Self-supervised Bayesian Deep Learning for Image Recovery with Applications to Compressive Sensing

Pang, Tongyao; Quan, Yuhui; Gao, Hui

doi:10.1007/978-3-030-58621-8_28

Cited by 18 publications

(9 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Several studies have investigated self-supervised learning for MRI reconstruction. [197][198][199][200][201][202][203][204][205][206][207][208][209][210][211][212] For instance, Yaman et al 197 introduced a self-supervised approach (SSDU) in which the acquired undersampled data indices are divided into a set of k-space positions used in the network's DC layer during training, and a set of k-space positions used within the loss function. This is a classic work in self-supervised MRI reconstruction, offering valuable insights for subsequent self-supervised learning methods.…”

Section: Unsupervised DL For Fast Mrimentioning

confidence: 99%

“…Several studies have investigated self‐supervised learning for MRI reconstruction 197–212 . For instance, Yaman et al 197 introduced a self‐supervised approach (SSDU) in which the acquired undersampled data indices are divided into a set of k‐space positions used in the network's DC layer during training, and a set of k‐space positions used within the loss function.…”

Section: Paradigm Shift and Applications For Mri Reconstructionmentioning

confidence: 99%

“…Furthermore, some works have incorporated Bayesian theory into their solution of inverse problems. Pang et al 208 introduced a self‐supervised DL recovery approach, which was implemented upon Bayesian deep networks and predicted the uncertainty of the target image by training networks with random weights. Aggarwal et al 209 introduced ENSURE, an ensemble stein's unbiased risk estimate framework, for unsupervised training of MRI reconstruction models.…”

Section: Paradigm Shift and Applications For Mri Reconstructionmentioning

confidence: 99%

See 2 more Smart Citations

Knowledge‐driven deep learning for fast MR imaging: Undersampled MR image reconstruction from supervised to un‐supervised learning

Wang,

Wu,

Jia

et al. 2024

Magnetic Resonance in Med

View full text Add to dashboard Cite

Deep learning (DL) has emerged as a leading approach in accelerating MRI. It employs deep neural networks to extract knowledge from available datasets and then applies the trained networks to reconstruct accurate images from limited measurements. Unlike natural image restoration problems, MRI involves physics‐based imaging processes, unique data properties, and diverse imaging tasks. This domain knowledge needs to be integrated with data‐driven approaches. Our review will introduce the significant challenges faced by such knowledge‐driven DL approaches in the context of fast MRI along with several notable solutions, which include learning neural networks and addressing different imaging application scenarios. The traits and trends of these techniques have also been given which have shifted from supervised learning to semi‐supervised learning, and finally, to unsupervised learning methods. In addition, MR vendors' choices of DL reconstruction have been provided along with some discussions on open questions and future directions, which are critical for the reliable imaging systems.

show abstract

Section: Unsupervised DL For Fast Mrimentioning

confidence: 99%

Section: Paradigm Shift and Applications For Mri Reconstructionmentioning

confidence: 99%

Section: Paradigm Shift and Applications For Mri Reconstructionmentioning

confidence: 99%

See 1 more Smart Citation

Knowledge‐driven deep learning for fast MR imaging: Undersampled MR image reconstruction from supervised to un‐supervised learning

Wang,

Wu,

Jia

et al. 2024

Magnetic Resonance in Med

View full text Add to dashboard Cite

show abstract

“…A closely related line of work known as plug-and-play (P&P) [49] used denoisers as regularizers for solving inverse problems. A number of recent extensions have used this concept to develop MAP solutions for inverse problems [50][51][52][53][54][55][56][57][58][59][60][61].…”

Section: Related Workmentioning

confidence: 99%

Image Reconstruction from Cone Excitations using the Implicit Prior in a Denoiser

et al. 2022

View full text Add to dashboard Cite

We re-examine the problem of reconstructing a high-dimensional signal from a small set of linear measurements, in combination with image prior from a diffusion probabilistic model. Well-established methods for optimizing such measurements include principal component analysis (PCA), independent component analysis (ICA) and compressed sensing (CS), all of which rely on axis-or subspace-aligned statistical characterization. But many naturally occurring signals, including photographic images, contain richer statistical structure. To exploit such structure, we introduce a general method for obtaining an optimized set of linear measurements, assuming a Bayesian inverse solution that leverages the prior implicit in a neural network trained to perform denoising. We demonstrate that these measurements are distinct from those of PCA and CS, with significant improvements in minimizing squared reconstruction error. In addition, we show that optimizing the measurements for the SSIM perceptual loss leads to perceptually improved reconstruction. Our results highlight the importance of incorporating the specific statistical regularities of natural signals when designing effective linear measurements.

show abstract

“…While most of the models we talked about are within the scope of supervised learning, unsupervised (and self-supervised) learning is another prevailing approach for image reconstruction [10,63,74,75,31,7,37,50,57,53,54]. In comparison with the supervised learning-based approach, unsupervised methods are more reliant on the regularization term (i.e., r(•) in (2.1)) due to the lack of labels.…”

Section: Improving Image Reconstruction With Deep Learningmentioning

confidence: 99%

A Note on Machine Learning Approach for Computational Imaging

Dong¹

2022

Preprint

View full text Add to dashboard Cite

Computational imaging has been playing a vital role in the development of natural sciences. Advances in sensory, information, and computer technologies have further extended the scope of influence of imaging, making digital images an essential component of our daily lives. For the past three decades, we have witnessed phenomenal developments of mathematical and machine learning methods in computational imaging. In this note, we will review some of the recent developments of the machine learning approach for computational imaging and discuss its differences and relations to the mathematical approach. We will demonstrate how we may combine the wisdom from both approaches, discuss the merits and potentials of such a combination and present some of the new computational and theoretical challenges it brings about.

show abstract

Self-supervised Bayesian Deep Learning for Image Recovery with Applications to Compressive Sensing

Cited by 18 publications

References 36 publications

Knowledge‐driven deep learning for fast MR imaging: Undersampled MR image reconstruction from supervised to un‐supervised learning

Knowledge‐driven deep learning for fast MR imaging: Undersampled MR image reconstruction from supervised to un‐supervised learning

Image Reconstruction from Cone Excitations using the Implicit Prior in a Denoiser

A Note on Machine Learning Approach for Computational Imaging

Contact Info

Product

Resources

About