McSTRA: A multi-branch cascaded swin transformer for point spread function-guided robust MRI reconstruction

Deep learning MRI reconstruction methods are often based on Convolutional neural network (CNN) models; however, they are limited in capturing global correlations among image features due to the intrinsic locality of the convolution operation. Conversely, the recent vision transformer models (ViT) are capable of capturing global correlations by applying self-attention operations on image patches. Nevertheless, the existing transformer models for MRI reconstruction rarely leverage the physics of MRI. In this paper, we propose a novel physics-based transformer model titled, the Multi-branch Cascaded Swin Transformers (McSTRA) for robust MRI reconstruction. McSTRA combines several interconnected MRI physics-related concepts with the Swin transformers: it exploits global MRI features via the shifted window self-attention mechanism; it extracts MRI features belonging to different spectral components via a multi-branch setup; it iterates between intermediate de-aliasing and data consistency via a cascaded network with intermediate loss computations; furthermore, we propose a point spread function-guided positional embedding generation mechanism for the Swin transformers which exploit the spread of the aliasing artifacts for effective reconstruction. With the combination of all these components, McSTRA outperforms the state-of-the-art methods while demonstrating robustness in adversarial conditions such as higher accelerations, noisy data, different undersampling protocols, out-of-distribution data, and abnormalities in anatomy.

Radial magnetic resonance image reconstruction with a deep unrolled projected fast iterative soft-thresholding network

Radially sampling of magnetic resonance imaging (MRI) is an effective way to accelerate the imaging. How to preserve the image details in reconstruction is always challenging. In this work, a deep unrolled neural network is designed to emulate the iterative sparse image reconstruction process of a projected fast soft-threshold algorithm (pFISTA). The proposed method, an unrolled pFISTA network for Deep Radial MRI (pFISTA-DR), include the preprocessing module to refine coil sensitivity maps and initial reconstructed image, the learnable convolution filters to extract image feature maps, and adaptive threshold to robustly remove image artifacts. Experimental results show that, among the compared methods, pFISTA-DR provides the best reconstruction and achieved the highest PSNR, the highest SSIM and the lowest reconstruction errors.

3D-EPI blip-up/down acquisition (BUDA) with CAIPI and joint Hankel structured low-rank reconstruction for rapid distortion-free high-resolution T 2 * mapping

PURPOSE

This work aims to develop a novel distortion-free 3D-EPI acquisition and image reconstruction technique for fast and robust, high-resolution, whole-brain imaging as well as quantitativeT 2 * $$ {\mathrm{T}}_2^{\ast } $$ mapping.

METHODS

3D Blip-up and -down acquisition (3D-BUDA) sequence is designed for both single- and multi-echo 3D gradient recalled echo (GRE)-EPI imaging using multiple shots with blip-up and -down readouts to encode B0 field map information. Complementary k-space coverage is achieved using controlled aliasing in parallel imaging (CAIPI) sampling across the shots. For image reconstruction, an iterative hard-thresholding algorithm is employed to minimize the cost function that combines field map information informed parallel imaging with the structured low-rank constraint for multi-shot 3D-BUDA data. Extending 3D-BUDA to multi-echo imaging permitsT 2 * $$ {\mathrm{T}}_2^{\ast } $$ mapping. For this, we propose constructing a joint Hankel matrix along both echo and shot dimensions to improve the reconstruction.

RESULTS

Experimental results on in vivo multi-echo data demonstrate that, by performing joint reconstruction along with both echo and shot dimensions, reconstruction accuracy is improved compared to standard 3D-BUDA reconstruction. CAIPI sampling is further shown to enhance image quality. ForT 2 * $$ {\mathrm{T}}_2^{\ast } $$ mapping, parameter values from 3D-Joint-CAIPI-BUDA and reference multi-echo GRE are within limits of agreement as quantified by Bland-Altman analysis.

CONCLUSIONS

The proposed technique enables rapid 3D distortion-free high-resolution imaging andT 2 * $$ {\mathrm{T}}_2^{\ast } $$ mapping. Specifically, 3D-BUDA enables 1-mm isotropic whole-brain imaging in 22 s at 3T and 9 s on a 7T scanner. The combination of multi-echo 3D-BUDA with CAIPI acquisition and joint reconstruction enables distortion-free whole-brainT 2 * $$ {\mathrm{T}}_2^{\ast } $$ mapping in 47 s at 1.1 × 1.1 × 1.0 mm3 resolution.

Blip up-down acquisition for spin- and gradient-echo imaging (BUDA-SAGE) with self-supervised denoising enables efficient T2 , T2 *, para- and dia-magnetic susceptibility mapping

PURPOSE

To rapidly obtain high resolution T₂ , T₂ *, and quantitative susceptibility mapping (QSM) source separation maps with whole-brain coverage and high geometric fidelity.

METHODS

We propose Blip Up-Down Acquisition for Spin And Gradient Echo imaging (BUDA-SAGE), an efficient EPI sequence for quantitative mapping. The acquisition includes multiple T₂ *-, T₂ '-, and T₂ -weighted contrasts. We alternate the phase-encoding polarities across the interleaved shots in this multi-shot navigator-free acquisition. A field map estimated from interim reconstructions was incorporated into the joint multi-shot EPI reconstruction with a structured low rank constraint to eliminate distortion. A self-supervised neural network (NN), MR-Self2Self (MR-S2S), was used to perform denoising to boost SNR. Using Slider encoding allowed us to reach 1 mm isotropic resolution by performing super-resolution reconstruction on volumes acquired with 2 mm slice thickness. Quantitative T₂ (=1/R₂ ) and T₂ * (=1/R₂ *) maps were obtained using Bloch dictionary matching on the reconstructed echoes. QSM was estimated using nonlinear dipole inversion on the gradient echoes. Starting from the estimated R₂ /R₂ * maps, R₂ ' information was derived and used in source separation QSM reconstruction, which provided additional para- and dia-magnetic susceptibility maps.

RESULTS

In vivo results demonstrate the ability of BUDA-SAGE to provide whole-brain, distortion-free, high-resolution, multi-contrast images and quantitative T₂ /T₂ * maps, as well as yielding para- and dia-magnetic susceptibility maps. Estimated quantitative maps showed comparable values to conventional mapping methods in phantom and in vivo measurements.

CONCLUSION

BUDA-SAGE acquisition with self-supervised denoising and Slider encoding enables rapid, distortion-free, whole-brain T₂ /T₂ * mapping at 1 mm isotropic resolution under 90 s.

Spatiotemporal Flexible Sparse Reconstruction for Rapid Dynamic Contrast-enhanced MRI

Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) is a tissue perfusion imaging technique. Some versatile free-breathing DCE-MRI techniques combining compressed sensing (CS) and parallel imaging with golden-angle radial sampling have been developed to improve motion robustness with high spatial and temporal resolution. These methods have demonstrated good diagnostic performance in clinical setting, but the reconstruction quality will degrade at high acceleration rates and overall reconstruction time remains long. In this paper, we proposed a new parallel CS reconstruction model for DCE-MRI that enforces flexible weighted sparse constraint along both spatial and temporal dimensions. Weights were introduced to flexibly adjust the importance of time and space sparsity, and we derived a fast-thresholding algorithm which was proven to be simple and efficient for solving the proposed reconstruction model. Results on both the brain tumor DCE and liver DCE show that, at relatively high acceleration factor of fast sampling, lowest reconstruction error and highest image structural similarity are obtained by the proposed method. Besides, the proposed method achieves faster reconstruction for liver datasets and better physiological measures are also obtained on tumor images.

Efficient T2 mapping with blip-up/down EPI and gSlider-SMS (T2 -BUDA-gSlider)

PURPOSE

To rapidly obtain high isotropic-resolution T₂ maps with whole-brain coverage and high geometric fidelity.

METHODS

A T₂ blip-up/down EPI acquisition with generalized slice-dithered enhanced resolution (T₂ -BUDA-gSlider) is proposed. A RF-encoded multi-slab spin-echo (SE) EPI acquisition with multiple TEs was developed to obtain high SNR efficiency with reduced TR. This was combined with an interleaved 2-shot EPI acquisition using blip-up/down phase encoding. An estimated field map was incorporated into the joint multi-shot EPI reconstruction with a structured low rank constraint to achieve distortion-free and robust reconstruction for each slab without navigation. A Bloch simulated subspace model was integrated into gSlider reconstruction and used for T₂ quantification.

RESULTS

In vivo results demonstrated that the T₂ values estimated by the proposed method were consistent with gold standard spin-echo acquisition. Compared to the reference 3D fast spin echo (FSE) images, distortion caused by off-resonance and eddy current effects were effectively mitigated.

CONCLUSION

BUDA-gSlider SE-EPI acquisition and gSlider-subspace joint reconstruction enabled distortion-free whole-brain T₂ mapping in 2 min at ~1 mm³ isotropic resolution, which could bring significant benefits to related clinical and neuroscience applications.

HF-SENSE: an improved partially parallel imaging using a high-pass filter

Background

One of the major limitations of MRI is its slow acquisition speed. To accelerate data acquisition, partially parallel imaging (PPI) methods have been widely used in clinical applications such as sensitivity encoding (SENSE) and generalized autocalibrating partially parallel acquisitions (GRAPPA). SENSE is a popular image-domain partially parallel imaging method, which suffers from residual aliasing artifacts when the reduction factor goes higher. Undersampling the k-space data and then reconstruct images with artificial sparsity is an efficient way to accelerate data acquisition. By exploiting artificial sparsity with a high-pass filter, an improved SENSE method is proposed in this work, termed high-pass filtered SENSE (HF-SENSE).

Methods

First, a high-pass filter was applied to the raw k-space data, the result of which was used as the inputs of sensitivity estimation and undersampling process. Second, the adaptive array coil combination method was adopted to calculate sensitivity maps on a block-by-block basis. Third, Tikhonov regularized SENSE was then used to reconstruct magnetic resonance images. Fourth, the reconstructed images were transformed into k-space data, which was filtered with the corresponding inverse filter.

Results

Both simulation and in vivo experiments demonstrate that HF-SENSE method significantly reduces noise level of the reconstructed images compared with SENSE. Furthermore, it is found that HF-SENSE can achieve lower normalized root-mean-square error value than SENSE.

Conclusions

The proposed method explores artificial sparsity with a high-pass filter. Experiments demonstrate that the proposed HF-SENSE method can improve the image quality of SENSE reconstruction. The high-pass filter parameters can be predefined. With this image reconstruction method, high acceleration factors can be achieved, which will improve the clinical applicability of SENSE.

This retrospective study (HF-SENSE: an improved partially parallel imaging using a high-pass filter) was approved by Institute Review Board of 2nd Affiliated Hospital of Zhejiang University (ethical approval number 2018–314). Participant for all images have informed consent that he knew the risks and agreed to participate in the research.

Rigid motion correction for magnetic resonance fingerprinting with sliding-window reconstruction and image registration

Magnetic resonance fingerprinting (MRF) can be used to simultaneously obtain multiple parameter maps from a single pulse sequence. However, patient motion during MRF acquisition may result in blurring and artifacts in estimated parameter maps. In this work, a novel motion correction method was proposed to correct for rigid motion in MRF. The proposed method involved sliding-window reconstruction to obtain intermediate images followed by image registration to estimate rigid motion information between these images. Finally, the motion-corrupted k-space data were corrected with the estimated motion parameters and then reconstructed to obtain the parameter maps via the conventional MRF processing pipeline. The proposed method was evaluated using both simulations and in vivo MRF experiments with intently different types of motion. For motion-corrupted data, the proposed method yielded brain T₁, T₂ and proton density maps with obviously reduced blurring and artifacts and lower normalized root-mean-square error, compared to MRF without motion correction. In conclusion, motion-corrected MRF using the proposed method has the potential to produce accurate parameter maps in the presence of in-plane rigid motion.

Technical Note: Clustering-based motion compensation scheme for multishot diffusion tensor imaging

PURPOSE

To extend image reconstruction using image-space sampling function (IRIS) to address large-scale motion in multishot diffusion-weighted imaging (DWI).

METHODS

A clustered IRIS (CIRIS) algorithm that would extend IRIS was proposed to correct for large-scale motion. For DWI, CIRIS initially groups the shots into clusters without intracluster large-scale motion and reconstructs each cluster by using IRIS. Then, CIRIS registers these cluster images and combines the registered images by using a weighted average to correct for voxel mismatch caused by intercluster large-scale motion. For diffusion tensor imaging (DTI), CIRIS further reduces the effect of motion on diffusion directions by treating motion-induced direction changes as additional diffusion directions. CIRIS also introduces the detection and rejection of motion-corrupted data to avoid corresponding image degradation. The proposed method was evaluated by simulation and in vivo diffusion datasets.

RESULTS

Experiments demonstrated that CIRIS can reduce motion-induced blurring and artifacts in DWI and provide more accurate DTI estimations in the presence of large-scale motion, compared with IRIS.

CONCLUSION

The proposed method presents a novel approach to correct for large-scale in-plane motion for multishot DWI and is expected to benefit the practical application of high-resolution diffusion imaging.

Improved k-t PCA Algorithm Using Artificial Sparsity in Dynamic MRI

The k-t principal component analysis (k-t PCA) is an effective approach for high spatiotemporal resolution dynamic magnetic resonance (MR) imaging. However, it suffers from larger residual aliasing artifacts and noise amplification when the reduction factor goes higher. To further enhance the performance of this technique, we propose a new method called sparse k-t PCA that combines the k-t PCA algorithm with an artificial sparsity constraint. It is a self-calibrated procedure that is based on the traditional k-t PCA method by further eliminating the reconstruction error derived from complex subtraction of the sampled k-t space from the original reconstructed k-t space. The proposed method is tested through both simulations and in vivo datasets with different reduction factors. Compared to the standard k-t PCA algorithm, the sparse k-t PCA can improve the normalized root-mean-square error performance and the accuracy of temporal resolution. It is thus useful for rapid dynamic MR imaging.