Oral health-related quality of life, adaptation/discomfort during open bite treatment with spurs: complementary analysis from a randomized clinical trial.

This single-center trial aimed to longitudinally compare the oral health-related quality of life (OHRQOL), adaptation and discomfort during anterior open bite (AOB) treatment with lingual spurs and build-ups (SBU) versus spurs only (S) approaches. Children (7-11 years) with AOB were randomly allocated into two treatment groups (SBU or S). The Child Perception Questionnaire (CPQ8-10) was applied 1 and 12 months after installation of the appliances. Questionnaires evaluating functional adaptation and discomfort during the first month of treatment were also applied. A visual analog scale (VAS) was used in these questionnaires. Generalized mixed models were used for analyzing OHRQOL and discomfort data. Generalized linear models were used to assess adaptation outcomes (α = 0.05). The SBU group included 24 patients (7 males and 17 females; mean age 8.2 years) and the S group included 25 patients (11 males and 14 females; mean age 8.3 years). Regardless of the treatment type, overall OHRQOL scores at 12 months were 0.69 times those recorded at 1 month after the appliances installation (i.e., ~ 31% reduction; exp (ß) = 0.69; 95% CI: 0.55, 0.88). A significant interaction between treatment and time was detected for the 'functional limitations' domain. For this domain, a significant improvement from the first to the twelfth month was observed in the S group (P < 0.001). Patients in both treatment groups showed similar and easy adaptation to the appliances. Independent of the type of treatment, tongue-related discomfort decreased over time. One week and one month after the appliance's delivery, the discomfort scores were 0.19 (i.e., ~ 81% reduction; exp (ß) = 0.19; 95% CI: 0.13, 0.28; P < 0.001) and 0.02 (i.e., ~ 98% reduction; exp (ß) = 0.02; 95% CI: 0.01, 0.07; P < 0.001) times, respectively, those issued immediately after the installation of the appliances. Regardless of treatment type; overall OHRQOL improved from the first to the twelfth month of AOB treatment. The functional limitations score decreased in the S group. Children showed easy adaptation, and their discomfort decreased 1 week after the installation of the appliances.Trial registration: Clinicaltrials.gov; NCT03702881, date of registration: October 11, 2018.

Automated artificial intelligence-based three-dimensional comparison of orthodontic treatment outcomes with and without piezocision surgery.

OBJECTIVE(S): This study aims to evaluate the influence of the piezocision surgery in the orthodontic biomechanics, as well as in the magnitude and direction of tooth movement in the mandibular arch using novel artificial intelligence (AI)-automated tools. MATERIALS AND METHODS: Nineteen patients, who had piezocision performed in the lower arch at the beginning of treatment with the goal of accelerating tooth movement, were compared to 19 patients who did not receive piezocision. Cone beam computed tomography (CBCT) and intraoral scans (IOS) were acquired before and after orthodontic treatment. AI-automated dental tools were used to segment and locate landmarks in dental crowns from IOS and root canals from CBCT scans to quantify 3D tooth movement. Differences in mesial-distal, buccolingual, intrusion and extrusion linear movements, as well as tooth long axis angulation and rotation were compared. RESULTS: The treatment time for the control and experimental groups were 13.2 ± 5.06 and 13 ± 5.52 months respectively (P = .176). Overall, anterior and posterior tooth movement presented similar 3D linear and angular changes in the groups. The piezocision group demonstrated greater (P = .01) mesial long axis angulation of lower right first premolar (4.4 ± 6°) compared with control group (0.02 ± 4.9°), while the mesial rotation was significantly smaller (P = .008) in the experimental group (0.5 ± 7.8°) than in the control (8.5 ± 9.8°) considering the same tooth. CONCLUSION: The open source-automated dental tools facilitated the clinicians' assessment of piezocision treatment outcomes. The piezocision surgery prior to the orthodontic treatment did not decrease the treatment time and did not influence in the orthodontic biomechanics, leading to similar tooth movements compared to conventional treatment.

Asymmetry in Class II subdivision malocclusion: Assessment based on 3D surface models.

INTRODUCTION: There is currently no consensus in the literature whether the aetiology of a Class II subdivision is dental, skeletal or both. The aim of this study was to identify and quantify skeletal and dental asymmetries in Class II subdivision malocclusions. METHODS: CBCTs from 33 Class II subdivision malocclusion patients were used to construct 3D volumetric label maps. Eighteen landmarks were identified. The original scan and associated 3D volumetric label map were mirrored. Registration of the original and mirrored images relative to the anterior cranial base, maxilla and mandible were performed. Surface models were generated, and 3D differences were quantified. Statistical analysis was performed. RESULTS: Anterior cranial base registration showed significant differences for fossa vertical difference, fossa roll, mandibular yaw, mandibular lateral displacement and lower midline displacement. Regional registrations showed significant differences for antero-posterior (A-P) mandibular length, maxillary roll, A-P maxillary first molar position, maxillary first molar yaw and maxillary first molar roll. Class II subdivision patients also show an asymmetric mandibular length as well as an asymmetric gonial angle. Moderate correlations were found between the A-P molar relationship and fossa A-P difference, mandibular first molar A-P difference, maxillary first molar A-P difference and maxillary first molar yaw. CONCLUSIONS: This study suggests that Class II subdivisions can result from both significant skeletal and dental factors. Skeletal factors include a shorter mandible as well as posterior and higher displacement of the fossa on the Class II side, resulting in mandibular yaw. Dental factors include maxillary and mandibular first molar antero-posterior asymmetry.

Interpretable artificial intelligence for classification of alveolar bone defect in patients with cleft lip and palate.

Cleft lip and/or palate (CLP) is the most common congenital craniofacial anomaly and requires bone grafting of the alveolar cleft. This study aimed to develop a novel classification algorithm to assess the severity of alveolar bone defects in patients with CLP using three-dimensional (3D) surface models and to demonstrate through an interpretable artificial intelligence (AI)-based algorithm the decisions provided by the classifier. Cone-beam computed tomography scans of 194 patients with CLP were used to train and test the performance of an automatic classification of the severity of alveolar bone defect. The shape, height, and width of the alveolar bone defect were assessed in automatically segmented maxillary 3D surface models to determine the ground truth classification index of its severity. The novel classifier algorithm renders the 3D surface models from different viewpoints and captures 2D image snapshots fed into a 2D Convolutional Neural Network. An interpretable AI algorithm was developed that uses features from each view and aggregated via Attention Layers to explain the classification. The precision, recall and F-1 score were 0.823, 0.816, and 0.817, respectively, with agreement ranging from 97.4 to 100% on the severity index within 1 group difference. The new classifier and interpretable AI algorithm presented satisfactory accuracy to classify the severity of alveolar bone defect morphology using 3D surface models of patients with CLP and graphically displaying the features that were considered during the deep learning model's classification decision.

Clear aligner mandibular advancement in growing patients with Class II malocclusion.

Treatment effects occurring during Class II malocclusion treatment with the clear aligner mandibular advancement protocol were evaluated in two growing patients: one male (12 years, 3 months) and one female (11 years, 9 months). Both patients presented with full cusp Class II molar and canine relationships. Intraoral scans and cone-beam computed tomography were acquired before treatment and after mandibular advancement. Three-dimensional skeletal and dental long-axis changes were quantified, in which the dental long axis was determined by registering the dental crowns obtained from intraoral scans to the root canals in cone-beam computed tomography scans obtained at the same time points. Class II correction was achieved by a combination of mandibular skeletal and dental changes. A similar direction of skeletal and dental changes was observed in both patients, with downward and forward displacement of the mandible resulting from the growth of the mandibular condyle and ramus. Dental changes in both patients included mesialization of the mandibular posterior teeth with flaring of mandibular anterior teeth. In these two patients, clear aligner mandibular advancement was an effective treatment modality for Class II malocclusion correction with skeletal and dental effects and facial profile improvement.

Three-dimensional changes after maxillary molar distalization with a miniscrew-anchored cantilever.

OBJECTIVES: To evaluate the changes after maxillary molar distalization in Class II malocclusion using the miniscrew-anchored cantilever with an extension arm. MATERIALS AND METHODS: The sample included 20 patients (9 male, 11 female; mean age 13.21 ± 1.54 years) with Class II malocclusion, treated with the miniscrew-anchored cantilever. Lateral cephalograms and dental models obtained before (T1) and after molar distalization (T2) were evaluated using Dolphin software and 3D Slicer. Superimposition of digital dental models using regions of interest on the palate was performed to evaluate three-dimensional displacement of maxillary teeth. Intragroup change comparisons were performed using dependent t-test and Wilcoxon test (P < 0.05). RESULTS: The maxillary first molars were distalized to overcorrected Class I. The mean distalization time was 0.43 ± 0.13 years. Cephalometric analysis demonstrated significant distal movement of the maxillary first premolar (-1.21 mm, 95% confidence interval [CI]: -0.45, -1.96) and maxillary first (-3.38 mm, 95% CI: -2.88, -3.87) and second molars (-2.12 mm, 95% CI: -1.53, -2.71). Distal movements increased progressively from the incisors to the molars. The first molar showed small intrusion (-0.72 mm, 95% CI: 0.49, -1.34). In the digital model analysis, the first and second molars showed a crown distal rotation of 19.31° ± 5.71° and 10.17° ± 3.84°, respectively. The increase in maxillary intermolar distance, evaluated at the mesiobuccal cusps, was 2.63 ± 1.56 mm. CONCLUSIONS: The miniscrew-anchored cantilever was effective for maxillary molar distalization. Sagittal, lateral, and vertical movements were observed for all maxillary teeth. Distal movement was progressively greater from anterior to posterior teeth.

Automatic landmark identification in cone-beam computed tomography.

OBJECTIVE: To present and validate an open-source fully automated landmark placement (ALICBCT) tool for cone-beam computed tomography scans. MATERIALS AND METHODS: One hundred and forty-three large and medium field of view cone-beam computed tomography (CBCT) were used to train and test a novel approach, called ALICBCT that reformulates landmark detection as a classification problem through a virtual agent placed inside volumetric images. The landmark agents were trained to navigate in a multi-scale volumetric space to reach the estimated landmark position. The agent movements decision relies on a combination of DenseNet feature network and fully connected layers. For each CBCT, 32 ground truth landmark positions were identified by 2 clinician experts. After validation of the 32 landmarks, new models were trained to identify a total of 119 landmarks that are commonly used in clinical studies for the quantification of changes in bone morphology and tooth position. RESULTS: Our method achieved a high accuracy with an average of 1.54 ± 0.87 mm error for the 32 landmark positions with rare failures, taking an average of 4.2 second computation time to identify each landmark in one large 3D-CBCT scan using a conventional GPU. CONCLUSION: The ALICBCT algorithm is a robust automatic identification tool that has been deployed for clinical and research use as an extension in the 3D Slicer platform allowing continuous updates for increased precision.

Three-dimensional comparison of bone-borne and tooth-bone-borne maxillary expansion in young adults with maxillary skeletal deficiency.

OBJECTIVE: To compare the transverse dental and skeletal changes in patients treated with bone-anchored palatal expander (bone-borne, BB) compared to patients treated with tooth and bone-anchored palatal expanders (tooth-bone-borne, TBB) using cone-beam computer tomography (CBCT) and 3D image analysis. METHODS: The sample comprised 30 patients with transverse maxillary discrepancy treated with two different types of appliances: bone-borne (Group BB) and tooth-bone-borne (Group TBB) expanders. CBCT scans were acquired before (T1) and after completion of maxillary expansion (T2); the interval was 5.4 ± 3.4 and 6.2 ± 2.1 months between the T1 and the T2 scans of Group TBB (tooth-bone-borne) and Group BB (bone-borne), respectively. Transverse, anteroposterior and vertical linear and angular three-dimensional dentoskeletal changes were assessed after cranial base superimposition. RESULTS: Both groups displayed marked transverse skeletal expansion with a greater ratio of skeletal to dental changes. Greater changes at the nasal cavity, zygoma and orbital levels were found in Group BB. A relatively parallel sutural opening in an anterior-posterior direction was observed in Group TBB; however, the Group BB presented a somewhat triangular (V-shaped) opening of the suture that was wider anteriorly. Small downward-forward displacements were observed in both groups. Asymmetric expansion occurred in approximately 50% of the patients in both groups. CONCLUSION: Greater skeletal vs dental expansion ratio and expansion of the circummaxillary regions were found in Group BB, the group in which a bone-borne expander was used. Both groups presented skeletal and dental changes, with a similar amount of posterior palate expansion. Asymmetric expansion was observed in both groups.

DENTALMODELSEG: FULLY AUTOMATED SEGMENTATION OF UPPER AND LOWER 3D INTRA-ORAL SURFACES.

In this paper, we present a deep learning-based method for surface segmentation. This technique consists of acquiring 2D views and extracting features from the surface such as the normal vectors. The rendered images are analyzed with a 2D convolutional neural network, such as a UNET. We test our method in a dental application for the segmentation of dental crowns. The neural network is trained for multi-class segmentation, using image labels as ground truth. A 5-fold cross-validation was performed, and the segmentation task achieved an average Dice of 0.97, sensitivity of 0.98 and precision of 0.98. Our method and algorithms are available as a 3DSlicer extension.

Automatic multi-anatomical skull structure segmentation of cone-beam computed tomography scans using 3D UNETR.

The segmentation of medical and dental images is a fundamental step in automated clinical decision support systems. It supports the entire clinical workflow from diagnosis, therapy planning, intervention, and follow-up. In this paper, we propose a novel tool to accurately process a full-face segmentation in about 5 minutes that would otherwise require an average of 7h of manual work by experienced clinicians. This work focuses on the integration of the state-of-the-art UNEt TRansformers (UNETR) of the Medical Open Network for Artificial Intelligence (MONAI) framework. We trained and tested our models using 618 de-identified Cone-Beam Computed Tomography (CBCT) volumetric images of the head acquired with several parameters from different centers for a generalized clinical application. Our results on a 5-fold cross-validation showed high accuracy and robustness with a Dice score up to 0.962±0.02. Our code is available on our public GitHub repository.

Presurgical orthodontic decompensation with clear aligners.

INTRODUCTION: Orthodontists, surgeons, and patients have taken an interest in using clear aligners in combination with orthognathic surgery. This study aimed to evaluate the accuracy of tooth movements with clear aligners during presurgical orthodontics using novel 3-dimensional superimposition techniques. METHODS: The study sample consisted of 20 patients who have completed presurgical orthodontics using Invisalign clear aligners. Initial (pretreatment) digital dental models, presurgical digital dental models, and ClinCheck prediction models were obtained. Presurgical models were superimposed onto initial ones using stable anatomic landmarks; ClinCheck models were superimposed onto presurgical models using surface best-fit superimposition. Five hundred forty-five teeth were measured for 3 angular movements (buccolingual torque, mesiodistal tip, and rotation) and 4 linear movements (buccolingual, mesiodistal, vertical, and total scalar displacement). The predicted tooth movement was compared with the achieved amount for each movement and tooth, using both percentage accuracy and numerical difference. RESULTS: Average percentage accuracy (63.4% ± 11.5%) was higher than in previously reported literature. The most accurate tooth movements were buccal torque and mesial displacement compared with lingual torque and distal displacement, particularly for mandibular posterior teeth. Clinically significant inaccuracies were found for the buccal displacement of maxillary second molars, lingual displacement of all molars, intrusion of mandibular second molars, the distal tip of molars, second premolars, and mandibular first premolars, buccal torque of maxillary central and lateral incisors, and lingual torque of premolars and molars. CONCLUSIONS: Superimposition techniques used in this study lay the groundwork for future studies to analyze advanced clear aligner patients. Invisalign is a treatment modality that can be considered for presurgical orthodontics-tooth movements involved in arch leveling and decompensation are highly accurate when comparing the simulated and the clinically achieved movements.

Three-dimensional dentoalveolar changes in open bite treatment in mixed dentition, spurs/posterior build-ups versus spurs alone: 1-year follow-up randomized clinical trial.

This randomized clinical trial aimed to compare the three-dimensional dentoalveolar maxillary changes after anterior open bite treatment with bonded spurs and build-ups versus bonded spurs alone. Patients from 7 to 11 years of age with anterior open bite were randomly allocated into two groups. Bonded spurs and posterior build-ups were used in the experimental group and only bonded spurs were used in the comparison group. The randomization sequence was generated at www.randomization.com . Opaque, sealed and sequentially numbered envelopes were part of the allocation concealment. Digital dental models were acquired before (T1) and after 12 months of treatment (T2) and de-identified for analysis purposes. Three-dimensional changes of maxillary permanent incisors and first molars were evaluated by means of T1 and T2 dental model superimposition. Landmark-based registration on the posterior teeth and registration on the palate using regions of interest were performed. T or Mann-Whitney U tests were used for intergroup comparisons (P < 0.05). Mean difference (MD) and 95% confidence interval (CI) were calculated. Twenty-four children (17 girls and 7 boys) were included in the experimental group (mean age 8.22 ± 1.06 years) and 25 children (14 girls and 11 boys) were included in the comparison group (mean age 8.30 ± 0.99 years). After 12 months of treatment, inferior displacements of maxillary incisors were similar in the experimental (1.55-2.92 mm) and comparison (1.40-2.65 mm) groups. Inferior displacement of the maxillary molars was also similar in both groups (MD: - 0.13 mm; 95% CI - 0.38, 0.12). The experimental and comparison groups showed medial and lateral displacements of the permanent first molars, respectively (MD, - 0.31 mm; 95% CI - 0.51, - 0.11). Lingual inclination of the permanent first molars were observed in the experimental group and buccal inclination in the comparison group (MD, - 2.16°; 95% CI - 3.72, - 0.60). Similar three-dimensional displacements of maxillary central and lateral incisors, and inferior displacements of maxillary permanent first molars were observed in both groups. Bonded spurs associated with posterior build-ups demonstrated some medial displacement and lingual inclination of the maxillary permanent first molars while opposite changes were noticed in the comparison group.Trial registration: Clinicaltrials.gov; NCT03702881, date of registration: October 11, 2018.

Interdisciplinary dental management of patient with oligodontia and maxillary hypoplasia: a case report.

BACKGROUND: The craniofacial developmental abnormality can significantly complicate the oral rehabilitation of patients with oligodontia. This case report describes an interdisciplinary approach that took 7 years to successfully treat a young patient with non-syndromic oligodontia and midface deficiency. CASE PRESENTATION: A 14-year-old patient with complex oral and maxillofacial conditions and diagnosis of oligodontia presented to our clinic. In addition to 4 retained deciduous teeth and congenitally missing 10 permanent teeth, dentofacial findings included maxillary and malar deficiency with a concave facial profile, Angle Class III malocclusion, and poor dental esthetics. The interdisciplinary treatment included pre-surgical orthodontic decompensation, high Le Fort I maxillary osteotomy, postsurgical orthodontic therapy, osseous ridge augmentation using recombinant human bone morphogenetic protein-2 (rhBMP-2), interim removable partial denture, dental implant installation, interim implant prostheses, and final prosthetic rehabilitation. CONCLUSIONS: The successful treatment of patients with oligodontia and complex dentofacial abnormalities requires the close and orderly collaboration among orthodontist, oral maxillofacial surgeon, and prosthodontist. Within the limitations of this case report, presented interdisciplinary approaches may optimize the oral rehabilitation outcome in patients with similar clinical challenges. A prospective clinical investigation is desired to verify the benefit of presented interdisciplinary approach.

Three-dimensional evaluation of skeletal and dental effects of treatment with maxillary skeletal expansion.

INTRODUCTION: The objective was to determine the skeletal and dental changes with microimplant assisted rapid palatal expansion (MARPE) appliances in growing (GR) and nongrowing (NG) patients using cone-beam computed tomography and 3-dimensional imaging analysis. METHODS: The sample consisted of 25 patients with transverse maxillary discrepancy treated with a maxillary skeletal expander, a type of MARPE appliance. Cone-beam computed tomography scans were taken before and after maxillary expansion; the interval was 6.0 ± 4.3 months. The sample was divided into GR and NG groups using cervical vertebral and midpalatal suture maturation. Linear and angular 3-dimensional dentoskeletal changes were assessed after cranial base superimposition. Groups were compared with independent-samples t test (P <0.05). RESULTS: Both groups displayed marked transverse changes with a similar ratio of skeletal to dental transverse changes and parallel sutural opening from the posterior nasal spine-anterior nasal spine; a similar amount of expansion occurred in the anterior and the posterior regions of the maxilla. The maxilla expanded skeletally without rotational displacements in both groups. The small downward-forward displacements were similar in both groups, except that the GR group had a significantly greater vertical displacement of the canines (GR, 1.7 ±1.0 mm; NG, 0.6 ± 0.8 mm; P = 0.02) and anterior nasal spine (GR, 1.1 ± 0.6 mm; NG, 0.5 ± 0.5 mm; P = 0.004). CONCLUSIONS: Treatment of patients with MARPE appliance is effective in GR and NG patients. Although greater skeletal and dental changes were observed in GR patients, a similar ratio of skeletal to dental transverse changes was observed in both groups.

Dental long axes using digital dental models compared to cone-beam computed tomography.

OBJECTIVE: Standard methods of evaluating tooth long axes are not comparable (digital dental models [DDMs], panoramic and cephalometric radiographs) or expose patients to more radiation (cone-beam computed tomography [CBCT]). This study aimed to compare angular changes in tooth long axes using DDMs vs using CBCTs. SETTINGS AND SAMPLE POPULATION: Secondary data analysis of DDMs and CBCTs, taken before and after orthodontic treatment with piezocision of 24 patients. METHODS: Angular changes in tooth long axes were evaluated using landmarks on first molars (centre of the occlusal surface and centre of the furcation), canines and incisors (cusp tip and centre of the root at the cementoenamel junction). Wilcoxon test, intraclass correlation coefficient (ICC) and Bland-Altman plots were used to test intra- and inter-rater agreement and compare DDM and CBCT measurements. RESULTS: The mesiodistal angulation and buccolingual inclination DDM measurements were reproducible. Overall mean differences between DDM and CBCT measurements of mesiodistal angulation, 1.9°±1.5°, and buccolingual inclination, 2.2 ± 2.2°, were not significant for all teeth. ICC between DDM and CBCT measurements ranged from good (0.85 molars) to excellent (0.94 canines; 0.96 incisors). The percentages of measurements outside the range of ±5 were 17.4% for molars, 13.8% for canines and 4.5% for incisors. CONCLUSIONS: DDM assessment of changes in tooth long axes has good reproducibility and yields comparable measurements to those obtained from CBCT within a 5° range. These findings lay the groundwork for machine learning approaches that synthesize crown and root canal information towards planning tooth movement without the need for ionizing radiation scans.

Comparison and reproducibility of three methods for maxillary digital dental model registration in open bite patients.

OBJECTIVE: To compare and assess the reproducibility of 3 methods for registration of maxillary digital dental models in patients with anterior open bite. Settings and sample population Digital dental models of 16 children with an anterior open bite in the mixed dentition were obtained before (T1) and after 12 months of treatment with bonded spurs (T2). METHODS: Landmarks were placed on all T2 models and 3 registration methods (R1, R2 and R3) were independently performed by 2 observers. R1 was based on 10 landmarks placed on posterior teeth. R2 was based on 5 landmarks on the palate (2 anterior, 2 posterior and 1 central). R3 used regions of interest around the 5 palatal landmarks used in R2. The differences between the registration methods were calculated by comparing the mean differences and standard deviations between the corresponding x, y and z coordinates of 6 corresponding landmarks in the T2 registered models. Repeated measures analysis of variance followed by post-hoc Bonferroni tests were used for comparisons (P < .05). The agreement between methods and the intra and interobserver reproducibility were assessed with Bland-Altman tests and intraclass correlation coefficients (ICC). RESULTS: Comparisons of R2 with R3 methods showed greater agreement, mean differences ≤0.50 mm for all landmarks, than comparisons of R1 with R2, and R1 with R3, mean differences >0.50 mm for most of the y and z coordinates (P < .05). The R1 and R3 methods presented excellent intra and interobserver reproducibility and R2 method had moderate interobserver reproducibility. CONCLUSIONS: Longitudinal assessments of open bite treatment using digital dental models could consider the posterior teeth and/or the palate as references. The R1 and R3 methods showed adequate reproducibility and yield different quantitative results. The choice will depend on the posterior teeth changes and dental models' characteristics.

Temporomandibular Joint Osteoarthritis Diagnosis Using Privileged Learning of Protein Markers.

Diagnosis of temporomandibular joint (TMJ) Osteoarthritis (OA) before serious degradation of cartilage and subchondral bone occurs can help prevent chronic pain and disability. Clinical, radiomic, and protein markers collected from TMJ OA patients have been shown to be predictive of OA onset. Since protein data can often be unavailable for clinical diagnosis, we harnessed the learning using privileged information (LUPI) paradigm to make use of protein markers only during classifier training. Three different LUPI algorithms are compared with traditional machine learning models on a dataset extracted from 92 unique OA patients and controls. The best classifier performance of 0.80 AUC and 75.6 accuracy was obtained from the KRVFL+ model using privileged protein features. Results show that LUPI-based algorithms using privileged protein data can improve final diagnostic performance of TMJ OA classifiers without needing protein microarray data during classifier diagnosis.

Automatic Segmentation of Dental Root Canal and Merging with Crown Shape.

In this paper, machine learning approaches are proposed to support dental researchers and clinicians to study the shape and position of dental crowns and roots, by implementing a Patient Specific Classification and Prediction tool that includes RootCanalSeg and DentalModelSeg algorithms and then merges the output of these tools for intraoral scanning and volumetric dental imaging. RootCanalSeg combines image processing and machine learning approaches to automatically segment the root canals of the lower and upper jaws from large datasets, providing clinical information on tooth long axis for orthodontics, endodontics, prosthodontic and restorative dentistry procedures. DentalModelSeg includes segmenting the teeth from the crown shape to provide clinical information on each individual tooth. The merging algorithm then allows users to integrate dental models for quantitative assessments. Precision in dentistry has been mainly driven by dental crown surface characteristics, but information on tooth root morphology and position is important for successful root canal preparation, pulp regeneration, planning of orthodontic movement, restorative and implant dentistry. In this paper we propose a patient specific classification and prediction of dental root canal and crown shape analysis workflow that employs image processing and machine learning methods to analyze crown surfaces, obtained by intraoral scanners, and three-dimensional volumetric images of the jaws and teeth root canals, obtained by cone beam computed tomography (CBCT).

Automatic Segmentation of Mandibular Ramus and Condyles.

In order to diagnose TMJ pathologies, we developed and tested a novel algorithm, MandSeg, that combines image processing and machine learning approaches for automatically segmenting the mandibular condyles and ramus. A deep neural network based on the U-Net architecture was trained for this task, using 109 cone-beam computed tomography (CBCT) scans. The ground truth label maps were manually segmented by clinicians. The U-Net takes 2D slices extracted from the 3D volumetric images. All the 3D scans were cropped depending on their size in order to keep only the mandibular region of interest. The same anatomic cropping region was used for every scan in the dataset. The scans were acquired at different centers with different resolutions. Therefore, we resized all scans to 512×512 in the pre-processing step where we also performed contrast adjustment as the original scans had low contrast. After the pre-processing, around 350 slices were extracted from each scan, and used to train the U-Net model. For the cross-validation, the dataset was divided into 10 folds. The training was performed with 60 epochs, a batch size of 8 and a learning rate of 2×10-5. The average performance of the models on the test set presented 0.95 ± 0.05 AUC, 0.93 ± 0.06 sensitivity, 0.9998 ± 0.0001 specificity, 0.9996 ± 0.0003 accuracy, and 0.91 ± 0.03 F1 score. This study findings suggest that fast and efficient CBCT image segmentation of the mandibular condyles and ramus from different clinical data sets and centers can be analyzed effectively. Future studies can now extract radiomic and imaging features as potentially relevant objective diagnostic criteria for TMJ pathologies, such as osteoarthritis (OA). The proposed segmentation will allow large datasets to be analyzed more efficiently for disease classification.