Deep Recurrent Partial Fourier Reconstruction in Diffusion MRI

Gadjimuradov F, Benkert T, Nickel MD, Maier A (2020)


Publication Type: Conference contribution

Publication year: 2020

Journal

Publisher: Springer Science and Business Media Deutschland GmbH

Book Volume: 12450 LNCS

Pages Range: 38-47

Conference Proceedings Title: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

Event location: Lima PE

ISBN: 9783030615970

DOI: 10.1007/978-3-030-61598-7_4

Abstract

Partial Fourier (PF) acquisition schemes are often employed to increase the inherently low SNR in diffusion-weighted (DW) images. The resulting ill-posed reconstruction problem can be tackled by an iterative Projection Onto Convex Sets (POCS). By relaxing the data constraint and replacing the heuristically chosen regularization by learned convolutional filters, we arrive at an unrolled recurrent network architecture which circumvents weaknesses of the conventional POCS. Further, knowledge on the pixel-wise noise level of MR images is incorporated into data consistency operations within the reconstruction network. We are able to demonstrate on DW images of the pelvis that the proposed model quantitatively and qualitatively outperforms conventional methods as well as a U-Net representing a direct image-to-image mapping.

Authors with CRIS profile

How to cite

APA:

Gadjimuradov, F., Benkert, T., Nickel, M.D., & Maier, A. (2020). Deep Recurrent Partial Fourier Reconstruction in Diffusion MRI. In Farah Deeba, Patricia Johnson, Tobias Würfl, Jong Chul Ye (Eds.), Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (pp. 38-47). Lima, PE: Springer Science and Business Media Deutschland GmbH.

MLA:

Gadjimuradov, Fasil, et al. "Deep Recurrent Partial Fourier Reconstruction in Diffusion MRI." Proceedings of the 3rd International Workshop on Machine Learning for Medical Image Reconstruction, MLMIR 2020, held in conjunction with the 23rd Medical Image Computing and Computer Assisted Intervention, MICCAI 2020, Lima Ed. Farah Deeba, Patricia Johnson, Tobias Würfl, Jong Chul Ye, Springer Science and Business Media Deutschland GmbH, 2020. 38-47.

BibTeX: Download