A fully automated classification of third molar development stages using deep learning.

Accurate classification of tooth development stages from orthopantomograms (OPG) is crucial for dental diagnosis, treatment planning, age assessment, and forensic applications. This study aims to develop an automated method for classifying third molar development stages using OPGs. Initially, our data consisted of 3422 OPG images, each classified and curated by expert evaluators. The dataset includes images from both Q3 (lower jaw left side) and Q4 (lower right side) regions extracted from panoramic images, resulting in a total of 6624 images for analysis. Following data collection, the methodology employs region of interest extraction, pre-filtering, and extensive data augmentation techniques to enhance classification accuracy. The deep neural network model, including architectures such as EfficientNet, EfficientNetV2, MobileNet Large, MobileNet Small, ResNet18, and ShuffleNet, is optimized for this task. Our findings indicate that EfficientNet achieved the highest classification accuracy at 83.7%. Other architectures achieved accuracies ranging from 71.57 to 82.03%. The variation in performance across architectures highlights the influence of model complexity and task-specific features on classification accuracy. This research introduces a novel machine learning model designed to accurately estimate the development stages of lower wisdom teeth in OPG images, contributing to the fields of dental diagnostics and treatment planning.

Interdisciplinary Role of Orthodontist in Screening and Managing Obstructive Sleep Apnea in Children and Adults.

Obstructive sleep apnea (OSA) can affect children and adults, and, if left untreated, could have a major impact on the general and overall well-being of the patient. Dental health care providers and orthodontists have an interdisciplinary role in screening patients at risk for OSA and make a referral to establish a definitive diagnosis by a sleep physician. The gold standard of diagnostic testing is polysomnography. The adeno-tonsillar hypertrophy is the primary cause of sleep apnea in children; therefore, adeno-tonsillectomy must be the first line of treatment. Post adeno-tonsillectomy, if there is residual OSA due to underlying skeletal discrepancy, the patient may be referred to an orthodontist for appropriate management. Currently the evidence in the literature for prophylactic growth modification in children to prevent OSA is weak. In adults, the gold standard for managing OSA is Positive Airway Pressure (PAP) therapy; however, adherence to this treatment is rather low. The oral appliance (OA) therapy is an alternate for PAP intolerant patients and for mild to moderate OSA patients. The OA therapy has to be administered by a qualified dentist or orthodontist after careful examination of dental and periodontal health as well as any pre-existing joint conditions. The OA therapy could cause OA-associated malocclusion and patients have to be made aware of prior to initiating treatment. In patients with severe OSA, surgical maxilla-mandibular advancement (MMA) is highly effective.

Recruitment strategies and retention rates for five National Dental PBRN studies.

Background: We describe a retrospective assessment of practitioner and patient recruitment strategies, patient retention strategies, and rates for five clinical studies conducted in the National Dental Practice-Based Research Network between 2012 and 2019, and practitioner and patient characteristics associated with retention. Methods: Similar recruitment strategies were adopted in the studies. The characteristics of the practitioners and patients are described. The proportion of patients who either attended a follow-up (FU) assessment or completed an online assessment was calculated. For studies with multiple FU visits or questionnaire assessments, rates for completing each FU were calculated, as were the rates for completing any and for completing all FU assessments. The associations of practitioner and patient characteristics with all clinic FU visits, and with the completion of all assessments for a study were ascertained. Results: Overall, 591 practitioners and 12,159 patients were included. FU rates by patients for any assessment varied from 91% to 96.5%, and rates for participating in all assessments ranged from 68% to 87%. The mean total number of patients each practitioner recruited was 21 (sd = 15); the mean number per study was 13 (sd = 7). For practitioners, practice type and patient enrollment were associated with greater clinic retention, while only race was associated with their patients completing post-visit online assessments. For patients, age was associated with clinic retention, while female gender, age, race, and education were all associated with greater completion of post-visit online assessments. Conclusion: The Network efficiently recruited practitioners and patients and achieved high patient retention rates for the five studies.

Clinical evaluation of Artificial Intelligence Driven Remote Monitoring technology for assessment of patient oral hygiene during orthodontic treatment.

INTRODUCTION: This study aimed to clinically evaluate the accuracy of Dental Monitoring's (DM) artificial intelligence (AI) image analysis and oral hygiene notification algorithm in identifying oral hygiene and mucogingival conditions. METHODS: Twenty-four patients seeking orthodontic therapy were monitored by DM oral hygiene protocol during their orthodontic treatment. During the bonding appointment and at each of 10 subsequent adjustment visits, a total of 232 clinical oral examinations were performed to assess the presence of the 3 oral hygiene parameters that DM monitors. In each clinical timepoint, the subjects took an oral DM scan and received a notification regarding their current oral status at that moment in time. Sensitivity, specificity, positive predictive value, and negative predictive value were calculated to evaluate AI and clinical assessment of plaque, gingivitis, and recession. RESULTS: A total of 232 clinical time points have been evaluated clinically and by the DM AI algorithm. For DM's AI detection of plaque and calculus, gingivitis, and recession, the sensitivity was 0.53, 0.35, and 0.22; the specificity was 0.94, 0.96, and 0.99; and the accuracy was 0.60, 0.49, and 0.72, respectively. CONCLUSIONS: DM's oral hygiene notification algorithm has low sensitivity, high specificity, and moderate accuracy. This indicates a tendency of DM to underreport the presence of plaque, gingivitis, and recession.

Pragmatic Return to Effective Dental Infection Control through Triage and Testing (PREDICT): an observational, feasibility study to improve dental office safety.

BACKGROUND: During the COVID-19 pandemic, there was a substantial interruption of care, with patients and workers fearful to return to the dental office. As dental practice creates a highly aerosolized environment, the potential for spread of airborne illness is magnified. As a means to increase safety and mitigate risk, pre-visit testing for SARS-CoV-2 has the potential to minimize disease transmission in dental offices. The Pragmatic Return to Effective Dental Infection Control through Testing (PREDICT) Feasibility Study examined the logistics and impact of two different testing mechanisms (laboratory-based PCR viral testing and point-of-care antigen testing) in dental offices. METHODS: Dental healthcare workers (DHCWs) and patients in four dental offices within the National Dental Practice-based Research Network participated in this prospective study. In addition to electronic surveys, participants in two offices completed POC testing, while participants in two offices used lab-based PCR methods to detect SARS-CoV-2 infection. Analysis was limited to descriptive measures, with median and interquartile ranges reported for Likert scale responses and mean and standard deviation for continuous variables. RESULTS: Of the total 72 enrolled, 28 DHCWs and 41 patients completed the protocol. Two patients (4.9%) tested positive prior to their visit, while 2 DHCWs (12.5%) tested positive for SARS-CoV-2 infection at the start of the study. DHCWs and patients shared similar degree of concern (69% and 63%, respectively) for contracting COVID-19 from patients, while patients feared contracting COVID-19 from DHCWs less (49%). Descriptive statistics calculations revealed that saliva, tongue epithelial cells, and nasal swabs were the most desirable specimen collection method; both testing (LAB and POC) protocols took similar amounts of total time to complete; and DHCWs and patients reported feeling more comfortable when both groups were tested. CONCLUSIONS: While a larger-scale, network study is necessary for generalizability of results, this feasibility study suggests that SARS-CoV-2 testing can be effectively implemented into dental practice workflows and positively impact perception of safety for DHCWs and patients. As new virulent infectious diseases emerge, preparing dental personnel to employ an entire toolbox of risk mitigation strategies, including testing, may have the potential to decrease dental practice closure time, maintaining continuity of dental care services for patients. TRIAL REGISTRATION: ClinicalTrials.gov: NCT05123742.

Content analysis of AI-generated (ChatGPT) responses concerning orthodontic clear aligners.

OBJECTIVES: To assess the accuracy of ChatGPT answers concerning orthodontic clear aligners. MATERIALS AND METHODS: A cross-sectional content analysis of ChatGPT generated responses to queries related to clear aligner treatment (CAT) was undertaken. A total of 111 questions were generated by three orthodontists based on a set of predefined domains and subdomains. The artificial intelligence (AI)-generated (ChatGPT) answers were extracted and their accuracy was determined independently by five orthodontists. The accuracy of answers was assessed using a prepiloted four-point scale scoring rubric. Descriptive statistics were performed. RESULTS: The total mean accuracy score for the entire set was 2.6 ± 1.1. It was noted that 58% of the AI-generated answers were scored as objectively true, 18% were selected facts, 9% were minimal facts, and 15% were false. False claims included the ability of CAT to reduce the need for orthognathic surgery (4.0 ± 0.0), improve airway function (3.8 ± 0.5), achieve root parallelism (3.6 ± 0.5), alleviate sleep apnea (3.8 ± 0.5), and produce more stable results compared to fixed appliances (3.8 ± 0.5). CONCLUSIONS: The overall level of accuracy of ChatGPT responses to questions concerning CAT was suboptimal and lacked citations to relevant literature. Ability of the software to offer current and precise information was limited. Therefore, clinicians and patients must be mindful of false claims and relevant facts omitted in the answers generated by ChatGPT.

Comparison of light transmittance and color changes between polyurethane and copolyester retainer materials after staining and destaining.

BACKGROUND: Retainers are the only effective approach to prevent orthodontic relapse. The aim of this study was to compare the changes in color and light-transmittance of rough and smooth thermoformed polyurethane and copolymer retainer samples after staining in different solutions and destaining with different approaches. METHODS: Four hundred copolyester (Essix® ACE) and 400 polyurethane (Zendura®) samples with different surface textures, smooth and rough, were stained in 4 different solutions (n = 100 per solution) over 28 days. Each of the four groups of 100 stained samples of each material was subdivided into 5 groups of 20 samples and subjected to different destaining solutions. Light transmittance and color changes were evaluated using a spectrometer and a spectrophotometer. Mean differences were compared using a two-way analysis of variance (ANOVA) and posthoc multiple comparison tests at P = 0.05. RESULTS: No significant differences in light transmittance were found between both untreated materials. Both materials were stained in a similar fashion and showed no significant differences between two materials after staining. Coffee and tea stained both materials more significantly than wine, but there was a significant difference of changes of color and light transmittance between rough and smooth surfaces during the destaining in coffee- and tea-stained samples of copolyester material. All destaining solutions were effective at removing all stains on the samples. The surface roughness of the material plays a significant role in the ability of the materials to be destained, demonstrating a more significant greater effect on cleaning rough samples for improvements in light-transmittance and greater changes in color. CONCLUSIONS: This study concluded that the surface of materials plays a significant role in the material destaining and staining. In addition, the different polymers used for retainer fabrication exhibited different responses during the destaining process depending on types of stains.

Long-term evaluation of soft-tissue changes after miniscrew-assisted and conventional rapid palatal expansion using voxel-based superimposition of cone-beam computed tomography scans.

INTRODUCTION: This study aimed to evaluate the soft-tissue changes in the long-term after miniscrew-assisted rapid palatal expansion (MARPE) and conventional rapid palatal expansion (RPE) appliances compared with a matched control group using voxel-based superimposition of cone-beam computed tomography (CBCT) scans. METHODS: A total of 180 CBCTs for 60 patients at 3-time points were evaluated: pretreatment (T1), postexpansion (T2), and posttreatment (T3) for 3 groups: (1) MARPE, (2) RPE, and (3) controls (time-period T1 to T3: MARPE, 2 years 8 months; RPE, 2 years 9 months; control, 2 years 7 months). The voxel-based superimposition technique was used to superimpose the CBCT scans, after which the soft-tissue surfaces were extracted from the superimposed T1-CBCT, T2-CBCT, and T3-CBCT scans. Nine landmarks were identified on the CBCT scans: nasion, A-point, pogonion, right and left alar base, right and left zygoma, and right and left gonion. The coordinates of the 9 parameters were obtained in the x-axis, y-axis, and z-axis for the CBCT scans and subjected to statistical analyses. The changes in the soft-tissue surfaces were also evaluated by color-coded maps for short-term (T2) and long-term (T3) changes. The mean changes from T1 to T2 and T1 to T3 were tested against no change within the groups by paired t test, and the mean changes among the 3 groups were compared with analysis of variance F test with Tukey's Honest significant difference used for adjusting P values for multiple testing. RESULTS: In the short term, both MARPE and RPE led to a significant downward movement of pogonion, left gonion, and lateral movement of the right and left alar base compared with controls at T2 (P <0.05). In addition, MARPE led to a significant downward movement of right gonion than controls at T2 (P <0.05). Moreover, RPE led to a significant downward and forward movement of A-point and downward movement of the right and left alar base than controls at T2 (P <0.05). However, in the long-term, there were no significant differences in the soft-tissue changes among the MARPE, RPE, and control groups. CONCLUSIONS: MARPE and RPE do not lead to significant soft-tissue changes in the long term when compared with controls.

A cone beam computed tomographic analysis of cervical vertebral nonsegmentation.

OBJECTIVE: Using cone beam computed tomography (CBCT), this study aimed to investigate the radiologic features of cervical vertebral nonsegmentation (CVN) in patients with no known syndromes or pathoses. STUDY DESIGN: In this retrospective study, we examined CBCT scans of patients with CVN for the following parameters: type of nonsegmentation (partial or complete); laterality of partial nonsegmentation; level of vertebrae affected in nonsegmentation; anatomic parts of the vertebrae involved; and the presence or absence of degenerative joint disease (DJD). RESULTS: From the structured reports of 13,458 CBCT scans, we found 110 CBCT scans (0.82%) with CVN. Of this total, 77.3% were partial and 22.7% were complete. Most were located at the level of the C2 and C3 vertebrae. The transverse process alone and the transverse process and body were most frequently involved in partial CVN, whereas most complete types occurred in the transverse process and body. Degenerative joint disease was present in 45.9% of partial CVN (clearly distinguishable from nonsegmentation in 36.5%) and 20% of complete CVN, with all cases clearly distinguishable. CONCLUSIONS: Cervical vertebral nonsegmentation has a low prevalence. It is mostly partial, occurs most commonly in C2-C3, and usually involves the transverse process and body. Identification of CVN on CBCT images is important because this condition can lead to DJD in older age and may increase the risk for muscle weakness, head and neck pain, limited movement, and neurologic complications.

Call for algorithmic fairness to mitigate amplification of racial biases in artificial intelligence models used in orthodontics and craniofacial health.

Machine Learning (ML), a subfield of Artificial Intelligence (AI), is being increasingly used in Orthodontics and craniofacial health for predicting clinical outcomes. Current ML/AI models are prone to accentuate racial disparities. The objective of this narrative review is to provide an overview of how AI/ML models perpetuate racial biases and how we can mitigate this situation. A narrative review of articles published in the medical literature on racial biases and the use of AI/ML models was undertaken. Current AI/ML models are built on homogenous clinical datasets that have a gross underrepresentation of historically disadvantages demographic groups, especially the ethno-racial minorities. The consequence of such AI/ML models is that they perform poorly when deployed on ethno-racial minorities thus further amplifying racial biases. Healthcare providers, policymakers, AI developers and all stakeholders should pay close attention to various steps in the pipeline of building AI/ML models and every effort must be made to establish algorithmic fairness to redress inequities.

Pragmatic Return to Effective Dental Infection Control through Triage and Testing (PREDICT): A feasibility study to improve dental office safety.

Background: The COVID-19 pandemic highlights the need for practitioners to enhance workflows to increase safety and mitigate risk. As dental practice creates a highly aerosolized environment, pre-visit testing for SARS-CoV-2 has the potential to be an effective mitigation strategy to minimize disease transmission in dental offices. The Pragmatic Return to Effective Dental Infection Control through Testing (PREDICT) Feasibility Study examined the potential, logistics, and impact related to laboratory-based PCR viral testing and point-of-care (POC) antigen testing. Methods: Dental healthcare workers (DHCWs) and patients in four dental offices within the National Dental Practice-based Research Network participated in this prospective study. In addition to electronic surveys, participants in two offices completed POC testing, while participants in two offices used lab based PCR methods to detect SARS-CoV-2 infection. For this feasibility study, analysis was limited to descriptive measures. Median and interquartile ranges were reported for Likert scale responses and mean and standard deviation for continuous variables. Results: Forty-one of forty-three consented patients and twenty-eight of twenty-nine DHCWs completed the protocol. Descriptive statistics calculations including median and interquartile ranges revealed (1) saliva, tongue epithelial cells and nasal swabs were the most desirable specimens for testing for groups (2) both LAB and POC protocols took similar amounts of total time to complete the full testing protocol and (3) DHCWs and patients reported feeling more comfortable when both groups were tested. Conclusions: This feasibility study suggests that pre-visit SARS-CoV-2 testing can be effectively implemented into dental practice workflows and positively impact perception of safety for DHCWs and patients, though a larger scale, network study is necessary for generalizability of results. As new virulent infectious diseases continue to emerge, preparing dental personnel to employ an entire toolbox of risk mitigation strategies, including testing, may have the potential to decrease dental practice closure time, maintaining continuity of dental care services for patients. Trial registration: This trial was registered on ClinicalTrials.gov: NCT05123742.

Characteristics of dentists in the National Dental Practice-Based Research Network.

OBJECTIVES: Our aims are to describe the characteristics of dentists, members of the US National Dental practice-based research network (PBRN) in the United States, and determine how often these dentists provide specific dental procedures. METHODS: Dentists completed a questionnaire when they enrolled in the Network about their demographic and training characteristics and characteristics of their practices and patients. Dentists also reported the frequency of providing specific dental procedures. Data were analyzed using descriptive statistics. RESULTS: Of 4,483 dentists in active clinical practice, 34% identified as females, 70% as white, and 73% as general dentists. Most dentists practiced in large metropolitan areas (87%) and in solo or small practices (72%). On average, they reported about one-half of their patients were children or older adults, a third were from historically underrepresented racial and ethnic groups, and one-quarter were covered by public insurance. Most dentists routinely performed restorations and fixed prosthetics (78%), extractions (59%), removable (44%) and implant (40%) prosthetics, and endodontics on incisor and premolar teeth (44%). CONCLUSIONS: Dentists participating in the National Dental PBRN have much in common with dentists at large. The network has a broad representation of dentists, practice types, patient populations, and treatments offered, including diversity regarding race/ethnicity, gender, insurance, and geography of its practitioners and patients. CLINICAL SIGNIFICANCE: Characteristics of National Dental PBRN dentists suggest that a broad range of dentists is interested in participating in national-level research studies, thereby enabling an array of clinical study settings and topics that can optimize the generalizability of study findings.

The use of teleorthodontics during the COVID-19 pandemic and beyond - perspectives of patients and providers.

BACKGROUND: The COVID-19 pandemic significantly impacted dental services, resulting in reduced staff availability, limited appointments, and some dental clinics even being forced to close their doors. Despite these challenges, the need for dental consultants remained present, particularly in emergency situations. One area of orthodontics that had seen a surge in demand during the pandemic is Teleorthodontics. With the help of Teleorthodontics, orthodontic consultations, assessments, and even treatment monitoring could be conducted remotely, making it a safe and convenient option for patients during those challenging times. AIM: This survey aimed to evaluate the acceptance of patients and their orthodontists on the use of different modes of communication through Teleorthodontics during the COVID-19 pandemic and their willingness to continue using this in the future. METHODS: An online survey instrument in Qualtrics was distributed to orthodontic patients at the University of Illinois, Chicago. The survey was available on a rolling basis for up to 6 months. A total number of 364 partients voluntarily participated in the survey. The Faculty and Residents were also asked to participate in a survey through recruitment via their UIC email addresses. RESULTS: According to our survey, both patients and providers showed acceptance of Teleorthodontics and have used it in different forms during orthodontic treatment. The application is easy-to-use, convenient, and not at all time-consuming. Overall satisfaction with using this application was recorded at 92%, with 66% of patients stating that it saved them time by eliminating the need to travel to the orthodontic clinic. 30% of providers found that the interaction with patients using Teleorthodontics was a positive experience and would recommend it in future. CONCLUSION: Teleorthodontics has shown great potential, particularly in follow-up cases, and holds promise as a valuable tool for online remote dental consultations in the future.

In-vivo evaluation of Artificial Intelligence Driven Remote Monitoring technology for tracking tooth movement and reconstruction of 3-dimensional digital models during orthodontic treatment.

INTRODUCTION: An in-vivo evaluation of the Dental Monitoring (DM; Paris, France) Artificial Intelligence Driven Remote Monitoring technology was conducted in an active clinical setting. Our objectives were to compare the accuracy and validity of the 3-dimensional (3D) digital models remotely generated from the DM application to 3D Digital Models generated from the iTero Element 5D intraoral scanner (Align Technologies, San Jose, Calif) of patients' dentition during in-vivo fixed orthodontic treatment. METHODS: The orthodontic treatment of 24 patients (aged 14-55 years) was tracked across an average of 13.4 months. Scans of the maxillary and mandibular arches of each patient were taken by an iTero intraoral scanner and with the DM application before treatment initiation without (T0) and with (T1) the fixed orthodontic appliances and at every in-person adjustment appointment (T2-T10). The global deviation between the reconstructed digital models from the DM and iTero scans was compared at each time point using Geomagic Control-X 2020 (3D Systems, Rock Hill, SC). Descriptive analysis was conducted to determine the mean deviation at each time point for the maxillary and mandibular arches, to compare the maxilla and mandible mean deviations at each time point to the null hypothesis mean of 0 mm and the paired mean of the average at each time point between the maxilla and mandible. RESULTS: The findings revealed no clinically significant difference between the reconstructed digital models generated by the iTero IOS and the remotely reconstructed digital dental models generated by the DM application. CONCLUSION: DM artificial intelligence tracking algorithm can track tooth movement and reconstruct 3D digital models to a clinically acceptable degree for orthodontic application.

Effectiveness of AI-driven remote monitoring technology in improving oral hygiene during orthodontic treatment.

OBJECTIVE: This study aimed to evaluate the effectiveness of Dental Monitoring™ (DM™) Artificial Intelligence Driven Remote Monitoring Technology (AIDRM) technology in improving the patient's oral hygiene during orthodontic treatment through AI-based personalized active notifications. METHODS: A prospective clinical study was conducted on two groups of orthodontic patients. DM Group: (n = 24) monitored by DM weekly scans and received personalized notifications on the DM smartphone application regarding their oral hygiene status. Control Group (n = 25) not monitored by DM. Both groups were clinically assessed using Plaque Index (OPI) and the Modified Gingival Index (MGI). DM Group was followed for 13 months and the Control Group was followed for 5 months. Student-independent t test and paired t tests were used to investigate the mean differences between study groups and between time points for each group respectively. RESULTS: At all time points, the mean differences indicated that the DM group had lower OPI and MGI values than the control group. The mean value for OPI and MGI were statistically significantly lower in the DM group (OPI = 1.96, MGI = 1.56) than in the control group (OPI = 2.41, MGI = 2.17) after 5 months. A rapid increase in mean OPI and MGI values was found between T0 and T1 for both study groups. A plateau effect for OPI scores appeared to occur from T1 to T5 for both study groups, but the plateau effect seemed to be more pronounced for the DM group than the study group. The MGI values for both study groups also increased dramatically from baseline to T5, however, a plateau effect was not observed. CONCLUSIONS: The oral hygiene of orthodontic patients rapidly worsens over the first 3 months and plateaus after about 5 months of treatment. AIDRM by weekly DM scans and personalized active notifications may improve oral hygiene over time in orthodontic patients.

Blockchain technology and federated machine learning for collaborative initiatives in orthodontics and craniofacial health.

There is a paucity of largescale collaborative initiatives in orthodontics and craniofacial health. Such nationally representative projects would yield findings that are generalizable. The lack of large-scale collaborative initiatives in the field of orthodontics creates a deficiency in study outcomes that can be applied to the population at large. The objective of this study is to provide a narrative review of potential applications of blockchain technology and federated machine learning to improve collaborative care. We conducted a narrative review of articles published from 2018 to 2023 to provide a high level overview of blockchain technology, federated machine learning, remote monitoring, and genomics and how they can be leveraged together to establish a patient centered model of care. To strengthen the empirical framework for clinical decision making in healthcare, we suggest use of blockchain technology and integrating it with federated machine learning. There are several challenges to adoption of these technologies in the current healthcare ecosystem. Nevertheless, this may be an ideal time to explore how best we can integrate these technologies to deliver high quality personalized care. This article provides an overview of blockchain technology and federated machine learning and how they can be leveraged to initiate collaborative projects that will have the patient at the center of care.

AggregateNet: A deep learning model for automated classification of cervical vertebrae maturation stages.

OBJECTIVE: A study of supervised automated classification of the cervical vertebrae maturation (CVM) stages using deep learning (DL) network is presented. A parallel structured deep convolutional neural network (CNN) with a pre-processing layer that takes X-ray images and the age as the input is proposed. METHODS: A total of 1018 cephalometric radiographs were labelled and classified according to the CVM stages. The images were separated according to gender for better model-fitting. The images were cropped to extract the cervical vertebrae automatically using an object detector. The resulting images and the age inputs were used to train the proposed DL model: AggregateNet with a set of tunable directional edge enhancers. After the features of the images were extracted, the age input was concatenated to the output feature vector. To have the parallel network not overfit, data augmentation was used. The performance of our CNN model was compared with other DL models, ResNet20, Xception, MobileNetV2 and custom-designed CNN model with the directional filters. RESULTS: The proposed innovative model that uses a parallel structured network preceded with a pre-processing layer of edge enhancement filters achieved a validation accuracy of 82.35% in CVM stage classification on female subjects, 75.0% in CVM stage classification on male subjects, exceeding the accuracy achieved with the other DL models investigated. The effectiveness of the directional filters is reflected in the improved performance attained in the results. If AggregateNet is used without directional filters, the test accuracy decreases to 80.0% on female subjects and to 74.03% on male subjects. CONCLUSION: AggregateNet together with the tunable directional edge filters is observed to produce higher accuracy than the other models that we investigated in the fully automated determination of the CVM stages.

Short-term and long-term effects of miniscrew-assisted and conventional rapid palatal expansion on the cranial and circummaxillary sutures.

INTRODUCTION: The objective of this study was to analyze the short-term and long-term effects of miniscrew-assisted rapid palatal expansion (MARPE) and conventional rapid palatal expansion (RPE) appliances on cranial and circummaxillary sutures as compared with a matched control group. METHODS: One hundred and eighty cone-beam computed tomography scans for 60 subjects were evaluated for the 3 groups: (1) MARPE (n = 20; aged 13.7 ± 1.74 years), (2) RPE (n = 21; age 13.9 ± 1.14 years), and (3) control (n = 19; age 13.3 ± 1.49 years) at pretreatment (T1), postexpansion (T2), and posttreatment (T3) (T1 to T3: MARPE, 2 years 8 months; RPE, 2 years 9 months; control, 2 years 7 months). Frontonasal suture, frontomaxillary suture, zygomaticomaxillary suture, zygomaticofrontal suture, intermaxillary suture, pterygomaxillary suture, nasomaxillary suture, and zygomaticotemporal suture were measured on the right and left sides for all 3 time labels. In addition, midpalatal suture was measured at the incisor, canine, and molar levels. RESULTS: Within-group analysis showed that MARPE and RPE led to a significant increase in the widths of frontonasal, frontomaxillary, intermaxillary, nasomaxillary, and midpalatal suture at incisor, canine, and molar levels at T2 compared with T1. Between-group analysis showed that MARPE and RPE significantly increased the width of the intermaxillary and midpalatal suture at the incisor, canine, and molar compared with controls at T2. In the long term, between-group comparisons showed no significant difference among the 3 groups except that MARPE led to a significant increase in the width of midpalatal suture at incisor, canine, and molar levels compared with RPE and controls at T3. CONCLUSIONS: MARPE led to a significant increase in the width of the midpalatal suture at incisor, canine, and molar levels compared with RPE and controls in the long term. There was no difference in the width of other cranial and circummaxillary sutures among the 3 groups in the long term.

COVID-19 infection rates and mitigation strategies in orthodontic practices.

BACKGROUND: COVID-19 has impacted and increased risks for all populations, including orthodontic patients and providers. It also changes the practice management and infection control landscape in the practices. This study aimed to investigate the COVID-19 infection and vaccination status of orthodontic providers and mitigation approaches in orthodontic practices in the United States during 2021. METHODS: A validated 50-question research electronic data capture (REDCap) browser-based questionnaire was distributed to 12,393 orthodontists and pediatric dentists who reported actively providing orthodontic treatment. Questions were designed to collect demographic data of respondents, evaluate the COVID-19 mitigation approaches, and evaluate the history of COVID-19 infection and vaccination status of the orthodontic providers. Associations of demographic and the COVID-19 mitigation approaches were assessed using chi-square tests at the significance level of 0.05. RESULTS: Four hundred fifty-seven returned the survey (response rate 3.69%) for analysis. Most respondents were vaccinated, and increased infection control measures in response to the pandemic. Half of the respondents practiced teledentistry and switched to digital impression systems. Two-thirds reported difficulties in attaining PPEs due to the increased cost and scarcity of PPEs. About 6% of respondents reported a history of COVID-19 infection, and 68.9% of their staff had COVID-19 infection. Statistically significant associations were found between increased practice experience with difficulties in acquiring PPE (p = .010). There were no significant associations between races of respondents, geographic location, and years of practicing when cross-tabulated with vaccination status or COVID-19 infection rate (p > .05). CONCLUSION: Increased infection control strategies were employed in almost all orthodontic practices in addition to existing universal precaution. Most of the orthodontic providers and their staff members were vaccinated. While staff's infection rates were an issue, doctors' infection rates remained low.