The Castang Foundation, Bath Unit for Research in Paediatrics, National Institute of Health Research, the Royal United Hospitals Bath NHS Foundation Trust, the BRONNER-BENDUNG Stifung/Gernsbach, and University Children's Hospital Zurich.

The Castang Foundation, Bath Unit for Research in Paediatrics, National Institute of Health Research, the Royal United Hospitals Bath NHS Foundation Trust, BRONNER-BENDER Stiftung/Gernsbach, University Children's Hospital Zurich.

No abstract

Objective:Changing trends in multiple sclerosis (MS) epidemiology may first be apparent in the childhood population affected with first onset acquired demyelinating syndromes (ADSs). We aimed to determine the incidence, clinical, investigative and magnetic resonance imaging (MRI) features of childhood central nervous system ADSs in the British Isles for the first time.Methods:We conducted a population active surveillance study. All paediatricians, and ophthalmologists (n = 4095) were sent monthly reporting cards (September 2009–September 2010). International Paediatric MS Study Group 2007 definitions and McDonald 2010 MS imaging criteria were used for acute disseminated encephalomyelitis (ADEM), clinically isolated syndrome (CIS) and neuromyelitis optica (NMO). Clinicians completed a standard questionnaire and provided an MRI copy for review.Results:Card return rates were 90%, with information available for 200/222 positive notifications (90%). After exclusion of cases, 125 remained (age range 1.3–15.9), with CIS in 66.4%, ADEM in 32.0% and NMO in 1.6%. The female-to-male ratio in children older than 10 years (n = 63) was 1.52:1 (p = 0.045). The incidence of first onset ADS in children aged 1–15 years old was 9.83 per million children per year (95% confidence interval [CI] 8.18–11.71). A trend towards higher incidence rates of ADS in children of South Asian and Black ethnicity was observed compared with White children. Importantly, a number of MRI characteristics distinguished ADEM from CIS cases. Of CIS cases with contrast imaging, 26% fulfilled McDonald 2010 MS diagnostic criteria.Conclusions:We report the highest surveillance incidence rates of childhood ADS. Paediatric MS diagnosis at first ADS presentation has implications for clinical practice and clinical trial design.

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.