Undersampled K-space Data Research Articles

LAPNet: Non-Rigid Registration Derived in k-Space for Magnetic Resonance Imaging.

Physiological motion, such as cardiac and respiratory motion, during Magnetic Resonance (MR) image acquisition can cause image artifacts. Motion correction techniques have been proposed to compensate for these types of motion during thoracic scans, relying on accurate motion estimation from undersampled motion-resolved reconstruction. A particular interest and challenge lie in the derivation of reliable non-rigid motion fields from the undersampled motion-resolved data. Motion estimation is usually formulated in image space via diffusion, parametric-spline, or optical flow methods. However, image-based registration can be impaired by remaining aliasing artifacts due to the undersampled motion-resolved reconstruction. In this work, we describe a formalism to perform non-rigid registration directly in the sampled Fourier space, i.e. k-space. We propose a deep-learning based approach to perform fast and accurate non-rigid registration from the undersampled k-space data. The basic working principle originates from the Local All-Pass (LAP) technique, a recently introduced optical flow-based registration. The proposed LAPNet is compared against traditional and deep learning image-based registrations and tested on fully-sampled and highly-accelerated (with two undersampling strategies) 3D respiratory motion-resolved MR images in a cohort of 40 patients with suspected liver or lung metastases and 25 healthy subjects. The proposed LAPNet provided consistent and superior performance to image-based approaches throughout different sampling trajectories and acceleration factors.

Joint reconstruction framework of compressed sensing and nonlinear parallel imaging for dynamic cardiac magnetic resonance imaging

Compressed Sensing (CS) and parallel imaging are two promising techniques that accelerate the MRI acquisition process. Combining these two techniques is of great interest due to the complementary information used in each. In this study, we proposed a novel reconstruction framework that effectively combined compressed sensing and nonlinear parallel imaging technique for dynamic cardiac imaging. Specifically, the proposed method decouples the reconstruction process into two sequential steps: In the first step, a series of aliased dynamic images were reconstructed from the highly undersampled k-space data using compressed sensing; In the second step, nonlinear parallel imaging technique, i.e. nonlinear GRAPPA, was utilized to reconstruct the original dynamic images from the reconstructed k-space data obtained from the first step. In addition, we also proposed a tailored k-space down-sampling scheme that satisfies both the incoherent undersampling requirement for CS and the structured undersampling requirement for nonlinear parallel imaging. The proposed method was validated using four in vivo experiments of dynamic cardiac cine MRI with retrospective undersampling. Experimental results showed that the proposed method is superior at reducing aliasing artifacts and preserving the spatial details and temporal variations, compared with the competing k-t FOCUSS and k-t FOCUSS with sensitivity encoding methods, with the same numbers of measurements.

Blind Primed Supervised (BLIPS) Learning for MR Image Reconstruction.

This paper examines a combined supervised-unsupervised framework involving dictionary-based blind learning and deep supervised learning for MR image reconstruction from under-sampled k-space data. A major focus of the work is to investigate the possible synergy of learned features in traditional shallow reconstruction using adaptive sparsity-based priors and deep prior-based reconstruction. Specifically, we propose a framework that uses an unrolled network to refine a blind dictionary learning-based reconstruction. We compare the proposed method with strictly supervised deep learning-based reconstruction approaches on several datasets of varying sizes and anatomies. We also compare the proposed method to alternative approaches for combining dictionary-based methods with supervised learning in MR image reconstruction. The improvements yielded by the proposed framework suggest that the blind dictionary-based approach preserves fine image details that the supervised approach can iteratively refine, suggesting that the features learned using the two methods are complementary.

Learning Data Consistency and its Application to Dynamic MR Imaging.

Magnetic resonance (MR) image reconstruction from undersampled k-space data can be formulated as a minimization problem involving data consistency and image prior. Existing deep learning (DL)-based methods for MR reconstruction employ deep networks to exploit the prior information and integrate the prior knowledge into the reconstruction under the explicit constraint of data consistency, without considering the real distribution of the noise. In this work, we propose a new DL-based approach termed Learned DC that implicitly learns the data consistency with deep networks, corresponding to the actual probability distribution of system noise. The data consistency term and the prior knowledge are both embedded in the weights of the networks, which provides an utterly implicit manner of learning reconstruction model. We evaluated the proposed approach with highly undersampled dynamic data, including the dynamic cardiac cine data with up to 24-fold acceleration and dynamic rectum data with the acceleration factor equal to the number of phases. Experimental results demonstrate the superior performance of the Learned DC both quantitatively and qualitatively than the state-of-the-art.

XCloud-pFISTA: A Medical Intelligence Cloud for Accelerated MRI.

Machine learning and artificial intelligence have shown remarkable performance in accelerated magnetic resonance imaging (MRI). Cloud computing technologies have great advantages in building an easily accessible platform to deploy advanced algorithms. In this work, we develop an open-access, easy-to-use and high-performance medical intelligence cloud computing platform (XCloud-pFISTA) to reconstruct MRI images from undersampled k-space data. Two state-of-the-art approaches of the Projected Fast Iterative Soft-Thresholding Algorithm (pFISTA) family have been successfully implemented on the cloud. This work can be considered as a good example of cloud-based medical image reconstruction and may benefit the future development of integrated reconstruction and online diagnosis system.

Synthetic T2-weighted images of the lumbar spine derived from an accelerated T2 mapping sequence: Comparison to conventional T2w turbo spin echo

ObjectivesTo evaluate the diagnostic usefulness of synthetic T2-weighted images of the lumbar spine derived from ten-fold undersampled k-space data using GRAPPATINI, a combination of a model-based approach for rapid T2 and M0 quantification (MARTINI) extended by generalized autocalibrating partial parallel acquistion (GRAPPA). Materials and methodsOverall, 58 individuals (26 female, mean age 23.3 ± 8.1 years) were examined at 3 Tesla with sagittal and axial T2w turbo spin echo (TSE) sequences compared to synthetic T2−weighted contrasts derived at identical effective echo times and spatial resolutions. Two blinded readers graded disk degeneration and evaluated the lumbar intervertebral disks for present herniation or annular tear. One reader reassessed all studies after four weeks. Weighted kappa statistics were calculated to assess inter-rater and intra-rater agreement. Also, all studies were segmented manually by one reader to compute contrast ratios (CR) and contrast-to-noise ratios (CNR) of the nucleus pulposus and the annulus fibrosus. ResultsOverall, the CRT2w was 4.45 ± 1.80 and CRT2synth was 4.71 ± 2.14. Both correlated (rsp = 0.768;p < 0.001) and differed (0.26 ± 1.38;p = 0.002) significantly. The CNRT2w was 1.73 ± 0.52 and CNRT2synth was 1.63 ± 0.50. Both correlated (rsp = 0.875;p < 0.001) and differed (−0.10 ± 0.25;p < 0.001) significantly. The inter-rater agreement was substantial to almost perfect (κ = 0.808–0.925) with the intra-rater agreement also substantial to almost perfect (κ = 0.862–0.963). The area under the curve of the receiver operating characteristics assessing disk herniation or annular tear ranged from 0.787 to 0.892. ConclusionsThis study concludes that synthetic images derived by GRAPPATINI can be used for clinical routine assessment with inter-rater and intra-rater agreements comparable to conventional T2w TSE.

Deep learning for radial SMS myocardial perfusion reconstruction using the 3D residual booster U-net

PurposeTo develop an end-to-end deep learning solution for quickly reconstructing radial simultaneous multi-slice (SMS) myocardial perfusion datasets with comparable quality to the pixel tracking spatiotemporal constrained reconstruction (PT-STCR) method. MethodsDynamic contrast enhanced (DCE) radial SMS myocardial perfusion data were obtained from 20 subjects who were scanned at rest and/or stress with or without ECG gating using a saturation recovery radial CAIPI turboFLASH sequence. Input to the networks consisted of complex coil combined images reconstructed using the inverse Fourier transform of undersampled radial SMS k-space data. Ground truth images were reconstructed using the PT-STCR pipeline. The performance of the residual booster 3D U-Net was tested by comparing it to state-of-the-art network architectures including MoDL, CRNN-MRI, and other U-Net variants. ResultsResults demonstrate significant improvements in speed requiring approximately 8 seconds to reconstruct one radial SMS dataset which is approximately 200 times faster than the PT-STCR method. Images reconstructed with the residual booster 3D U-Net retain quality of ground truth PT-STCR images (0.963 SSIM/40.238 PSNR/0.147 NRMSE). The residual booster 3D U-Net has superior performance compared to existing network architectures in terms of image quality, temporal dynamics, and reconstruction time. ConclusionResidual and booster learning combined with the 3D U-Net architecture was shown to be an effective network for reconstructing high-quality images from undersampled radial SMS datasets while bypassing the reconstruction time of the PT-STCR method.

Joint total variation-based reconstruction of multiparametric magnetic resonance images for mapping tissue types.

Multispectral analysis of coregistered multiparametric magnetic resonance (MR) images provides a powerful method for tissue phenotyping and segmentation. Acquisition of a sufficiently varied set of multicontrast MR images and parameter maps to objectively define multiple normal and pathologic tissue types can require long scan times. Accelerated MRI on clinical scanners with multichannel receivers exploits techniques such as parallel imaging, while accelerated preclinical MRI scanning must rely on alternate approaches. In this work, tumor-bearing mice were imaged at 7T to acquire k-space data corresponding to a series of images with varying T1-, T2- and T2*-weighting. A joint reconstruction framework is proposed to reconstruct a series of T1-weighted images and corresponding T1 maps simultaneously from undersampled Cartesian k-space data. The ambiguity introduced by undersampling was resolved by using model-based constraints and structural information from a reference fully sampled image as the joint total variation prior. This process was repeated to reconstruct T2-weighted and T2*-weighted images and corresponding maps of T2 and T2* from undersampled Cartesian k-space data. Validation of the reconstructed images and parameter maps was carried out by computing tissue-type maps, as well as maps of the proton density fat fraction (PDFF), proton density water fraction (PDwF), fat relaxation rate ( and water relaxation rate ( from the reconstructed data, and comparing them with ground truth (GT) equivalents. Tissue-type maps computed using 18% k-space data were visually similar to GT tissue-type maps, with dice coefficients ranging from 0.43 to 0.73 for tumor, fluid adipose and muscle tissue types. The mean T1 and T2 values within each tissue type computed using only 18% k-space data were within 8%-10% of the GT values from fully sampled data. The PDFF and PDwF maps computed using 27% k-space data were within 3%-15% of GT values and showed good agreement with the expected values for the four tissue types.

Accelerate gas diffusion-weighted MRI for lung morphometry with deep learning.

ObjectivesMultiple b-value gas diffusion-weighted MRI (DW-MRI) enables non-invasive and quantitative assessment of lung morphometry, but its long acquisition time is not well-tolerated by patients. We aimed to accelerate multiple b-value gas DW-MRI for lung morphometry using deep learning.MethodsA deep cascade of residual dense network (DC-RDN) was developed to reconstruct high-quality DW images from highly undersampled k-space data. Hyperpolarized 129Xe lung ventilation images were acquired from 101 participants and were retrospectively collected to generate synthetic DW-MRI data to train the DC-RDN. Afterwards, the performance of the DC-RDN was evaluated on retrospectively and prospectively undersampled multiple b-value 129Xe MRI datasets.ResultsEach slice with size of 64 × 64 × 5 could be reconstructed within 7.2 ms. For the retrospective test data, the DC-RDN showed significant improvement on all quantitative metrics compared with the conventional reconstruction methods (p < 0.05). The apparent diffusion coefficient (ADC) and morphometry parameters were not significantly different between the fully sampled and DC-RDN reconstructed images (p > 0.05). For the prospectively accelerated acquisition, the required breath-holding time was reduced from 17.8 to 4.7 s with an acceleration factor of 4. Meanwhile, the prospectively reconstructed results showed good agreement with the fully sampled images, with a mean difference of −0.72% and −0.74% regarding global mean ADC and mean linear intercept (Lm) values.ConclusionsDC-RDN is effective in accelerating multiple b-value gas DW-MRI while maintaining accurate estimation of lung microstructural morphometry, facilitating the clinical potential of studying lung diseases with hyperpolarized DW-MRI.Key Points• The deep cascade of residual dense network allowed fast and high-quality reconstruction of multiple b-value gas diffusion-weighted MRI at an acceleration factor of 4.• The apparent diffusion coefficient and morphometry parameters were not significantly different between the fully sampled images and the reconstructed results (p > 0.05).• The required breath-holding time was reduced from 17.8 to 4.7 s and each slice with size of 64 × 64 × 5 could be reconstructed within 7.2 ms.Supplementary InformationThe online version contains supplementary material available at 10.1007/s00330-021-08126-y.

Adaptive convolutional neural networks for accelerating magnetic resonance imaging via k-space data interpolation.

Deep learning in k-space has demonstrated great potential for image reconstruction from undersampled k-space data in fast magnetic resonance imaging (MRI). However, existing deep learning-based image reconstruction methods typically apply weight-sharing convolutional neural networks (CNNs) to k-space data without taking into consideration the k-space data's spatial frequency properties, leading to ineffective learning of the image reconstruction models. Moreover, complementary information of spatially adjacent slices is often ignored in existing deep learning methods. To overcome such limitations, we have developed a deep learning algorithm, referred to as adaptive convolutional neural networks for k-space data interpolation (ACNN-k-Space), which adopts a residual Encoder-Decoder network architecture to interpolate the undersampled k-space data by integrating spatially contiguous slices as multi-channel input, along with k-space data from multiple coils if available. The network is enhanced by self-attention layers to adaptively focus on k-space data at different spatial frequencies and channels. We have evaluated our method on two public datasets and compared it with state-of-the-art existing methods. Ablation studies and experimental results demonstrate that our method effectively reconstructs images from undersampled k-space data and achieves significantly better image reconstruction performance than current state-of-the-art techniques. Source code of the method is available at https://gitlab.com/qgpmztmf/acnn-k-space.

Projection-Based cascaded U-Net model for MR image reconstruction

Background and ObjectiveBackground and Objective: Recent studies in deep learning reveal that the U-Net stands out among the diverse set of deep models as an effective network structure, especially for imaging inverse problems. Initially, the U-Net model was developed to solve segmentation problems for biomedical images while using an annotated dataset. In this paper, we will study a novel application of the U-Net structure for the important inverse problem of MRI reconstruction. Deep networks are particularly efficient for the speed-up of the MR image reconstruction process by decreasing the data acquisition time, and they can significantly reduce the aliasing artifacts caused by the undersampling in the k-space. Our aim is to develop a novel and efficient cascaded U-Net framework for reconstructing MR images from undersampled k-space data. The new framework should have improved reconstruction performance when compared to competing methodologies. MethodsIn this paper, a novel cascaded framework utilizing the U-Net as a sub-block is being proposed. The introduced U-Net cascade structure is applied to the magnetic resonance image reconstruction problem. The connection between the cascaded U-Nets is realized in the form of a recently developed projection-based updated data consistency layer. The novel structure is implemented in the PyTorch environment, which is one of the standards for deep learning implementations. The recently created fastMRI dataset which forms an important benchmark for MRI reconstruction is used for training and testing purposes. ResultsWe present simulation results comparing the novel method with a variety of competitive deep networks. The new cascaded U-Net structures PSNR performance stands on average 1.28 dB higher than the baseline U-Net. The improvement, when compared to the standard CNN, is on average 3.32 dB. ConclusionsThe proposed cascaded U-Net configuration results in an improved reconstruction performance when compared to the CNN, the cascaded CNN, and also the singular U-Net structures, where the singular U-Net forms the baseline reconstruction method from the fastMRI package. The use of the projection-based updated data consistency layer also leads to improved quantitative (including SSIM, PSNR, and NMSE results) and qualitative results when compared to the use of the conventional data consistency layer.

CALIBRATIONLESS MRI RECONSTRUCTION WITH A PLUG-IN DENOISER.

Magnetic Resonance Imaging (MRI) is a noninvasive imaging technique that provides excellent soft-tissue contrast without using ionizing radiation. MRI's clinical application may be limited by long data acquisition time; therefore, MR image reconstruction from highly under-sampled k-space data has been an active research area. Calibrationless MRI not only enables a higher acceleration rate but also increases flexibility for sampling pattern design. To leverage non-linear machine learning priors, we pair our High-dimensional Fast Convolutional Framework (HICU) [1] with a plug-in denoiser and demonstrate its feasibility using 2D brain data.

Deep model-based magnetic resonance parameter mapping network (DOPAMINE) for fast T1 mapping using variable flip angle method.

Quantitative tissue characteristics, which provide valuable diagnostic information, can be represented by magnetic resonance (MR) parameter maps using magnetic resonance imaging (MRI); however, a long scan time is necessary to acquire them, which prevents the application of quantitative MR parameter mapping to real clinical protocols. For fast MR parameter mapping, we propose a deep model-based MR parameter mapping network called DOPAMINE that combines a deep learning network with a model-based method to reconstruct MR parameter maps from undersampled multi-channel k-space data. DOPAMINE consists of two networks: 1) an MR parameter mapping network that uses a deep convolutional neural network (CNN) that estimates initial parameter maps from undersampled k-space data (CNN-based mapping), and 2) a reconstruction network that removes aliasing artifacts in the parameter maps with a deep CNN (CNN-based reconstruction) and an interleaved data consistency layer by an embedded MR model-based optimization procedure. We demonstrated the performance of DOPAMINE in brain T1 map reconstruction with a variable flip angle (VFA) model. To evaluate the performance of DOPAMINE, we compared it with conventional parallel imaging, low-rank based reconstruction, model-based reconstruction, and state-of-the-art deep-learning-based mapping methods for three different reduction factors (R = 3, 5, and 7) and two different sampling patterns (1D Cartesian and 2D Poisson-disk). Quantitative metrics indicated that DOPAMINE outperformed other methods in reconstructing T1 maps for all sampling patterns and reduction factors. DOPAMINE exhibited quantitatively and qualitatively superior performance to that of conventional methods in reconstructing MR parameter maps from undersampled multi-channel k-space data. The proposed method can thus reduce the scan time of quantitative MR parameter mapping that uses a VFA model.

K sparse autoencoder-based accelerated reconstruction of magnetic resonance imaging

Owing to the sequential collection of phase encoded data in k-space, magnetic resonance (MR) imaging suffers from long acquisition time. One possible measure to reduce the long acquisition time is to reconstruct MR image using a subset of k-space MR data rather than the complete set. In this work, we propose to implement a K sparse autoencoder model for reconstruction of MR image from undersampled k-space data. Autoencoder models, which have shown great ability in capturing the complex features of input data, can be used to reconstruct high-quality MR image. The reconstruction process involved solving an optimization problem whose solution was expected to satisfy the data consistency and also lie in close proximity to the output space of trained K sparse autoencoder. Observing the effect of sparsity value enforced by K sparse autoencoder on the reconstructed output, we implemented the cascaded form of reconstruction, incorporating three K sparse autoencoders with three different K values. Using MR-PD and MR-T1 images, reconstruction performance of the proposed approach was compared with those of the conventional reconstruction approaches. The quantitative as well as the qualitative analysis of the reconstructed images, obtained using the proposed approach, validates the efficiency of the proposed approach.

Compressed sensing time-of-flight magnetic resonance angiography with high spatial resolution for evaluating intracranial aneurysms: comparison with digital subtraction angiography.

Compressed sensing is used for accelerated acquisitions with incoherently under-sampled k-space data, and intracranial time-of-flight magnetic resonance angiography is suitable for compressed sensing. Compressed sensing time-of-flight is beneficial in decreasing acquisition time and increasing spatial resolution while maintaining acquisition time. In this retrospective study, we aimed to evaluate the image quality and diagnostic performance of compressed sensing time-of-flight with high spatial resolution and compare with parallel imaging time-of-flight using digital subtraction angiography as a reference. In total, 39 patients with 46 intracranial aneurysms underwent parallel imaging and compressed sensing time-of-flight in the same imaging session and digital subtraction angiography before or after magnetic resonance angiography. The overall image quality, artefacts and diagnostic confidence were assessed by two observers. The contrast ratio, maximal aneurysm diameters and diagnostic performance were evaluated. Compressed sensing time-of-flight showed significantly better overall image quality, degree of artefacts and diagnostic confidence in both observers, with better inter-observer agreement. The contrast ratio was significantly higher for compressed sensing time-of-flight than for parallel imaging time-of-flight in both observers (source images, P < 0.001; maximum intensity projection images, P < 0.05 for both observers); however, the measured maximal diameters of aneurysms were not significantly different. Compressed sensing time-of-flight showed higher sensitivity, specificity, accuracy and positive and negative predictive values for detecting aneurysms than parallel imaging time-of-flight in both observers, with better inter-observer agreement. Compressed sensing time-of-flight was preferred over parallel imaging time-of-flight by both observers; however, parallel imaging time-of-flight was preferred in cases of giant and large aneurysms. Compressed sensing-time-of-flight provides better image quality and diagnostic performance than parallel imaging time-of-flight. However, neuroradiologists should be aware of under-sampling artefacts caused by compressed sensing.

A Modified Generative Adversarial Network Using Spatial and Channel-Wise Attention for CS-MRI Reconstruction

Compressed sensing (CS) can speed up the magnetic resonance imaging (MRI) process and reconstruct high-quality images from under-sampled k-space data. However, traditional CS-MRI suffers from slow iterations and artifacts caused by noise when the acceleration factor is high. Currently, deep learning has been introduced to address these issues. Although some improvements have been achieved, the reconstruction problem under high under-sampling rates has not been solved. Thus, our study proposed a novel CS-MRI reconstruction method called RSCA-GAN. The generator of RSCA-GAN is a residual U-net consisting of both spatial and channel-wise attention. The proposed RSCA-GAN was compared to the zero-filling, DAGAN, RefineGAN, and RCA-GAN using both cartesian and non-cartesian under-sampling masks on brain and knee datasets. The sampling rates of cartesian masks are 25%, 16.7%, and 12.5% and the sampling rates of spiral and radial masks are 30%, 20%, and 10%. Peak signal-to-noise ratio (PSNR), structural similarity (SSIM) and normalized mean square error (NMSE) were used to evaluate the reconstructed image quality, and the rank-sum test was adopted to evaluate the significant difference among different approaches. P <; 0.05 indicated statistical significance. The results demonstrated that RSCA-GAN outperforms the other approaches for all the quantitative metrics. Thus, RSCA-GAN exhibits excellent reconstruction performance at high under-sampling rates and is suitable for clinical applications.

Rosette Trajectories for Fast MRI Based on an Adaptive Reconstruction Method

Non-Cartesian MRI k-space trajectories provide faster and more motion-robust data acquisitions than those of Cartesian trajectories. In this paper, we focus on the rosette trajectory and generalize the weighted rosette trajectories for fast undersampled k-space data acquisition. However, single-slice imaging using the rosette trajectory will be affected by the off-resonance effect. To reduce the artifacts from off-resonance slices, this paper introduced an adaptive iterative reconstruction method derived from the TV and nuclear norm regularization terms for raw MRI data reconstruction. This reconstruction problem was separated into two subproblems and solved with an adaptive regularization parameter. The results show that the weighted rosette trajectory that offers short acquisition times and good off-resonance behaviors with little blurring can be used for reconstruction with adaptive regularization parameters and achieve a superior performance.

A k-space-to-image reconstruction network for MRI using recurrent neural network.

Reconstructing the images from undersampled k-space data are an ill-posed inverse problem. As a solution to this problem, we propose a method to reconstruct magnetic resonance (MR) images directly from k-space data using a recurrent neural network. A novel neural network architecture named "ETER-net" is developed as a unified solution to reconstruct MR images from undersampled k-space data, where two bi-RNNs and convolutional neural network (CNN) are utilized to perform domain transformation and de-aliasing. To demonstrate the practicality of the proposed method, we conducted model optimization, cross-validation, and network pruning using in-house data from a 3T MRI scanner and public dataset called "FastMRI." The experimental results showed that the proposed method could be utilized for accurate image reconstruction from undersampled k-space data. The size of the proposed model was optimized and cross-validation was performed to show the robustness of the proposed method. For in-house dataset (R=4), the proposed method provided nMSE=1.09% and SSIM=0.938. For "FastMRI" dataset, the proposed method provided nMSE=1.05 % and SSIM=0.931 for R=4, and nMSE=3.12 % and SSIM=0.884 for R=8. The performance of the pruned model trained the loss function including with L2 regularization was consistent for a pruning ratio of up to 70%. The proposed method is an end-to-end MR image reconstruction method based on recurrent neural networks. It performs direct mapping of the input k-space data and the reconstructed images, operating as a unified solution that is applicable to various scanning trajectories.

Parallel imaging with a combination of sensitivity encoding and generative adversarial networks.

Magnetic resonance imaging (MRI) has the limitation of low imaging speed. Acceleration methods using under-sampled k-space data have been widely exploited to improve data acquisition without reducing the image quality. Sensitivity encoding (SENSE) is the most commonly used method for multi-channel imaging. However, SENSE has the drawback of severe g-factor artifacts when the under-sampling factor is high. This paper applies generative adversarial networks (GAN) to remove g-factor artifacts from SENSE reconstructions. Our method was evaluated on a public knee database containing 20 healthy participants. We compared our method with conventional GAN using zero-filled (ZF) images as input. Structural similarity (SSIM), peak signal to noise ratio (PSNR), and normalized mean square error (NMSE) were calculated for the assessment of image quality. A paired student's t-test was conducted to compare the image quality metrics between the different methods. Statistical significance was considered at P<0.01. The proposed method outperformed SENSE, variational network (VN), and ZF + GAN methods in terms of SSIM (SENSE + GAN: 0.81±0.06, SENSE: 0.40±0.07, VN: 0.79±0.06, ZF + GAN: 0.77±0.06), PSNR (SENSE + GAN: 31.90±1.66, SENSE: 22.70±1.99, VN: 31.35±2.01, ZF + GAN: 29.95±1.59), and NMSE (×10-7) (SENSE + GAN: 0.95±0.34, SENSE: 4.81±1.33, VN: 0.97±0.30, ZF + GAN: 1.60±0.84) with an under-sampling factor of up to 6-fold. This study demonstrated the feasibility of using GAN to improve the performance of SENSE reconstruction. The improvement of reconstruction is more obvious for higher under-sampling rates, which shows great potential for many clinical applications.

Motion correction for PET data using subspace-based real-time MR imaging in simultaneous PET/MR

Image quality of positron emission tomography (PET) reconstructions is degraded by subject motion occurring during the acquisition. Magnetic resonance (MR)-based motion correction approaches have been studied for PET/MR scanners and have been successful at capturing regular motion patterns, when used in conjunction with surrogate signals (e.g. navigators) to detect motion. However, handling irregular respiratory motion and bulk motion remains challenging. In this work, we propose an MR-based motion correction method relying on subspace-based real-time MR imaging to estimate motion fields used to correct PET reconstructions. We take advantage of the low-rank characteristics of dynamic MR images to reconstruct high-resolution MR images at high frame rates from highly undersampled k-space data. Reconstructed dynamic MR images are used to determine motion phases for PET reconstruction and estimate phase-to-phase nonrigid motion fields able to capture complex motion patterns such as irregular respiratory and bulk motion. MR-derived binning and motion fields are used for PET reconstruction to generate motion-corrected PET images. The proposed method was evaluated on in vivo data with irregular motion patterns. MR reconstructions accurately captured motion, outperforming state-of-the-art dynamic MR reconstruction techniques. Evaluation of PET reconstructions demonstrated the benefits of the proposed method in terms of motion artifacts reduction, improving the contrast-to-noise ratio by up to a factor 3 and achieveing a target-to-background ratio up to 90% superior compared to standard/uncorrected methods. The proposed method can improve the image quality of motion-corrected PET reconstructions in clinical applications.