An untrained deep learning method for reconstructing dynamic MR images from accelerated model‐based data

Slavkova, Kalina P.; DiCarlo, Julie; Viraj, Wadhwa,; Wu, Chengyue; Virostko, John; Kumar, Sidharth; Yankeelov, Thomas E.; Tamir, Jonathan I.

doi:10.1002/mrm.29547

Cited by 4 publications

(7 citation statements)

References 74 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…When stopping the iteration process of the UNN on blade reconstruction needs more study. The recent work (Slavkova et al 2023) combining the UNN with a physicsbased regularization loss may be helpful in determining the optimal stopping point on each blade reconstruction and achieving the optimal PROPELLER UNN image quality.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Suppressing image blurring of PROPELLER MRI via untrained method

Saju

Mao

et al. 2023

Phys. Med. Biol.

View full text Add to dashboard Cite

Objective: Periodically Rotated Overlapping ParallEL Lines with Enhanced Reconstruction (PROPELLER) used in magnetic resonance imaging (MRI) is inherently insensitive to motion artifacts but with an expense of around 60% increase in minimum scan time. An untrained deep learning method is proposed to accelerate PROPELLER MRI while suppressing image blurring. Approach: Several reconstruction methods have been developed to accelerate PROPELLER with reduced sampling on blades. However, image quality is degraded due to blurring. Deep learning has been applied to enhance MRI reconstruction quality, and external training data are therefore needed. In addition, the distribution shift problem in deep learning also exists between the external training data and to-be-reconstructed target blade data. This paper introduces an untrained neural network (UNN) to suppress image blurring, which is applied to improve PROPELLER MRI. This network structure was then incorporated into blade k-space.Results: The untrained method improved the blade image quality from brain MRI data. Furthermore, it enhanced the sharpness of the reconstructed image compared to PROPELLER reconstructions using parallel imaging methods and supervised learning methods using external training data. PROPELLER blade acquisition was accelerated by undersampling data with reduction factors 2, 3, and 4.Significance: The reported UNN enhanced PROPELLER method can improve image quality by suppressing blurring. External training data are not needed to mitigate the challenge of collecting high-quality clinical data for training without affecting clinical workflow and the standard care for patients.

show abstract

Section: Discussionmentioning

confidence: 99%

“…Additionally, our UNN can be integrated with model-based reconstruction methods. This unique feature allows us to determine the optimal training stopping point, all without relying on fully-sampled ground truth data (Slavkova et al 2023).…”

Section: Introductionmentioning

confidence: 99%

Suppressing image blurring of PROPELLER MRI via untrained method

Saju

Mao

et al. 2023

Phys. Med. Biol.

View full text Add to dashboard Cite

show abstract

“…∥ • ∥ 1 is a sparse promoting regularizer. Instead of relying on the sparse nature of CS, UNN has proven effective by parameterizing it through a carefully designed deep neural network (Lempitsky et al 2018, Darestani and Heckel 2021, Slavkova et al 2023. Then, equation (7) can be reformulated in k-space as follows…”

Section: Latent Image Characterizationmentioning

confidence: 99%

“…Unsupervised generative convolutional neural networks (named as UNN) without the need for any training data have made promising strides in approximating low-level statistical priors, relying on the neural network structure for sparse regularization (Heckel and Hand 2018, Lempitsky et al 2018, Slavkova et al 2023. UNN has made significant progress in effectively addressing the challenging inverse problem encountered in one of the MR reconstructions (Darestani andHeckel 2021, Liu et al 2023).…”

Section: Introductionmentioning

confidence: 99%

“…Specifically, compressed sensing (CS) fixes the sparse transform and seeks a sparse coefficient to represent the desired solution, while UNN alternatively fixes a sparse coefficient and seeks an adaptive transformation to represent the desired solution (Cui et al 2023). Furthermore, improvements have been made to accelerate dynamic imaging by utilizing UNN regularized by spoiled gradient echo (SPGR) (Slavkova et al 2023). Sequentially, leveraging UNN to generalize the degradation process of blurred images provides a solid foundation and also demonstrates the insensitivity of UNN to distribution shifts (Saju et al 2023), which can be viewed as the characterization of image low-level statistical priors through a parameterized convolutional neural network.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

DPP: deep phase prior for parallel imaging with wave encoding

Liu,

Cui,

Jia

et al. 2024

Phys. Med. Biol.

View full text Add to dashboard Cite

Objective: In MR parallel imaging with virtual channel-expanded Wave encoding, limitations are imposed on the ability to comprehensively and accurately characterize the background phase. These limitations are primarily attributed to the calibrationprocess relying solely on center low-frequency ACS data for calibration. Approach: To tackle the challenge of accurately estimating the background phase in wave encoding, a novel deep neural network model guided by deep phase priors (DPP) is proposed with integrated virtual conjugate coil (VCC) extension. Concretely, within the proposed framework, the background phase is implicitly characterized by employing a carefully designed decoder convolutional neural network, leveraging the inherent characteristics of phase smoothness and compact support in the transformed domain. Furthermore, the proposed model with wave encoding benefits from additional priors, which incorporate transmission sparsity of the latent image and coil sensitivity smoothness.Main results: Ablation experiments were conducted to ascertain the proposed method’s capability to implicitly represent CSM and the background phase. Subsequently, the superiority of the proposed method is demonstrated through confidence comparisons with competing methods, employing 4-fold and 5-fold acceleration experiments. In achieving 4-fold and 5-fold acceleration, the optimalquantitative metrics (PSNR/SSIM/NMSE) are 44.1359 dB/0.9863/0.0008 (4-fold) and 41.2074/0.9846/0.0017 (5-fold), respectively. Furthermore, the generalizability of the proposed method is further validated by conducting acceleration experiments with T1, T2, T2*, and various undersampling patterns. In addition, the DPP delivered much better performance than the conventional methods byexploring accelerated phase-sensitive SWI imaging. In SWI accelerated imaging, it also surpasses the optimal competing method in terms of (PSNR/SSIM/NMSE) with 0.096%/0.009%/0.0017%.Significance: The proposed method enables precise characterization of the background phase in the integrated VCC and wave encoding framework, supported via theoretical analysis and empirical findings. Our code is available at: https://github.com/sober235/DPP.

show abstract

Enhancing quality and speed in database‐free neural network reconstructions of undersampled MRI with SCAMPI

Siedler,

Jakob,

Herold

2024

Magnetic Resonance in Med

View full text Add to dashboard Cite

PurposeWe present SCAMPI (Sparsity Constrained Application of deep Magnetic resonance Priors for Image reconstruction), an untrained deep Neural Network for MRI reconstruction without previous training on datasets. It expands the Deep Image Prior approach with a multidomain, sparsity‐enforcing loss function to achieve higher image quality at a faster convergence speed than previously reported methods.MethodsTwo‐dimensional MRI data from the FastMRI dataset with Cartesian undersampling in phase‐encoding direction were reconstructed for different acceleration rates for single coil and multicoil data.ResultsThe performance of our architecture was compared to state‐of‐the‐art Compressed Sensing methods and ConvDecoder, another untrained Neural Network for two‐dimensional MRI reconstruction. SCAMPI outperforms these by better reducing undersampling artifacts and yielding lower error metrics in multicoil imaging. In comparison to ConvDecoder, the U‐Net architecture combined with an elaborated loss‐function allows for much faster convergence at higher image quality. SCAMPI can reconstruct multicoil data without explicit knowledge of coil sensitivity profiles. Moreover, it is a novel tool for reconstructing undersampled single coil k‐space data.ConclusionOur approach avoids overfitting to dataset features, that can occur in Neural Networks trained on databases, because the network parameters are tuned only on the reconstruction data. It allows better results and faster reconstruction than the baseline untrained Neural Network approach.

show abstract

An untrained deep learning method for reconstructing dynamic MR images from accelerated model‐based data

Cited by 4 publications

References 74 publications

Suppressing image blurring of PROPELLER MRI via untrained method

Suppressing image blurring of PROPELLER MRI via untrained method

DPP: deep phase prior for parallel imaging with wave encoding

Enhancing quality and speed in database‐free neural network reconstructions of undersampled MRI with SCAMPI

Contact Info

Product

Resources

About