Index by author

MR imaging–pathologic studies have reported that paramagnetic rims on 7T susceptibility-based MR imaging identify, in vivo, a subset of MS lesions with compartmentalized inflammation at the lesion edge and associated remyelination failure. High-resolution T2* and phase MR imaging were collected in 20 patients with MS at 3T and 7T. Phase rims were seen in 34 lesions at 7T and in 36 lesions at 3T by consensus. Inter- and intra-rater reliability were “substantial/good” both at 3T and 7T analysis. Nearly all 7T paramagnetic rims can also be seen at 3T. Imaging at 3T opens the possibility of implementing paramagnetic rims as an outcome measure.

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.