casts, where patient's teeth changed ever since. In this paper, we propose a reconstruction framework for patient-specific teeth based on an integration of 2-D radiographs and digitized casts. The reconstruction is under a template-fitting framework. The shape and orientation of teeth templates are tuned in accordance with patient's radiographs. Specially, the tooth root morphology is controlled by 2-D contours in radiographs. With ray tracing and a contour plane assumption, 2-D root contours in radiographs are projected back to 3-D space, and guide tooth root deformations. Moreover, the template's crown is deformed nonrigidly to fit digitized casts that bear patient's crown details. The system allows 3-D tooth reconstruction with patient-specific geometric details from just casts and 2-D radiographs.