The University of Arizona
Please note that this event has ended!

A machine learning approach for accelerating High Angular Resolution Diffusion Imaging Acquisitions

Sparsity in Imaging

A machine learning approach for accelerating High Angular Resolution Diffusion Imaging Acquisitions
Series: Sparsity in Imaging
Location: MATH 402
Presenter: Ethan Lockhart, University of Arizona

High Angular Resolution Diffusion Imaging (HARDI) is a diffusion weighted magnetic resonance imaging (DW-MRI) modality in which the diffusion  of water molecules in the brain is sampled along many diffusion directions and shells. This results in long patient scan times. Instead of requiring the acquisition of volumetric images for all directions needed, neural networks are trained to predict unsampled diffusion direction images using a significantly reduced set of measurements, reducing scan times by a factor of 6 for some applications.

 

(This is also an organizational meeting for this seminar.)