A Conditional Normalizing Flow for Accelerated Multi-Coil MR Imaging

Jeffrey Wen, Rizwan Ahmad, Philip Schniter
Proceedings of the 40th International Conference on Machine Learning, PMLR 202:36926-36939, 2023.

Abstract

Accelerated magnetic resonance (MR) imaging attempts to reduce acquisition time by collecting data below the Nyquist rate. As an ill-posed inverse problem, many plausible solutions exist, yet the majority of deep learning approaches generate only a single solution. We instead focus on sampling from the posterior distribution, which provides more comprehensive information for downstream inference tasks. To do this, we design a novel conditional normalizing flow (CNF) that infers the signal component in the measurement operator’s nullspace, which is later combined with measured data to form complete images. Using fastMRI brain and knee data, we demonstrate fast inference and accuracy that surpasses recent posterior sampling techniques for MRI. Code is available at https://github.com/jwen307/mri_cnf

Cite this Paper

BibTeX
@InProceedings{pmlr-v202-wen23a, title = {A Conditional Normalizing Flow for Accelerated Multi-Coil {MR} Imaging}, author = {Wen, Jeffrey and Ahmad, Rizwan and Schniter, Philip}, booktitle = {Proceedings of the 40th International Conference on Machine Learning}, pages = {36926--36939}, year = {2023}, editor = {Krause, Andreas and Brunskill, Emma and Cho, Kyunghyun and Engelhardt, Barbara and Sabato, Sivan and Scarlett, Jonathan}, volume = {202}, series = {Proceedings of Machine Learning Research}, month = {23--29 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v202/wen23a/wen23a.pdf}, url = {https://proceedings.mlr.press/v202/wen23a.html}, abstract = {Accelerated magnetic resonance (MR) imaging attempts to reduce acquisition time by collecting data below the Nyquist rate. As an ill-posed inverse problem, many plausible solutions exist, yet the majority of deep learning approaches generate only a single solution. We instead focus on sampling from the posterior distribution, which provides more comprehensive information for downstream inference tasks. To do this, we design a novel conditional normalizing flow (CNF) that infers the signal component in the measurement operator’s nullspace, which is later combined with measured data to form complete images. Using fastMRI brain and knee data, we demonstrate fast inference and accuracy that surpasses recent posterior sampling techniques for MRI. Code is available at https://github.com/jwen307/mri_cnf} }
Endnote
%0 Conference Paper %T A Conditional Normalizing Flow for Accelerated Multi-Coil MR Imaging %A Jeffrey Wen %A Rizwan Ahmad %A Philip Schniter %B Proceedings of the 40th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2023 %E Andreas Krause %E Emma Brunskill %E Kyunghyun Cho %E Barbara Engelhardt %E Sivan Sabato %E Jonathan Scarlett %F pmlr-v202-wen23a %I PMLR %P 36926--36939 %U https://proceedings.mlr.press/v202/wen23a.html %V 202 %X Accelerated magnetic resonance (MR) imaging attempts to reduce acquisition time by collecting data below the Nyquist rate. As an ill-posed inverse problem, many plausible solutions exist, yet the majority of deep learning approaches generate only a single solution. We instead focus on sampling from the posterior distribution, which provides more comprehensive information for downstream inference tasks. To do this, we design a novel conditional normalizing flow (CNF) that infers the signal component in the measurement operator’s nullspace, which is later combined with measured data to form complete images. Using fastMRI brain and knee data, we demonstrate fast inference and accuracy that surpasses recent posterior sampling techniques for MRI. Code is available at https://github.com/jwen307/mri_cnf
APA
Wen, J., Ahmad, R. & Schniter, P.. (2023). A Conditional Normalizing Flow for Accelerated Multi-Coil MR Imaging. Proceedings of the 40th International Conference on Machine Learning, in Proceedings of Machine Learning Research 202:36926-36939 Available from https://proceedings.mlr.press/v202/wen23a.html.

Related Material