ModelArray is a scalable R package for statistical analysis of fixel-wise data derived from diffusion MRI (and beyond). It supports linear and nonlinear modeling and is extensible to more models. Full documentation: https://pennlinc.github.io/ModelArray/

If you're a neuroscientist who wants to use DSI to look at some cool microstructural details but can't afford to add the extra 20 minutes of scan time to your protocol, compressed sensing might be your new best friend!

We show that CS-DSI schemes can generate white matter derivatives comparable to those generated by a full DSI scheme, allowing up to a 60% decrease in scan time with minimal loss in accuracy or reliability! This could allow both researchers and even clinicians to harness the advantages of DSI sequences that were previously impractical to deploy.

Hierarchical processing requires activity propagating between higher and lower-order cortical areas. Here, we leveraged advances in neuroimaging and computer vision to describe propagations robustly ascend and descend the cortical hierarchy. Notably, top-down propagations become both more prevalent with cognitive control demands and with development in youth.