GroupCDL: Interpretable Denoising and Compressed Sensing MRI via Learned Group-Sparsity and Circulant Attention
- 5 February 2025
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computational Imaging
- Vol. 11 (25730436) , 201-212
Abstract
Nonlocal self-similarity within images has become an increasingly popular prior in deep-learning models. Despite their successful image restoration performance, such models remain largely uninterpretable due to their black-box construction. Our previous studies have shown that interpretable construction of a fully convolutional denoiser (CDLNet), with performance on par with state-of-the-art black-box counterparts, is achievable by unrolling a convolutional dictionary learning algorithm. In this manuscript, we seek an interpretable construction of a convolutional network with a nonlocal self-similarity prior that performs on par with black-box nonlocal models. We show that such an architecture can be effectively achieved by upgrading the $\ell _{1}$ sparsity prior (soft-thresholding) of CDLNet to an image-adaptive group-sparsity prior (group-thresholding). The proposed learned group-thresholding makes use of nonlocal attention to perform spatially varying soft-thresholding on the latent representation. To enable effective training and inference on large images with global artifacts, we propose a novel circulant-sparse attention. We achieve competitive natural-image denoising performance compared to black-box nonlocal DNNs and transformers. The interpretable construction of our network allows for a straightforward extension to Compressed Sensing MRI (CS-MRI), yielding state-of-the-art performance. Lastly, we show robustness to noise-level mismatches between training and inference for denoising and CS-MRI reconstruction.Keywords
All Related Versions
Funding Information
- National Institute of Biomedical Imaging and Bioengineering (R01EB030549)
- National Institute of Biomedical Imaging and Bioengineering (R01 EB031083)
- National Institute of Biomedical Imaging and Bioengineering (R21 EB032917)
This publication has 36 references indexed in Scilit:
- Revisiting ℓ1-wavelet compressed-sensing MRI in the era of deep learningProceedings of the National Academy of Sciences, 2022
- Fully Trainable and Interpretable Non-local Sparse Models for Image Restoration2020
- End-to-End Variational Networks for Accelerated MRI Reconstruction2020
- VS-Net: Variable Splitting Network for Accelerated Parallel MRI Reconstruction2019
- ISTA-Net: Interpretable Optimization-Inspired Deep Network for Image Compressive Sensing2018
- Deep Residual Learning for Image Recognition2016
- U-Net: Convolutional Networks for Biomedical Image Segmentation2015
- Single image super-resolution from transformed self-exemplars2015
- k‐t group sparse: A method for accelerating dynamic MRIMagnetic Resonance in Medicine, 2011
- A Wavelet Tour of Signal Processing1999