On February 20, 2020, the incidence of Covid-19 began to rapidly escalate in Italy. By March 25, Italy had the second highest number of Covid-19 infections worldwide and the greatest number of deaths.¹ Children younger than 18 years of age who had Covid-19 composed only 1% of the total number of patients; 11% of these children were hospitalized, and none died.² The Coronavirus Infection in Pediatric Emergency Departments (CONFIDENCE) study involved a cohort of 100 Italian children younger than 18 years of age with Covid-19 confirmed by reverse-transcriptase–polymerase-chain-reaction testing of nasal or nasopharyngeal swabs who were assessed between March 3 and March 27 in 17 pediatric emergency departments. Here, we describe the results of the CONFIDENCE study and compare them with those from three cohorts in previously published analyses.^3-5

Progressive functional decline in the epilepsies is largely unexplained. We formed the ENIGMA-Epilepsy consortium to understand factors that influence brain measures in epilepsy, pooling data from 24 research centres in 14 countries across Europe, North and South America, Asia, and Australia. Structural brain measures were extracted from MRI brain scans across 2149 individuals with epilepsy, divided into four epilepsy subgroups including idiopathic generalized epilepsies (n =367), mesial temporal lobe epilepsies with hippocampal sclerosis (MTLE; left, n = 415; right, n = 339), and all other epilepsies in aggregate (n = 1026), and compared to 1727 matched healthy controls. We ranked brain structures in order of greatest differences between patients and controls, by meta-analysing effect sizes across 16 subcortical and 68 cortical brain regions. We also tested effects of duration of disease, age at onset, and age-by-diagnosis interactions on structural measures. We observed widespread patterns of altered subcortical volume and reduced cortical grey matter thickness. Compared to controls, all epilepsy groups showed lower volume in the right thalamus (Cohen’s d = −0.24 to −0.73; P < 1.49 × 10⁻⁴), and lower thickness in the precentral gyri bilaterally (d = −0.34 to −0.52; P < 4.31 × 10⁻⁶). Both MTLE subgroups showed profound volume reduction in the ipsilateral hippocampus (d = −1.73 to −1.91, P < 1.4 × 10⁻¹⁹), and lower thickness in extrahippocampal cortical regions, including the precentral and paracentral gyri, compared to controls (d = −0.36 to −0.52; P < 1.49 × 10⁻⁴). Thickness differences of the ipsilateral temporopolar, parahippocampal, entorhinal, and fusiform gyri, contralateral pars triangularis, and bilateral precuneus, superior frontal and caudal middle frontal gyri were observed in left, but not right, MTLE (d = −0.29 to −0.54; P < 1.49 × 10⁻⁴). Contrastingly, thickness differences of the ipsilateral pars opercularis, and contralateral transverse temporal gyrus, were observed in right, but not left, MTLE (d = −0.27 to −0.51; P < 1.49 × 10⁻⁴). Lower subcortical volume and cortical thickness associated with a longer duration of epilepsy in the all-epilepsies, all-other-epilepsies, and right MTLE groups (beta, b < −0.0018; P < 1.49 × 10⁻⁴). In the largest neuroimaging study of epilepsy to date, we provide information on the common epilepsies that could not be realistically acquired in any other way. Our study provides a robust ranking of brain measures that can be further targeted for study in genetic and neuropathological studies. This worldwide initiative identifies patterns of shared grey matter reduction across epilepsy syndromes, and distinctive abnormalities between epilepsy syndromes, which inform our understanding of epilepsy as a network disorder, and indicate that certain epilepsy syndromes involve more widespread structural compromise than previously assumed.

One outstanding challenge for machine learning in diagnostic biomedical imaging is algorithm interpretability. A key application is the identification of subtle epileptogenic focal cortical dysplasias (FCDs) from structural MRI. FCDs are difficult to visualize on structural MRI but are often amenable to surgical resection. We aimed to develop an open-source, interpretable, surface-based machine-learning algorithm to automatically identify FCDs on heterogeneous structural MRI data from epilepsy surgery centres worldwide. The Multi-centre Epilepsy Lesion Detection (MELD) Project collated and harmonized a retrospective MRI cohort of 1015 participants, 618 patients with focal FCD-related epilepsy and 397 controls, from 22 epilepsy centres worldwide. We created a neural network for FCD detection based on 33 surface-based features. The network was trained and cross-validated on 50% of the total cohort and tested on the remaining 50% as well as on 2 independent test sites. Multidimensional feature analysis and integrated gradient saliencies were used to interrogate network performance. Our pipeline outputs individual patient reports, which identify the location of predicted lesions, alongside their imaging features and relative saliency to the classifier. On a restricted ‘gold-standard’ subcohort of seizure-free patients with FCD type IIB who had T₁ and fluid-attenuated inversion recovery MRI data, the MELD FCD surface-based algorithm had a sensitivity of 85%. Across the entire withheld test cohort the sensitivity was 59% and specificity was 54%. After including a border zone around lesions, to account for uncertainty around the borders of manually delineated lesion masks, the sensitivity was 67%. This multicentre, multinational study with open access protocols and code has developed a robust and interpretable machine-learning algorithm for automated detection of focal cortical dysplasias, giving physicians greater confidence in the identification of subtle MRI lesions in individuals with epilepsy.

Detailed data on clinical presentations and outcomes of children with COVID-19 in Europe are still lacking. In this descriptive study, we report on 130 children with confirmed COVID-19 diagnosed by 28 centers (mostly hospitals), in 10 regions in Italy, during the first months of the pandemic. Among these, 67 (51.5%) had a relative with COVID-19 while 34 (26.2%) had comorbidities, with the most frequent being respiratory, cardiac, or neuromuscular chronic diseases. Overall, 98 (75.4%) had an asymptomatic or mild disease, 11 (8.5%) had moderate disease, 11 (8.5%) had a severe disease, and 9 (6.9%) had a critical presentation with infants below 6 months having significantly increased risk of critical disease severity (OR 5.6, 95% CI 1.3 to 29.1). Seventy-five (57.7%) children were hospitalized, 15 (11.5%) needed some respiratory support, and nine (6.9%) were treated in an intensive care unit. All recovered.Conclusion:This descriptive case series of children with COVID-19, mostly encompassing of cases enrolled at hospital level, suggest that COVID-19 may have a non-negligible rate of severe presentations in selected pediatric populations with a relatively high rates of comorbidities. More studies are needed to further understand the presentation and outcomes of children with COVID-19 in children with special needs. What is Known:• There is limited evidence on the clinical presentation and outcomes of children with COVID-19 in Europe, and almost no evidence on characteristics and risk factors of severe cases. What is New:• Among a case series of 130 children, mostly diagnosed at hospital level, and with a relatively high rate (26.2%) of comorbidities, about three-quarter had an asymptomatic or mild disease.• However, 57.7% were hospitalized, 11.5% needed some respiratory support, and 6.9% were treated in an intensive care unit.

Background and Objective To test the hypothesis that a multicenter-validated computer deep learning algorithm detects MRI-negative focal cortical dysplasia (FCD). Methods We used clinically acquired 3-dimensional (3D) T1-weighted and 3D fluid-attenuated inversion recovery MRI of 148 patients (median age 23 years [range 2–55 years]; 47% female) with histologically verified FCD at 9 centers to train a deep convolutional neural network (CNN) classifier. Images were initially deemed MRI-negative in 51% of patients, in whom intracranial EEG determined the focus. For risk stratification, the CNN incorporated bayesian uncertainty estimation as a measure of confidence. To evaluate performance, detection maps were compared to expert FCD manual labels. Sensitivity was tested in an independent cohort of 23 cases with FCD (13 ± 10 years). Applying the algorithm to 42 healthy controls and 89 controls with temporal lobe epilepsy disease tested specificity. Results Overall sensitivity was 93% (137 of 148 FCD detected) using a leave-one-site-out cross-validation, with an average of 6 false positives per patient. Sensitivity in MRI-negative FCD was 85%. In 73% of patients, the FCD was among the clusters with the highest confidence; in half, it ranked the highest. Sensitivity in the independent cohort was 83% (19 of 23; average of 5 false positives per patient). Specificity was 89% in healthy and disease controls. Discussion This first multicenter-validated deep learning detection algorithm yields the highest sensitivity to date in MRI-negative FCD. By pairing predictions with risk stratification, this classifier may assist clinicians in adjusting hypotheses relative to other tests, increasing diagnostic confidence. Moreover, generalizability across age and MRI hardware makes this approach ideal for presurgical evaluation of MRI-negative epilepsy. Classification of Evidence This study provides Class III evidence that deep learning on multimodal MRI accurately identifies FCD in patients with epilepsy initially diagnosed as MRI negative.

Epilepsy is associated with genetic risk factors and cortico-subcortical network alterations, but associations between neurobiological mechanisms and macroscale connectomics remain unclear. This multisite ENIGMA-Epilepsy study examined whole-brain structural covariance networks in patients with epilepsy and related findings to postmortem epilepsy risk gene expression patterns. Brain network analysis included 578 adults with temporal lobe epilepsy (TLE), 288 adults with idiopathic generalized epilepsy (IGE), and 1328 healthy controls from 18 centres worldwide. Graph theoretical analysis of structural covariance networks revealed increased clustering and path length in orbitofrontal and temporal regions in TLE, suggesting a shift towards network regularization. Conversely, people with IGE showed decreased clustering and path length in fronto-temporo-parietal cortices, indicating a random network configuration. Syndrome-specific topological alterations reflected expression patterns of risk genes for hippocampal sclerosis in TLE and for generalized epilepsy in IGE. These imaging-transcriptomic signatures could potentially guide diagnosis or tailor therapeutic approaches to specific epilepsy syndromes.