Search

Cerebellofaciodental syndrome is characterized by facial dysmorphisms, intellectual disability, cerebellar hypoplasia, and dental anomalies. It is an autosomal-recessive condition described in 2015 caused by pathogenic variants in BRF1. Here, we report a Brazilian patient who faced a diagnostic challenge beginning at 11 months of age. Fortunately, whole-exome sequencing (WES) was performed, detecting the BRF1 variants NM_001519.3:c.1649delG:p.(Gly550Alafs*36) and c.421C>T:p.(Arg141Cys) in compound heterozygosity, thus finally achieving a diagnosis of cerebellofaciodental syndrome. The patient is currently 25 years old and is the oldest patient yet reported. The clinical report and a review of published cases are presented. Atlanto-occipital fusion, a reduced foramen magnum and basilar invagination leading to compression of the medulla-spinal cord transition are skeletal findings not reported in previous cases. The description of syndromes with dental findings shows that such anomalies can be an important clue to relevant differential diagnoses. The cooperation of groups from different international centers made possible the resolution of this and other cases and is one of the strategies to bring medical advances to developing countries, where many patients with rare diseases are difficult to diagnose definitively.

Radiologists have been at the forefront of the digitization process in medicine. Artificial intelligence (AI) is a promising area of innovation, particularly in medical imaging. The number of applications of AI in neuroradiology has also grown. This article illustrates some of these applications. This article reviews machine learning challenges related to neuroradiology. The first approval of reimbursement for an AI algorithm by the Centers for Medicare and Medicaid Services, covering a stroke software for early detection of large vessel occlusion, is also discussed.

Myelomeningocele (MMC) is often related to hydrocephalus and Chiari malformation (CM) type 2; however, other brain abnormalities have been reported in this population. In order to better understand and quantify other forebrain abnormalities, we analyzed magnetic resonance imaging (MRI) of MMC patients treated in utero or postnatal. Between January 2014 and March 2017, 59 MMC were treated in our hospital. Thirty-seven patients (32 postnatal and 5 intrautero repair) had brain MRI and were enrolled at the study. MRI was analyzed by two experienced neuroradiologists to identify the supra and infratentorial brain abnormalities. A wide range of brain abnormalities was consistently identified in MMC patients. As expected, the most common were hydrocephalus (94.5%) and CM type II (89.1%). Of note, we found high incidence of corpus callosum abnormalities (86.4%), mostly represented by dysplasia (46%). The data are consistent with the concept that brain abnormalities related to MMC can be both infratentorial and supratentorial, cortical, and subcortical. More studies are needed to correlate these forebrain abnormalities to long-term functional outcome and their prognostic value for these patients.

Summary Background The CNS manifestations of COVID-19 in children have primarily been described in case reports, which limit the ability to appreciate the full spectrum of the disease in paediatric patients. We aimed to identify enough cases that could be evaluated in aggregate to better understand the neuroimaging manifestations of COVID-19 in the paediatric population. Methods An international call for cases of children with encephalopathy related to severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection and abnormal neuroimaging findings was made. Clinical history and associated plasma and cerebrospinal fluid data were requested. These data were reviewed by a central neuroradiology panel, a child neurologist, and a paediatric infectious diseases expert. The children were categorised on the basis of their time of probable exposure to SARS-CoV-2. In addition, cases were excluded when a direct link to SARS-CoV-2 infection could not be established or an established alternate diagnostic cause could be hypothesised. The accepted referral centre imaging data, from ten countries, were remotely reviewed by a central panel of five paediatric neuroradiologists and a consensus opinion obtained on the imaging findings. Findings 38 children with neurological disease related to SARS-CoV-2 infection were identified from France (n=13), the UK (n=8), the USA (n=5), Brazil (n=4), Argentina (n=4), India (n=2), Peru (n=1), and Saudi Arabia (n=1). Recurring patterns of disease were identified, with neuroimaging abnormalities ranging from mild to severe. The most common imaging patterns were postinfectious immune-mediated acute disseminated encephalomyelitis-like changes of the brain (16 patients), myelitis (eight patients), and neural enhancement (13 patients). Cranial nerve enhancement could occur in the absence of corresponding neurological symptoms. Splenial lesions (seven patients) and myositis (four patients) were predominantly observed in children with multisystem inflammatory syndrome. Cerebrovascular complications in children were less common than in adults. Significant pre-existing conditions were absent and most children had favourable outcomes. However, fatal atypical CNS co-infections developed in four previously healthy children infected with SARS-CoV-2. Interpretation Acute-phase and delayed-phase SARS-CoV-2-related CNS abnormalities are seen in children. Recurring patterns of disease and atypical neuroimaging manifestations can be found and should be recognised being as potentially due to SARS-CoV-2 infection as an underlying aetiological factor. Studies of paediatric specific cohorts are needed to better understand the effects of SARS-CoV-2 infection on the CNS at presentation and on long-term follow-up in children. Funding American Society of Pediatric Neuroradiology, University of Manchester (Manchester, UK). Video Abstract Neuroimaging manifestations in children with SARS-CoV-2 infection

PurposeTo improve and test the generalizability of a deep learning-based model for assessment of COVID-19 lung disease severity on chest radiographs (CXRs) from different patient populations.Materials and MethodsA published convolutional Siamese neural network-based model previously trained on hospitalized patients with COVID-19 was tuned using 250 outpatient CXRs. This model produces a quantitative measure of COVID-19 lung disease severity (pulmonary x-ray severity (PXS) score). The model was evaluated on CXRs from four test sets, including 3 from the United States (patients hospitalized at an academic medical center (N=154), patients hospitalized at a community hospital (N=113), and outpatients (N=108)) and 1 from Brazil (patients at an academic medical center emergency department (N=303)). Radiologists from both countries independently assigned reference standard CXR severity scores, which were correlated with the PXS scores as a measure of model performance (Pearson r). The Uniform Manifold Approximation and Projection (UMAP) technique was used to visualize the neural network results.ResultsTuning the deep learning model with outpatient data improved model performance in two United States hospitalized patient datasets (r=0.88 and r=0.90, compared to baseline r=0.86). Model performance was similar, though slightly lower, when tested on the United States outpatient and Brazil emergency department datasets (r=0.86 and r=0.85, respectively). UMAP showed that the model learned disease severity information that generalized across test sets.ConclusionsPerformance of a deep learning-based model that extracts a COVID-19 severity score on CXRs improved using training data from a different patient cohort (outpatient versus hospitalized) and generalized across multiple populations.

PURPOSE: To develop a deep learning model for detecting brain abnormalities on MR images. MATERIALS AND METHODS: In this retrospective study, a deep learning approach using T2-weighted fluid-attenuated inversion recovery images was developed to classify brain MRI findings as “likely normal” or “likely abnormal.” A convolutional neural network model was trained on a large, heterogeneous dataset collected from two different continents and covering a broad panel of pathologic conditions, including neoplasms, hemorrhages, infarcts, and others. Three datasets were used. Dataset A consisted of 2839 patients, dataset B consisted of 6442 patients, and dataset C consisted of 1489 patients and was only used for testing. Datasets A and B were split into training, validation, and test sets. A total of three models were trained: model A (using only dataset A), model B (using only dataset B), and model A + B (using training datasets from A and B). All three models were tested on subsets from dataset A, dataset B, and dataset C separately. The evaluation was performed by using annotations based on the images, as well as labels based on the radiology reports. RESULTS: Model A trained on dataset A from one institution and tested on dataset C from another institution reached an F1 score of 0.72 (95% CI: 0.70, 0.74) and an area under the receiver operating characteristic curve of 0.78 (95% CI: 0.75, 0.80) when compared with findings from the radiology reports. CONCLUSION: The model shows relatively good performance for differentiating between likely normal and likely abnormal brain examination findings by using data from different institutions. Keywords: MR-Imaging, Head/Neck, Computer Applications-General (Informatics), Convolutional Neural Network (CNN), Deep Learning Algorithms, Machine Learning Algorithms © RSNA, 2021 Supplemental material is available for this article.