Iterative SENSE Reconstruction Research Articles

DeepSENSE: Learning coil sensitivity functions for SENSE reconstruction using deep learning.

To improve the estimation of coil sensitivity functions from limited auto-calibration signals (ACS) in SENSE-based reconstruction for brain imaging. We propose to use deep learning to estimate coil sensitivity functions by leveraging information from previous scans obtained using the same RF receiver system. Specifically, deep convolutional neural networks were designed to learn an end-to-end mapping from the initial sensitivity to the high-resolution counterpart. Sensitivity alignment was further proposed to reduce the geometric variation caused by different subject positions and imaging FOVs. Cross-validation with a small set of datasets was performed to validate the learned neural network. Iterative SENSE reconstruction was adopted to evaluate the utility of the sensitivity functions from the proposed and conventional methods. The proposed method produced improved sensitivity estimates and SENSE reconstructions compared to the conventional methods in terms of aliasing and noise suppression with very limited ACS data. Cross-validation with a small set of data demonstrated the feasibility of learning coil sensitivity functions for brain imaging. The network learned on the spoiled GRE data can be applied to predict sensitivity functions for spin-echo and MPRAGE datasets. A deep learning-based method has been proposed for improving the estimation of coil sensitivity functions. Experimental results have demonstrated the feasibility and potential of the proposed method for improving SENSE-based reconstructions especially when the ACS data are limited.

Noise reduction in diffusion weighted MRI of the pancreas using an L1-regularized iterative SENSE reconstruction

ObjectivesTo prospectively evaluate an L1 regularized iterative SENSE reconstruction (L1-R SENSE) to eliminate band-like artifacts frequently seen with parallel imaging (SENSE) at high acceleration factors in high resolution diffusion weighted magnetic resonance imaging of the pancreas. MethodsFourteen patients with pancreatic ductal adenocarcinoma (PDAC) underwent respiratory triggered DWI ss-EPI at a resolution of 2.5 × 2.5 × 3 mm3 with uniform undersampling in the phase encoding direction (AP axis) with an acceleration factor of 4. Data were reconstructed using the standard SENSE reconstruction routine of the vendor and an iterative SENSE reconstruction employing L1 regularization after a wavelet sparsifying transformation (L1-R SENSE). Retrospective reconstruction of the data with a lower number of averages was performed using both reconstruction methods. Two radiologists independently assessed noise artifacts, anatomical details and image quality (IQ) subjectively with a 4-point scale. Apparent diffusion coefficient (ADC) and covariance (CV) of ADC estimated from images reconstructed at a different number of averages for PDAC and the normal pancreas were assessed. ResultsL1-R SENSE resulted in higher IQ and less noise artifacts than SENSE. Anatomical details were significantly higher for SENSE in one reader. Mean ADC of PDAC and normal pancreas were significantly higher for L1-R SENSE than SENSE. L1-R SENSE revealed lower CV of ADC for normal pancreas compared to SENSE, whereas no difference was noted for PDAC. ConclusionCompared with traditional SENSE reconstruction, L1-R SENSE effectively reduces band-like noise and improves the robustness of the ADC estimation from acquisitions using single-shot DW-EPI of the pancreas.

Development of a spiral spin- and gradient-echo (spiral-SAGE) approach for improved multi-parametric dynamic contrast neuroimaging.

The purpose of this study was to develop a spiral-based combined spin- and gradient-echo (spiral-SAGE) method for simultaneous dynamic contrast-enhanced (DCE-MRI) and dynamic susceptibility contrast MRI (DSC-MRI). Using this sequence, we obtained gradient-echo TEs of 1.69 and 26 ms, a SE TE of 87.72 ms, with a TR of 1663 ms. Using an iterative SENSE reconstruction followed by deblurring, spiral-induced image artifacts were minimized. Healthy volunteer images are shown to demonstrate image quality using the optimized reconstruction, as well as for comparison with EPI-based SAGE. A bioreactor phantom was used to compare dynamic-contrast time courses with both spiral-SAGE and EPI-SAGE. A proof-of-concept cohort of patients with brain tumors shows the range of hemodynamic maps available using spiral-SAGE. Comparison of spiral-SAGE images with conventional EPI-SAGE images illustrates substantial reductions of image distortion and artifactual image intensity variations. Bioreactor phantom data show similar dynamic contrast time courses between standard EPI-SAGE and spiral-SAGE for the second and third echoes, whereas first-echo data show improvements in quantifying T1 changes with shorter echo times. In a cohort of patients with brain tumors, spiral-SAGE-based perfusion and permeability maps are shown with comparison with the standard single-echo EPI perfusion map. Spiral-SAGE provides a substantial improvement for the assessment of perfusion and permeability by mitigating artifacts typically encountered with EPI and by providing a shorter echo time for improved characterization of permeability. Spiral-SAGE enables quantification of perfusion, permeability, and vessel architectural parameters, as demonstrated in brain tumors.

Accelerated real-time cardiac MRI using iterative sparse SENSE reconstruction: comparing performance in patients with sinus rhythm and atrial fibrillation.

To compare accelerated real-time cardiac MRI (CMR) using sparse spatial and temporal undersampling and non-linear iterative SENSE reconstruction (RT IS SENSE) with real-time CMR (RT) and segmented CMR (SEG) in a cohort that included atrial fibrillation (AF) patients. We evaluated 27 subjects, including 11 AF patients, by acquiring steady-state free precession cine images covering the left ventricle (LV) at 1.5 T with SEG (acceleration factor 2, TR 42 ms, 1.8 × 1.8 × 6 mm3), RT (acceleration factor 3, TR 62 ms, 3.0 × 3.0 × 7 mm3), and RT IS SENSE (acceleration factor 9.9-12, TR 42 ms, 2.0 × 2.0 × 7 mm3). We performed quantitative LV functional analysis in sinus rhythm (SR) patients and qualitatively scored image quality, noise and artefact using a 5-point Likert scale in the complete cohort and AF and SR subgroups. There was no difference between LV functional parameters between acquisitions in SR patients. RT IS SENSE short-axis image quality was superior to SEG (4.5 ± 0.6 vs. 3.9 ± 1.1, p = 0.007) and RT (3.8 ± 0.4, p = 0.003). There was reduced artefact in RT IS SENSE compared to SEG (4.4 ± 0.6 vs. 3.8 ± 1.2, p = 0.04), driven by arrhythmia performance. RT IS SENSE short-axis image quality was superior to SEG (4.6 ± 0.5 vs. 3.1 ± 1.0, p < 0.001) in the AF subgroup. Accelerated real-time CMR with iterative sparse SENSE provides excellent clinical performance, especially in patients with AF. • Iterative sparse SENSE significantly accelerates real-time cardiovascular MRI acquisitions. • It provides excellent qualitative and quantitative performance in sinus rhythm patients. • It outperforms standard segmented acquisitions in patients with atrial fibrillation. • It improves the trade-off between temporal and spatial resolution in real-time imaging.

Respiratory motion correction for free-breathing 3D abdominal MRI using CNN-based image registration: a feasibility study.

Free-breathing abdomen imaging requires non-rigid motion registration of unavoidable respiratory motion in three-dimensional undersampled data sets. In this work, we introduce an image registration method based on the convolutional neural network (CNN) to obtain motion-free abdominal images throughout the respiratory cycle. Abdominal data were acquired from 10 volunteers using a 1.5 T MRI system. The respiratory signal was extracted from the central-space spokes, and the acquired data were reordered in three bins according to the corresponding breathing signal. Retrospective image reconstruction of the three near-motion free respiratory phases was performed using non-Cartesian iterative SENSE reconstruction. Then, we trained a CNN to analyse the spatial transform among the different bins. This network could generate the displacement vector field and be applied to perform registration on unseen image pairs. To demonstrate the feasibility of this registration method, we compared the performance of three different registration approaches for accurate image fusion of three bins: non-motion corrected (NMC), local affine registration method (LREG) and CNN. Visualization of coronal images indicated that LREG had caused broken blood vessels, while the vessels of the CNN were sharper and more consecutive. As shown in the sagittal view, compared to NMC and CNN, distorted and blurred liver contours were caused by LREG. At the same time, zoom-in axial images presented that the vessels were delineated more clearly by CNN than LREG. The statistical results of the signal-to-noise ratio, visual score, vessel sharpness and registration time over all volunteers were compared among the NMC, LREG and CNN approaches. The SNR indicated that the CNN acquired the best image quality (207.42 ± 96.73), which was better than NMC (116.67 ± 44.70) and LREG (187.93 ± 96.68). The image visual score agreed with SNR, marking CNN (3.85 ± 0.12) as the best, followed by LREG (3.43 ± 0.13) and NMC (2.55 ± 0.09). A vessel sharpness assessment yielded similar values between the CNN (0.81 ± 0.03) and LREG (0.80 ± 0.04), differentiating them from the NMC (0.78 ± 0.06). When compared with the LREG-based reconstruction, the CNN-based reconstruction reduces the registration time from 1 h to 1 min. Our preliminary results demonstrate the feasibility of the CNN-based approach, and this scheme outperforms the NMC- and LREG-based methods. Advances in knowledge: This method reduces the registration time from ~1 h to ~1 min, which has promising prospects for clinical use. To the best of our knowledge, this study shows the first convolutional neural network-based registration method to be applied in abdominal images.

Free breathing whole-heart 3D CINE MRI with self-gated Cartesian trajectory

PurposeTo present a method that uses a novel free-running self-gated acquisition to achieve isotropic resolution in whole heart 3D Cartesian cardiac CINE MRI. Material and methods3D cardiac CINE MRI using navigator gating results in long acquisition times. Recently, several frameworks based on self-gated non-Cartesian trajectories have been proposed to accelerate this acquisition. However, non-Cartesian reconstructions are computationally expensive due to gridding, particularly in 3D. In this work, we propose a novel highly efficient self-gated Cartesian approach for 3D cardiac CINE MRI. Acquisition is performed using CArtesian trajectory with Spiral PRofile ordering and Tiny golden angle step for eddy current reduction (so called here CASPR-Tiger). Data is acquired continuously under free breathing (retrospective ECG gating, no preparation pulses interruption) for 4–5min and 4D whole-heart volumes (3D+cardiac phases) with isotropic spatial resolution are reconstructed from all available data using a soft gating technique combined with temporal total variation (TV) constrained iterative SENSE reconstruction. ResultsFor data acquired on eight healthy subjects and three patients, the reconstructed images using the proposed method had good contrast and spatio-temporal variations, correctly recovering diastolic and systolic cardiac phases. Non-significant differences (P>0.05) were observed in cardiac functional measurements obtained with proposed 3D approach and gold standard 2D multi-slice breath-hold acquisition. ConclusionThe proposed approach enables isotropic 3D whole heart Cartesian cardiac CINE MRI in 4 to 5min free breathing acquisition.

Right-ventricular assessment using a segmented cine acquisition employing iterative SENSE reconstruction with spatio-temporal L1 regularization: Initial clinical experience

Background Cardiac MR (CMR) has emerged as the gold standard for assessing right-ventricular (RV) size and function with cine acquisitions. However, dyspnea often accompanies RV dysfunction, limiting patient ability to breathhold and consequently image quality at CMR. The application of a novel iterative reconstruction technique to segmented CMR cine acquisitions may enable higher acceleration factors, shortening image acquisitions, while maintaining image quality for RV assessment. The purpose of this study is to evaluate the clinical utility of a prototype iterative reconstruction algorithm utilizing spatio-temporal L1-regularization (iteratively reconstructed Sparse-SENSE) in the quantitative assessment of RV systolic function.

Evaluation of accelerated real-time CMR using sparse sampling with iterative SENSE reconstruction in patients and volunteers

Background The use of gated CMR can be limited by motion artifacts secondary to cardiac and respiratory motion. Imaging is especially challenging in patients with arrhythmias or those who cannot perform adequate breath-holds. Realtime CMR is a non-gated technique that has been successfully applied in scenarios where standard segmented acquisitions break down. In this study, we sought to accelerate real-time acquisition by using sparse sampling with an iterative SENSE reconstruction.

Biventricular strain analysis at 1.5T cardiac MR imaging: preliminary results in volunteers using an iterative SENSE reconstruction with L1 regularization

Biventricular strain analysis at 1.5T cardiac MR imaging: preliminary results in volunteers using an iterative SENSE reconstruction with L1 regularization

Biventricular strain analysis at 1.5T cardiac MR imaging: preliminary results in volunteers using an iterative SENSE reconstruction with L1 regularization

Background Changes in myocardial strain have been shown to precede onset of systolic dysfunction in patients with cardiomyopathy. Traditionally performed with echocardiography, acoustic windows can limit strain evaluation. Preliminary work has shown good agreement between myocardial strain derived from deformation field analysis at balanced steady-state free-precession (bSSFP) cine MRI and speckle tracking echocardiography. The application of a novel prototype iterative SENSE reconstruction with L1 regularization (IRSENSE) to highly accelerated segmented bSSFP cine acquisitions can maintain or improve the effective temporal resolution with a shorter breath-hold. The purpose of this study is to evaluate left- and right-ventricular stain analysis from bSSFP cine acquisitions using conventional GRAPPA and highly accelerated T-PAT with IR-SENSE. We hypothesize that myocardial strain parameters derived from accelerated T-PAT cine acquisitions with IR-SENSE are similar to those obtained using conventional segmented bSSFP with comparable effective temporal resolutions.

Segmented CMR acquisition with iterative SENSE reconstruction using L1-regularization in the evaluation of right ventricular systolic function

Background Cardiac MR (CMR) has emerged as the gold standard in assessing biventricular size and systolic function with segmented balanced steady-state free-precession (bSSFP) cine acquisitions. The application of a novel prototype iterative reconstruction technique to sparsely undersampled segmented bSSFP cine acquisitions may enable higher acceleration factors while shortening image acquisitions, thus maintaining adequate image quality for quantitative analysis. The purpose of this study is to evaluate the clinical utility of a segmented sparsely sampled 2D CINE imaging technique, with a prototype iterative SENSE reconstruction using L1-regularization, in the quantitative assessment of right ventricular (RV) systolic function.

An improved iterative SENSE reconstruction method

AbstractThe conjugate‐gradient (CG)‐based non‐Cartesian SENSE reconstruction usually exhibits unstable convergence behavior due to the ill conditioning of the generalized encoding matrix (GEM). To overcome this difficulty, an improved iterative SENSE approach is presented. During a so‐called Lanczos iteration process, which is equally efficient as CG, the inversion of GEM can be gradually approximated by calculating inversions of a series of small tridiagonal matrices. In this fashion, inner regularization can be incorporated into the reconstruction without touching the iteration process. The degree of regularization can be determined based on the eigenvalue information provided by the Lanczos process. With inner regularization adaptively applied for every iteration vector, the convergence behavior of iterative SENSE can be significantly improved and noise amplification can be avoided. The feasibility of this novel iterative SENSE technique is demonstrated by radial and spiral MRI experiments. © 2007 Wiley Periodicals, Inc. Concepts Magn Reson Part B (Magn Reson Engineering) 31B: 44–50, 2007

Augmented generalized SENSE reconstruction to correct for rigid body motion

The correction of motion artifacts continues to be a significant problem in MRI. In the case of uncooperative patients, such as children, or patients who are unable to remain stationary, the accurate determination and correction of motion artifacts becomes a very important prerequisite for achieving good image quality. The application of conventional motion-correction strategies often produces inconsistencies in k-space data. As a result, significant residual artifacts can persist. In this work a formalism is introduced for parallel imaging in the presence of motion. The proposed method can improve overall image quality because it diminishes k-space inconsistencies by exploiting the complementary image encoding capacity of individual receiver coils. Specifically, an augmented version of an iterative SENSE reconstruction is used as a means of synthesizing the missing data in k-space. Motion is determined from low-resolution navigator images that are coregistered by an automatic registration routine. Navigator data can be derived from self-navigating k-space trajectories or in combination with other navigation schemes that estimate patient motion. This correction method is demonstrated by interleaved spiral images collected from volunteers. Conventional spiral scans and scans corrected with proposed techniques are shown, and the results illustrate the capacity of this new correction approach.

Convergence behavior of iterative SENSE reconstruction with non‐Cartesian trajectories

In non-Cartesian SENSE reconstruction based on the conjugate gradient (CG) iteration method, the iteration very often exhibits a "semi-convergence" behavior, which can be characterized as initial convergence toward the exact solution and later divergence. This phenomenon causes difficulties in automatic implementation of this reconstruction strategy. In this study, the convergence behavior of the iterative SENSE reconstruction is analyzed based on the mathematical principle of the CG method. It is revealed that the semi-convergence behavior is caused by the ill-conditioning of the underlying generalized encoding matrix (GEM) and the intrinsic regularization effect of CG iteration. From the perspective of regularization, each iteration vector is a regularized solution and the number of iterations plays the role of the regularization parameter. Therefore, the iteration count controls the compromise between the SNR and the residual aliasing artifact. Based on this theory, suggestions with respect to the stopping rule for well-behaved reconstructions are provided. Simulated radial imaging and in vivo spiral imaging are performed to demonstrate the theoretical analysis on the semi-convergence phenomenon and the stopping criterion. The dependence of convergence behavior on the undersampling rate and the noise level in samples is also qualitatively investigated.