More articles from ADULT BRAIN

MR imaging data and molecular information were retrospectively obtained from The Cancer Imaging Archives for 259 patients with either low- or high-grade gliomas. A convolutional neural network was trained to classify IDH1 mutation status, 1p/19q codeletion, and MGMT promotor methylation status. Classification had high accuracy: IDH1 mutation status, 94%; 1p/19q codeletion, 92%; and MGMT promotor methylation status, 83%. The authors conclude that this shows the feasibility of a deep-learning CNN approach for the accurate classification of individual genetic mutations of both low- and high-grade gliomas and that the relevant MR imaging features acquired from an added dimensionality-reduction technique are concordant with existing literature, showing that neural networks are capable of learning key imaging components without prior feature selection or human directed training.

Patients with basilar artery stenosis (n=126; 66 symptomatic and 60 asymptomatic) underwent high-resolution MR imaging. The relationship between imaging findings (intraplaque hemorrhage, contrast enhancement, degree of stenosis, minimal lumen area, and plaque burden) and symptoms was analyzed. Intraplaque hemorrhage was identified in 22 patients (17.5%), including 21 (31.8%) symptomatic patients and 1 (1.7%) asymptomatic patient. Multivariate analysis showed that intraplaque hemorrhage was the strongest independent marker of symptomatic status. Contrast enhancement was also independently associated with symptomatic status. The authors conclude that intraplaque hemorrhage is present in both low- and high-grade stenotic basilar artery plaques and is independently associated with symptomatic stroke status. Intraplaque hemorrhage may identify high-risk plaque and provide new insight into the management of patients with stroke without significant stenosis.