A Path Towards Clinical Adaptation of Accelerated MRI

Michael S. Yao

[email protected]

Microsoft Research

University of Pennsylvania, Department of Bioengineering

University of Pennsylvania, School of Medicine

Michael S. Hansen

michael.han[email protected]

Microsoft Research

Corresponding author

Project Links


Accelerated MRI reconstructs images of clinical anatomies from sparsely sampled signal data to reduce patient scan times. While recent works have leveraged deep learning to accomplish this task, such approaches have often only been explored in simulated environments where there is no signal corruption or resource limitations. In this work, we explore augmentations to neural network MRI image reconstructors to enhance their clinical relevancy. Namely, we propose a neural network-based approach for detecting sources of image artifacts during signal preprocessing. We also demonstrate that training reconstructors on MR signal data with variable acceleration factors can improve their average performance at any time step of a clinical patient scan. We offer a new loss function to overcome catastrophic forgetting when models learn to reconstruct MR images of multiple anatomies and orientations. Finally, we propose a framework using simulated phantom data to leverage transfer learning for learning to reconstruct other anatomies with limited clinically acquired datasets and compute capabilities.

Abstract Figure Abstract Figure


  title={A Path Towards Clinical Adaptation of Accelerated {MRI}},
  authors={Yao, Michael S. and Hansen, Michael S.},