Dental Anomaly Detection Using Intraoral Photos Via Deep Learning

Children with orofacial clefting (OFC) present with a wide range of dental anomalies. Identifying these anomalies is vital to understand their etiology and to discern the complex phenotypic spectrum of OFC. Such anomalies are currently identified using intra-oral exams by dentists, a costly and time-consuming process. We claim that automating the process of anomaly detection using deep neural networks (DNNs) could increase efficiency and provide reliable anomaly detection while potentially increasing the speed of research discovery. This study characterizes the use of` DNNs to identify dental anomalies by training a DNN model using intraoral photographs from the largest international cohort to date of children with nonsyndromic OFC and controls (OFC1). In this project, the intraoral images were submitted to a Convolutional Neural Network (CNN) model to perform multi-label multi-class classification of 10 dental anomalies. The network predicts whether an individual exhibits any of the 10 anomalies and is able to do so significantly faster than a human rater. For every anomaly except mammalons, F1 scores suggest that our model performs competitively at anomaly detection when compared to a dentist with 8 years of clinical experience. In addition, we use saliency maps to provide a post-hoc interpretation for our model’s predictions. This enables dentists to examine and verify our model’s predictions.

Orofacial clefts

Orofacial clefting is an umbrella term for common human malformations. The most common of these malformations result from failure of tissue fusion events during formation of the embryonic face. Single-gene causes of orofacial clefting are rare and often associated with other clinical features in the form of a recognizable syndromal diagnosis. Isolated orofacial clefting is most likely caused by gene–environment interactions but the exact nature of these interactions is not yet known

RERE deficiency contributes to the development of orofacial clefts in humans and mice

Deletions of chromosome 1p36 are the most common telomeric deletions in humans and are associated with an increased risk of orofacial clefting. Deletion/phenotype mapping, combined with data from human and mouse studies, suggests the existence of multiple 1p36 genes associated with orofacial clefting including SKI, PRDM16, PAX7 and GRHL3. The arginine–glutamic acid dipeptide (RE) repeats gene (RERE) is located in the proximal critical region for 1p36 deletion syndrome and encodes a nuclear receptor co-regulator. Pathogenic RERE variants have been shown to cause neurodevelopmental disorder with or without anomalies of the brain, eye or heart (NEDBEH). Cleft lip has previously been described in one individual with NEDBEH. Here we report the first individual with NEDBEH to have a cleft palate. We confirm that RERE is broadly expressed in the palate during mouse embryonic development, and we demonstrate that the majority of RERE-deficient mouse embryos on C57BL/6 background have cleft palate. We go on to show that ablation of Rere in cranial neural crest (CNC) cells, mediated by a Wnt1-Cre, leads to delayed elevation of the palatal shelves and cleft palate and that proliferation of mesenchymal cells in the palatal shelves is significantly reduced in Rereflox/flox; Wnt1-Cre embryos. We conclude that loss of RERE function contributes to the development of orofacial clefts in individuals with proximal 1p36 deletions and NEDBEH and that RERE expression in CNC cells and their derivatives is required for normal palatal development.

Adoption of Telemedicine for Multidisciplinary Care in Pediatric Otolaryngology

Objective: The COVID-19 pandemic has introduced a period of social isolation that has challenged the ability of providers to uphold in-person patient care. Although commonplace in pediatric otolaryngology, multidisciplinary clinics pose a unique challenge during this time due to increased infection risk from multiple patient-provider interactions. Guidance on the application of telemedicine for multidisciplinary clinics in pediatric otolaryngology is limited. Methods: We provide comprehensive guidance on best practices for conducting telemedicine visits for a number of multidisciplinary otolaryngology clinics using our experiences at a single tertiary care children’s hospital. A review of literature to support the adoption of telemedicine in multidisciplinary pediatric otolaryngology is also incorporated. Results: Telemedicine was successfully adopted for 7 multidisciplinary pediatric clinics with a variety of specialists: aerodigestive disorders, congenital hearing loss, microtia/aural atresia, orofacial clefting, sleep disorders, tracheostomy care, and velopharyngeal dysfunction. Conclusions: Telemedicine is feasible for a variety of multidisciplinary clinics and its optimization is critical for providing care to complex pediatric otolaryngology patients during the COVID-19 pandemic and beyond.

The Intersection of the Genetic Architectures of Orofacial Clefts and Normal Facial Variation

Unaffected relatives of individuals with non-syndromic cleft lip with or without cleft palate (NSCL/P) show distinctive facial features. The presence of this facial endophenotype is potentially an expression of underlying genetic susceptibility to NSCL/P in the larger unselected population. To explore this hypothesis, we first partitioned the face into 63 partially overlapping regions representing global-to-local facial morphology and then defined endophenotypic traits by contrasting the 3D facial images from 264 unaffected parents of individuals with NSCL/P versus 3,171 controls. We observed distinct facial features between parents and controls across 59 global-to-local facial segments at nominal significance (p ≤ 0.05) and 52 segments at Bonferroni corrected significance (p < 1.2 × 10–3), respectively. Next, we quantified these distinct facial features as univariate traits in another dataset of 8,246 unaffected European individuals and performed a genome-wide association study. We identified 29 independent genetic loci that were associated (p < 5 × 10–8) with at least one of the tested endophenotypic traits, and nine genetic loci also passed the study-wide threshold (p < 8.47 × 10–10). Of the 29 loci, 22 were in proximity of loci previously associated with normal facial variation, 18 were near genes that show strong evidence in orofacial clefting (OFC), and another 10 showed some evidence in OFC. Additionally, polygenic risk scores for NSCL/P showed associations with the endophenotypic traits. This study thus supports the hypothesis of a shared genetic architecture of normal facial development and OFC.

Parents of Children With Nonsyndromic Orofacial Clefting Show Altered Palate Shape

Objective: The unaffected relatives of individuals with nonsyndromic orofacial clefts have been shown to exhibit subtle craniofacial differences compared with the general population. Here, we investigate whether these morphological differences extend to the shape of the palate. Design: We conducted a geometric morphometric analysis to compare palate shape in the clinically unaffected parents of children with nonsyndromic cleft lip with or without cleft palate and adult controls of European, Asian, and African ancestry. We conducted pairwise group comparisons using canonical variates analysis, and then confirmed and characterized findings of shape differences using Euclidean distance matrix analysis. Results: Significant differences in palate shape were detected in unaffected mothers (but not fathers) compared to demographically matched controls. The differences in shape were ancestry-specific; mothers of Asian-derived and African-derived ancestry showed wider and shorter palates with higher posterior palatal vaults, while mothers of European-derived ancestry showed narrower palates with higher anterior palatal vaults. Conclusions: Our findings suggest that altered palate shape is a subclinical phenotypic feature, which may be indicative of elevated orofacial cleft risk. The risk phenotype varied by sex and ancestry, suggesting possible etiologic heterogeneity among demographic groups. Understanding the genetic basis of these informative palate shape traits may reveal new genes and pathways relevant to nonsyndromic orofacial clefting.

Cleft Lip and Palate in Ectodermal Dysplasia

Objective: Ectodermal dysplasia (ED) comprises multiple syndromes that affect skin, hair, nails, and teeth, and sometimes are associated with orofacial clefting. The purpose of this study is to (1) identify the prevalence and characteristics of cleft lip and/or palate (CL/P) in patients with ED and (2) describe the management and outcomes. Design: Retrospective review from 1990 to 2019. Patients: All patients with ED treated at Boston Children's Hospital. Main Outcomes Measures: Prevalence of CL/P was calculated and clinical details recorded: phenotypic anomalies, cleft type, operative treatment, and results of repair. Results: Of 170 patients with a purported diagnosis of ED, 24 (14%) had CL/P. Anatomic categories were bilateral CL/P (67%), unilateral CL/P (8%), and cleft palate only (25%). The most common ED syndrome (37%) was ectrodactyly, ectodermal dysplasia, and cleft lip/palate (EEC). Pathogenic variants in TP63 were the most frequent finding in the 11 patients who had genetic testing. Aberrations from a typical clinical course included failure of presurgical dentofacial orthopedics, dehiscence of nasolabial adhesion, and total palatal absence requiring free-flap construction. Two patients had prolonged postoperative admission for respiratory infection. High fistula (8%) and velopharyngeal insufficiency (33%) rates reflected the predominance of bilateral complete forms. Conclusions: As in other types of syndromic CL/P, cleft phenotypic expression in ED is more severe than the general cleft population. Further studies are needed to correlate genotype and phenotype for the distinct syndromes included in the ED spectrum.