Scale-Equivariant Unrolled Neural Networks for Data-Efficient Accelerated MRI Reconstruction
This work addresses the challenge of data scarcity and scale variability in clinical MRI reconstruction, offering an incremental improvement over existing unrolled neural networks.
The paper tackles the problem of data-efficient accelerated MRI reconstruction by proposing scale-equivariant unrolled neural networks, which reduce dependence on costly fully-sampled scans and improve robustness to scale variations, achieving strong improvements over state-of-the-art methods without significantly increasing train or inference time.
Unrolled neural networks have enabled state-of-the-art reconstruction performance and fast inference times for the accelerated magnetic resonance imaging (MRI) reconstruction task. However, these approaches depend on fully-sampled scans as ground truth data which is either costly or not possible to acquire in many clinical medical imaging applications; hence, reducing dependence on data is desirable. In this work, we propose modeling the proximal operators of unrolled neural networks with scale-equivariant convolutional neural networks in order to improve the data-efficiency and robustness to drifts in scale of the images that might stem from the variability of patient anatomies or change in field-of-view across different MRI scanners. Our approach demonstrates strong improvements over the state-of-the-art unrolled neural networks under the same memory constraints both with and without data augmentations on both in-distribution and out-of-distribution scaled images without significantly increasing the train or inference time.