CG-SENSE revisited: Results from the first ISMRM reproducibility challenge

Prospective acceleration of whole-brain CEST imaging by golden-angle view ordering in Cartesian coordinates and joint k-space and image-space parallel imaging (KIPI)

PURPOSE

To prospectively accelerate whole-brain CEST acquisition by joint k-space and image-space parallel imaging (KIPI) with a proposed golden-angle view ordering technique (GAVOT) in Cartesian coordinates.

THEORY AND METHODS

The T2-decay effect will vary across frames with variable acceleration factors (AF) in the prospective acquisition using sequences with long echo trains. The GAVOT method uses a subset strategy to eliminate the T2-decay inconsistency, where all frames use a subset of shots from the calibration frame to form their k-space view ordering. The golden-angle rule is adapted to ensure uniform k-space coverage for arbitrary AFs in Cartesian coordinates. Phantom and in vivo studies were conducted on a 3 T scanner.

RESULTS

The GAVOT view ordering yielded a higher g-factor than conventional uniformly centric ordering, whereas the noise propagation in amide proton transfer (APT) weighted images was similar between different view ordering. Compared to centric ordering, GAVOT successfully eliminated the T2-decay inconsistency across all frames, resulting in fewer image artifacts for both KIPI and conventional parallel imaging methods. The synergy of GAVOT and KIPI mitigated strong aliasing artifacts and achieved high-quality reconstruction of prospective variable-AF datasets. GAVOT-KIPI reduced the scan time to 2.1 min for whole-brain APT weighted imaging and 4.7 min for quantitative APT signal (APT#) mapping.

CONCLUSION

GAVOT makes the prospective variable AF strategy flexible and practical, and, in conjunction with KIPI, ensures high-quality reconstruction from highly undersampled data, facilitating the clinical translation of whole-brain CEST imaging.

MRI at low field: A review of software solutions for improving SNR

Low magnetic field magnetic resonance imaging (MRI) (B 0 $$ {B}_0 $$  < 1 T) is regaining interest in the magnetic resonance (MR) community as a complementary, more flexible, and cost-effective approach to MRI diagnosis. Yet, the impaired signal-to-noise ratio (SNR) per square root of time, or SNR efficiency, leading in turn to prolonged acquisition times, still challenges its relevance at the clinical level. To address this, researchers investigate various hardware and software solutions to improve SNR efficiency at low field, including the leveraging of latest advances in computing hardware. However, there may not be a single recipe for improving SNR at low field, and it is key to embrace the challenges and limitations of each proposed solution. In other words, suitable solutions depend on the final objective or application envisioned for a low-field scanner and, more importantly, on the characteristics of a specific lowB 0 $$ {B}_0 $$ field. In this review, we aim to provide an overview on software solutions to improve SNR efficiency at low field. First, we cover techniques for efficient k-space sampling and reconstruction. Then, we present post-acquisition techniques that enhance MR images such as denoising and super-resolution. In addition, we summarize recently introduced electromagnetic interference cancellation approaches showing great promises when operating in shielding-free environments. Finally, we discuss the advantages and limitations of these approaches that could provide directions for future applications.

Coil sketching for computationally efficient MR iterative reconstruction

PURPOSE

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.

THEORY AND METHODS

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.

RESULTS

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.

CONCLUSION

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

Digital reference objects for evaluating algorithm performance in MR image formation

PURPOSE

Digital Reference Objects (DROs) are mathematical phantoms that can serve as a basis for evaluating MR image quality (IQ) in an objective way. Their main purpose is to facilitate the establishment of fully automated and perfectly reproducible IQ metrics to objectively compare different algorithms in MR image formation in a standardized manner. They also allow to re-build parts of standard phantoms.

METHODS

We sample DROs directly in k-space, using analytical formulas for the continuous Fourier transform of primitive shapes. We demonstrate this DRO approach by applying a state-of-the-art CNN-based denoising algorithm that is robust to varying noise levels to noisy images of the resolution section of the well-known ACR phantom for IQ assessment, reconstructed from both measured and simulated k-space data.

RESULTS

Applying the CNN-based denoising algorithm to the measured and simulated version of the ACR phantom resolution section produced virtually identical results, as confirmed by visual and quantitative comparison.

CONCLUSIONS

DROs can help guide technology selection during the development of new algorithms in MR image formation, e.g., via deep learning. This could be an important step towards reproducible MR image formation.

Model-based Deep Learning Reconstruction Using a Folded Image Training Strategy for Abdominal 3D T1-weighted Imaging

PURPOSE

To evaluate the feasibility of folded image training strategy (FITS) and the quality of images reconstructed using the improved model-based deep learning (iMoDL) network trained with FITS (FITS-iMoDL) for abdominal MR imaging.

METHODS

This retrospective study included abdominal 3D T1-weighted images of 122 patients. In the experimental analyses, peak SNR (PSNR) and structure similarity index (SSIM) of images reconstructed with FITS-iMoDL were compared with those with the following reconstruction methods: conventional model-based deep learning (conv-MoDL), MoDL trained with FITS (FITS-MoDL), total variation regularized compressed sensing (CS), and parallel imaging (CG-SENSE). In the clinical analysis, SNR and image contrast were measured on the reference, FITS-iMoDL, and CS images. Three radiologists evaluated the image quality using a 5-point scale to determine the mean opinion score (MOS).

RESULTS

The PSNR of FITS-iMoDL was significantly higher than that of FITS-MoDL, conv-MoDL, CS, and CG-SENSE (P < 0.001). The SSIM of FITS-iMoDL was significantly higher than those of the others (P < 0.001), except for FITS-MoDL (P = 0.056). In the clinical analysis, the SNR of FITS-iMoDL was significantly higher than that of the reference and CS (P < 0.0001). Image contrast was equivalent within an equivalence margin of 10% among these three image sets (P < 0.0001). MOS was significantly improved in FITS-iMoDL (P < 0.001) compared with CS images in terms of liver edge and vessels conspicuity, lesion depiction, artifacts, blurring, and overall image quality.

CONCLUSION

The proposed method, FITS-iMoDL, allowed a deeper MoDL reconstruction network without increasing memory consumption and improved image quality on abdominal 3D T1-weighted imaging compared with CS images.

Time-Resolved 3D cardiopulmonary MRI reconstruction using spatial transformer network

The accurate visualization and assessment of the complex cardiac and pulmonary structures in 3D is critical for the diagnosis and treatment of cardiovascular and respiratory disorders. Conventional 3D cardiac magnetic resonance imaging (MRI) techniques suffer from long acquisition times, motion artifacts, and limited spatiotemporal resolution. This study proposes a novel time-resolved 3D cardiopulmonary MRI reconstruction method based on spatial transformer networks (STNs) to reconstruct the 3D cardiopulmonary MRI acquired using 3D center-out radial ultra-short echo time (UTE) sequences. The proposed reconstruction method employed an STN-based deep learning framework, which used a combination of data-processing, grid generator, and sampler. The reconstructed 3D images were compared against the start-of-the-art time-resolved reconstruction method. The results showed that the proposed time-resolved 3D cardiopulmonary MRI reconstruction using STNs offers a robust and efficient approach to obtain high-quality images. This method effectively overcomes the limitations of conventional 3D cardiac MRI techniques and has the potential to improve the diagnosis and treatment planning of cardiopulmonary disorders.

Stochastic optimization of three-dimensional non-Cartesian sampling trajectory

PURPOSE

Optimizing three-dimensional (3D) k-space sampling trajectories is important for efficient MRI yet presents a challenging computational problem. This work proposes a generalized framework for optimizing 3D non-Cartesian sampling patterns via data-driven optimization.

METHODS

We built a differentiable simulation model to enable gradient-based methods for sampling trajectory optimization. The algorithm can simultaneously optimize multiple properties of sampling patterns, including image quality, hardware constraints (maximum slew rate and gradient strength), reduced peripheral nerve stimulation (PNS), and parameter-weighted contrast. The proposed method can either optimize the gradient waveform (spline-based freeform optimization) or optimize properties of given sampling trajectories (such as the rotation angle of radial trajectories). Notably, the method can optimize sampling trajectories synergistically with either model-based or learning-based reconstruction methods. We proposed several strategies to alleviate the severe nonconvexity and huge computation demand posed by the large scale. The corresponding code is available as an open-source toolbox.

RESULTS

We applied the optimized trajectory to multiple applications including structural and functional imaging. In the simulation studies, the image quality of a 3D kooshball trajectory was improved from 0.29 to 0.22 (NRMSE) with Stochastic optimization framework for 3D NOn-Cartesian samPling trajectorY (SNOPY) optimization. In the prospective studies, by optimizing the rotation angles of a stack-of-stars (SOS) trajectory, SNOPY reduced the NRMSE of reconstructed images from 1.19 to 0.97 compared to the best empirical method (RSOS-GR). Optimizing the gradient waveform of a rotational EPI trajectory improved participants' rating of the PNS from "strong" to "mild."

CONCLUSION

SNOPY provides an efficient data-driven and optimization-based method to tailor non-Cartesian sampling trajectories.

Efficient Approximation of Jacobian Matrices Involving a Non-Uniform Fast Fourier Transform (NUFFT)

There is growing interest in learning Fourier domain sampling strategies (particularly for magnetic resonance imaging, MRI) using optimization approaches. For non-Cartesian sampling, the system models typically involve non-uniform fast Fourier transform (NUFFT) operations. Commonly used NUFFT algorithms contain frequency domain interpolation, which is not differentiable with respect to the sampling pattern, complicating the use of gradient methods. This paper describes an efficient and accurate approach for computing approximate gradients involving NUFFTs. Multiple numerical experiments validate the improved accuracy and efficiency of the proposed approximation. As an application to computational imaging, the NUFFT Jacobians were used to optimize non-Cartesian MRI sampling trajectories via data-driven stochastic optimization. Specifically, the sampling patterns were learned with respect to various model-based image reconstruction (MBIR) algorithms. The proposed approach enables sampling optimization for image sizes that are infeasible with standard auto-differentiation methods due to memory limits. The synergistic acquisition and reconstruction design leads to remarkably improved image quality. In fact, we show that model-based image reconstruction methods with suitably optimized imaging parameters can perform nearly as well as CNN-based methods.

Open-source MR imaging and reconstruction workflow

PURPOSE

This work presents an end-to-end open-source MR imaging workflow. It is highly flexible in rapid prototyping across the whole imaging process and integrates vendor-independent openly available tools. The whole workflow can be shared and executed on different MR platforms. It is also integrated in the JEMRIS simulation framework, which makes it possible to generate simulated data from the same sequence that runs on the MRI scanner using the same pipeline for image reconstruction.

METHODS

MRI sequences can be designed in Python or JEMRIS using the Pulseq framework, allowing simplified integration of new sequence design tools. During the sequence design process, acquisition metadata required for reconstruction is stored in the MR raw data format. Data acquisition is possible on MRI scanners supported by Pulseq and in simulations through JEMRIS. An image reconstruction and postprocessing pipeline was implemented into a Python server that allows real-time processing of data as it is being acquired. The Berkeley Advanced Reconstruction Toolbox is integrated into this framework for image reconstruction. The reconstruction pipeline supports online integration through a vendor-dependent interface.

RESULTS

The flexibility of the workflow is demonstrated with different examples, containing 3D parallel imaging with controlled aliasing in volumetric parallel imaging (CAIPIRINHA) acceleration, spiral imaging, and B₀ mapping. All sequences, data, and the corresponding processing pipelines are publicly available.

CONCLUSION

The proposed workflow is highly flexible and allows integration of advanced tools at all stages of the imaging process. All parts of this workflow are open-source, simplifying collaboration across different MR platforms or sites and improving reproducibility of results.

Open and reproducible neuroimaging: From study inception to publication

Empirical observations of how labs conduct research indicate that the adoption rate of open practices for transparent, reproducible, and collaborative science remains in its infancy. This is at odds with the overwhelming evidence for the necessity of these practices and their benefits for individual researchers, scientific progress, and society in general. To date, information required for implementing open science practices throughout the different steps of a research project is scattered among many different sources. Even experienced researchers in the topic find it hard to navigate the ecosystem of tools and to make sustainable choices. Here, we provide an integrated overview of community-developed resources that can support collaborative, open, reproducible, replicable, robust and generalizable neuroimaging throughout the entire research cycle from inception to publication and across different neuroimaging modalities. We review tools and practices supporting study inception and planning, data acquisition, research data management, data processing and analysis, and research dissemination. An online version of this resource can be found at https://oreoni.github.io. We believe it will prove helpful for researchers and institutions to make a successful and sustainable move towards open and reproducible science and to eventually take an active role in its future development.

qMRI-BIDS: An extension to the brain imaging data structure for quantitative magnetic resonance imaging data

The Brain Imaging Data Structure (BIDS) established community consensus on the organization of data and metadata for several neuroimaging modalities. Traditionally, BIDS had a strong focus on functional magnetic resonance imaging (MRI) datasets and lacked guidance on how to store multimodal structural MRI datasets. Here, we present and describe the BIDS Extension Proposal 001 (BEP001), which adds a range of quantitative MRI (qMRI) applications to the BIDS. In general, the aim of qMRI is to characterize brain microstructure by quantifying the physical MR parameters of the tissue via computational, biophysical models. By proposing this new standard, we envision standardization of qMRI through multicenter dissemination of interoperable datasets. This way, BIDS can act as a catalyst of convergence between qMRI methods development and application-driven neuroimaging studies that can help develop quantitative biomarkers for neural tissue characterization. In conclusion, this BIDS extension offers a common ground for developers to exchange novel imaging data and tools, reducing the entrance barrier for qMRI in the field of neuroimaging.

Vendor-neutral sequences and fully transparent workflows improve inter-vendor reproducibility of quantitative MRI

PURPOSE

We developed an end-to-end workflow that starts with a vendor-neutral acquisition and tested the hypothesis that vendor-neutral sequences decrease inter-vendor variability of T1, magnetization transfer ratio (MTR), and magnetization transfer saturation-index (MTsat) measurements.

METHODS

We developed and deployed a vendor-neutral 3D spoiled gradient-echo (SPGR) sequence on three clinical scanners by two MRI vendors. We then acquired T1 maps on the ISMRM-NIST system phantom, as well as T1, MTR, and MTsat maps in three healthy participants. We performed hierarchical shift function analysis in vivo to characterize the differences between scanners when the vendor-neutral sequence is used instead of commercial vendor implementations. Inter-vendor deviations were compared for statistical significance to test the hypothesis.

RESULTS

In the phantom, the vendor-neutral sequence reduced inter-vendor differences from 8% to 19.4% to 0.2% to 5% with an overall accuracy improvement, reducing ground truth T1 deviations from 7% to 11% to 0.2% to 4%. In vivo, we found that the variability between vendors is significantly reduced (p = 0.015) for all maps (T1, MTR, and MTsat) using the vendor-neutral sequence.

CONCLUSION

We conclude that vendor-neutral workflows are feasible and compatible with clinical MRI scanners. The significant reduction of inter-vendor variability using vendor-neutral sequences has important implications for qMRI research and for the reliability of multicenter clinical trials.

Advances in spiral fMRI: A high-resolution study with single-shot acquisition

Spiral fMRI has been put forward as a viable alternative to rectilinear echo-planar imaging, in particular due to its enhanced average k-space speed and thus high acquisition efficiency. This renders spirals attractive for contemporary fMRI applications that require high spatiotemporal resolution, such as laminar or columnar fMRI. However, in practice, spiral fMRI is typically hampered by its reduced robustness and ensuing blurring artifacts, which arise from imperfections in both static and dynamic magnetic fields. Recently, these limitations have been overcome by the concerted application of an expanded signal model that accounts for such field imperfections, and its inversion by iterative image reconstruction. In the challenging ultra-high field environment of 7 Tesla, where field inhomogeneity effects are aggravated, both multi-shot and single-shot 2D spiral imaging at sub-millimeter resolution was demonstrated with high depiction quality and anatomical congruency. In this work, we further these advances towards a time series application of spiral readouts, namely, single-shot spiral BOLD fMRI at 0.8 mm in-plane resolution. We demonstrate that high-resolution spiral fMRI at 7 T is not only feasible, but delivers both excellent image quality, BOLD sensitivity, and spatial specificity of the activation maps, with little artifactual blurring. Furthermore, we show the versatility of the approach with a combined in/out spiral readout at a more typical resolution (1.5 mm), where the high acquisition efficiency allows to acquire two images per shot for improved sensitivity by echo combination.

Results of the 2020 fastMRI Challenge for Machine Learning MR Image Reconstruction

Accelerating MRI scans is one of the principal outstanding problems in the MRI research community. Towards this goal, we hosted the second fastMRI competition targeted towards reconstructing MR images with subsampled k-space data. We provided participants with data from 7,299 clinical brain scans (de-identified via a HIPAA-compliant procedure by NYU Langone Health), holding back the fully-sampled data from 894 of these scans for challenge evaluation purposes. In contrast to the 2019 challenge, we focused our radiologist evaluations on pathological assessment in brain images. We also debuted a new Transfer track that required participants to submit models evaluated on MRI scanners from outside the training set. We received 19 submissions from eight different groups. Results showed one team scoring best in both SSIM scores and qualitative radiologist evaluations. We also performed analysis on alternative metrics to mitigate the effects of background noise and collected feedback from the participants to inform future challenges. Lastly, we identify common failure modes across the submissions, highlighting areas of need for future research in the MRI reconstruction community.

Revealing the mechanisms behind novel auditory stimuli discrimination: An evaluation of silent functional MRI using looping star

Looping Star is a near‐silent, multi‐echo, 3D functional magnetic resonance imaging (fMRI) technique. It reduces acoustic noise by at least 25dBA, with respect to gradient‐recalled echo echo‐planar imaging (GRE‐EPI)‐based fMRI. Looping Star has successfully demonstrated sensitivity to the cerebral blood‐oxygen‐level‐dependent (BOLD) response during block design paradigms but has not been applied to event‐related auditory perception tasks. Demonstrating Looping Star's sensitivity to such tasks could (a) provide new insights into auditory processing studies, (b) minimise the need for invasive ear protection, and (c) facilitate the translation of numerous fMRI studies to investigations in sound‐averse patients. We aimed to demonstrate, for the first time, that multi‐echo Looping Star has sufficient sensitivity to the BOLD response, compared to that of GRE‐EPI, during a well‐established event‐related auditory discrimination paradigm: the “oddball” task. We also present the first quantitative evaluation of Looping Star's test–retest reliability using the intra‐class correlation coefficient. Twelve participants were scanned using single‐echo GRE‐EPI and multi‐echo Looping Star fMRI in two sessions. Random‐effects analyses were performed, evaluating the overall response to tones and differential tone recognition, and intermodality analyses were computed. We found that multi‐echo Looping Star exhibited consistent sensitivity to auditory stimulation relative to GRE‐EPI. However, Looping Star demonstrated lower test–retest reliability in comparison with GRE‐EPI. This could reflect differences in functional sensitivity between the techniques, though further study is necessary with additional cognitive paradigms as varying cognitive strategies between sessions may arise from elimination of acoustic scanner noise.