More Like this

1. Network Accelerated Motion Estimation and Reduction (NAMER): Convolutional neural network guided retrospective motion correction using a separable motion model

   https://doi.org/10.1002/mrm.27771  

   Haskell, Melissa W. ; Cauley, Stephen F. ; Bilgic, Berkin ; Hossbach, Julian ; Splitthoff, Daniel N. ; Pfeuffer, Josef ; Setsompop, Kawin ; Wald, Lawrence L. ( May 2019 , Magnetic Resonance in Medicine)

   We introduce and validate a scalable retrospective motion correction technique for brain imaging that incorporates a machine learning component into a model‐based motion minimization.

   A convolutional neural network (CNN) trained to remove motion artifacts from 2D T₂‐weighted rapid acquisition with refocused echoes (RARE) images is introduced into a model‐based data‐consistency optimization to jointly search for 2D motion parameters and the uncorrupted image. Our separable motion model allows for efficient intrashot (line‐by‐line) motion correction of highly corrupted shots, as opposed to previous methods which do not scale well with this refinement of the motion model. Final image generation incorporates the motion parameters within a model‐based image reconstruction. The method is tested in simulations and in vivo motion experiments of in‐plane motion corruption.

   While the convolutional neural network alone provides some motion mitigation (at the expense of introduced blurring), allowing it to guide the iterative joint‐optimization both improves the search convergence and renders the joint‐optimization separable. This enables rapid mitigation within shots in addition to between shots. For 2D in‐plane motion correction experiments, the result is a significant reduction of both image space root mean square error in simulations, and a reduction of motion artifacts in the in vivo motion tests.

   The separability and convergence improvements afforded by the combined convolutional neural network+model‐based method shows the potential for meaningful postacquisition motion mitigation in clinical MRI.

    

   more » « less
2. Coil sketching for computationally efficient MR iterative reconstruction

   https://doi.org/10.1002/mrm.29883  

   Oscanoa, Julio A. ; Ong, Frank ; Iyer, Siddharth S. ; Li, Zhitao ; Sandino, Christopher M. ; Ozturkler, Batu ; Ennis, Daniel B. ; Pilanci, Mert ; Vasanawala, Shreyas S. ( October 2023 , Magnetic Resonance in Medicine)

   Parallel imaging and compressed sensing reconstructions of large MRI datasets often have a prohibitive computational cost that bottlenecks clinical deployment, especially for three‐dimensional (3D) non‐Cartesian acquisitions. One common approach is to reduce the number of coil channels actively used during reconstruction as in coil compression. While effective for Cartesian imaging, coil compression inherently loses signal energy, producing shading artifacts that compromise image quality for 3D non‐Cartesian imaging. We propose coil sketching, a general and versatile method for computationally‐efficient iterative MR image reconstruction.

   We based our method on randomized sketching algorithms, a type of large‐scale optimization algorithms well established in the fields of machine learning and big data analysis. We adapt the sketching theory to the MRI reconstruction problem via a structured sketching matrix that, similar to coil compression, considers high‐energy virtual coils obtained from principal component analysis. But, unlike coil compression, it also considers random linear combinations of the remaining low‐energy coils, effectively leveraging information from all coils.

   First, we performed ablation experiments to validate the sketching matrix design on both Cartesian and non‐Cartesian datasets. The resulting design yielded both improved computatioanal efficiency and preserved signal‐to‐noise ratio (SNR) as measured by the inverse g‐factor. Then, we verified the efficacy of our approach on high‐dimensional non‐Cartesian 3D cones datasets, where coil sketching yielded up to three‐fold faster reconstructions with equivalent image quality.

   Coil sketching is a general and versatile reconstruction framework for computationally fast and memory‐efficient reconstruction.

    

   more » « less
3. Scan‐specific artifact reduction in k‐space (SPARK) neural networks synergize with physics‐based reconstruction to accelerate MRI

   https://doi.org/10.1002/mrm.29036  

   Arefeen, Yamin ; Beker, Onur ; Cho, Jaejin ; Yu, Heng ; Adalsteinsson, Elfar ; Bilgic, Berkin ( October 2021 , Magnetic Resonance in Medicine)

   To develop a scan‐specific model that estimates and corrects k‐space errors made when reconstructing accelerated MRI data.

   Scan‐specific artifact reduction in k‐space (SPARK) trains a convolutional‐neural‐network to estimate and correct k‐space errors made by an input reconstruction technique by back‐propagating from the mean‐squared‐error loss between an auto‐calibration signal (ACS) and the input technique’s reconstructed ACS. First, SPARK is applied to generalized autocalibrating partially parallel acquisitions (GRAPPA) and demonstrates improved robustness over other scan‐specific models, such as robust artificial‐neural‐networks for k‐space interpolation (RAKI) and residual‐RAKI. Subsequent experiments demonstrate that SPARK synergizes with residual‐RAKI to improve reconstruction performance. SPARK also improves reconstruction quality when applied to advanced acquisition and reconstruction techniques like 2D virtual coil (VC‐) GRAPPA, 2D LORAKS, 3D GRAPPA without an integrated ACS region, and 2D/3D wave‐encoded imaging.

   SPARK yields SSIM improvement and 1.5 – 2× root mean squared error (RMSE) reduction when applied to GRAPPA and improves robustness to ACS size for various acceleration rates in comparison to other scan‐specific techniques. When applied to advanced reconstruction techniques such as residual‐RAKI, 2D VC‐GRAPPA and LORAKS, SPARK achieves up to 20% RMSE improvement. SPARK with 3D GRAPPA also improves RMSE performance by ~2×, SSIM performance, and perceived image quality without a fully sampled ACS region. Finally, SPARK synergizes with non‐Cartesian, 2D and 3D wave‐encoding imaging by reducing RMSE between 20% and 25% and providing qualitative improvements.

   SPARK synergizes with physics‐based acquisition and reconstruction techniques to improve accelerated MRI by training scan‐specific models to estimate and correct reconstruction errors in k‐space.

    

   more » « less
4. Eddy current nulled constrained optimization of isotropic diffusion encoding gradient waveforms

   https://doi.org/10.1002/mrm.27539  

   Yang, Grant ; McNab, Jennifer A. ( October 2018 , Magnetic Resonance in Medicine)

   Isotropic diffusion encoding efficiently encodes additional microstructural information in combination with conventional linear diffusion encoding. However, the gradient‐intensive isotropic diffusion waveforms generate significant eddy currents, which cause image distortions. The purpose of this study is to present a method for designing isotropic diffusion encoding waveforms with intrinsic eddy current nulling.

   Eddy current nulled gradient waveforms were designed using a constrained optimization waveform for a 3T GE Premier MRI system. Encoding waveforms were designed for a variety of eddy current null times and sequence timings to evaluate the achievable b‐value. Waveforms were also tested with both eddy current nulling and concomitant field compensation. Distortion reduction was tested in both phantoms and the in vivo human brain.

   The feasibility of isotropic diffusion encoding with intrinsic correction of eddy current distortion and signal bias from concomitant fields was demonstrated. The constrained optimization algorithm produced gradient waveforms with the specified eddy current null times. The reduction in the achievable diffusion weighting was dependent on the number of eddy current null times. A reduction in the eddy current–induced image distortions was observed in both phantoms and in vivo human subjects.

   The proposed framework allows the design of isotropic diffusion‐encoding sequences with reduced image distortion.

    

   more » « less
5. Whole‐heart coronary MR angiography using a 3D cones phyllotaxis trajectory

   https://doi.org/10.1002/mrm.27475  

   Malavé, Mario O. ; Baron, Corey A. ; Addy, Nii Okai ; Cheng, Joseph Y. ; Yang, Phillip C. ; Hu, Bob S. ; Nishimura, Dwight G. ( October 2018 , Magnetic Resonance in Medicine)

   To develop a 3D cones steady‐state free precession sequence with improved robustness to respiratory motion while mitigating eddy current artifacts for free‐breathing whole‐heart coronary magnetic resonance angiography.

   The proposed sequence collects cone interleaves using a phyllotaxis pattern, which allows for more distributed k‐space sampling for each heartbeat compared to a typical sequential collection pattern. A Fibonacci number of segments is chosen to minimize eddy current effects with the trade‐off of an increased number of acquisition heartbeats. For verification, phyllotaxis‐cones is compared to sequential‐cones through simulations, phantom studies, andin vivocoronary scans with 8 subjects using 2D image‐based navigators for retrospective motion correction.

   Simulated point spread functions and moving phantom results show less coherent motion artifacts for phyllotaxis‐cones compared to sequential‐cones. Assessment of the right and left coronary arteries using reader scores and the image edge profile acutance vessel sharpness metric indicate superior image quality and sharpness for phyllotaxis‐cones.

   Phyllotaxis 3D cones results in improved qualitative image scores and coronary vessel sharpness for free‐breathing whole‐heart coronary magnetic resonance angiography compared to standard sequential ordering when using a steady‐state free precession sequence.

    

   more » « less