View Presentation Video

Aniket Tolpadi¹, Upasana Bharadwaj¹, Kenneth Gao¹, Rupsa Bhattacharjee¹, Felix Gassert¹, Johanna Luitjens¹, Jan Nikolas Morshuis^2,3, Paul Fischer², Matthias Hein², Christian F. Baumgartner², Artem Razumov⁴, Dmitry Dylov⁴, Quintin van Lohuizen⁵, Stefan Fransen⁵, Xiaoxia Zhang⁶, Radhika Tibrewaka⁶, Hector Lise de Moura⁶, Kangning Liu⁶, Marcelo Zibetti⁶, Ravinder Regatte⁶, Sharmila Majumdar¹, and Valentina Pedoia^1
1Radiology and Biomedical Imaging, UCSF, San Francisco, CA, United States, ²Cluster of Excellence Machine Learning, University of Tübingen, Tübingen, Germany, ³International Max Planck Research School for Intelligent Systems, Tübingen, Germany, ⁴Skolkovo Institute of Science and Technology, Moscow, Russian Federation, ⁵Department of Radiology, University Medical Center Groningen, Groningen, Netherlands, ⁶Center for Advanced Imaging Innovation and Research, New York University Grossman School of Medicine, New York, NY, United States

Keywords: Image Reconstruction, MSK

Image reconstruction and downstream tasks have typically been treated independently by the image processing community, but we hypothesized performing them end-to-end could facilitate further optimization. To these ends, UCSF organized the K2S challenge, where challenge participants were tasked with segmenting bone and cartilage from 8X undersampled knee MRI acquisitions. Top challenge submissions produced high-quality segmentations maintaining fidelity to ground truth, but strong reconstruction performance proved not to be required for accurate tissue segmentation, and there was no correlation between reconstruction and segmentation performance. This challenge showed reconstruction algorithms can be optimized for downstream tasks in an end-to-end fashion.

View Presentation Video

Keywords: Sparse & Low-Rank Models, Quantitative Imaging

Subspace or low-rank models have been demonstrated useful in producing efficient representations and reconstructions for high dimensional imaging problems. In this study, we propose an unsupervised learning-based approach by generalizing the subspace model using a deep generative network. The generative subspace model can then be incorporated into the physics-based reconstruction formalism. The network parameters can be self-trained by minimizing the cost function with the flexibility to integrate with conventional constraints. We demonstrated the effectiveness of the proposed method over standard linear subspace and deep image prior models using in vivo T2 mapping dataset.

View Presentation Video

Keywords: Sparse & Low-Rank Models, Cardiovascular, Motion estimation; Motion correction; Low-Rank & Sparse;Time-resolved imaging

In this work we present a time-resolved (ungated, free-breathing) dynamic cardiac MRI method which disentangles and jointly estimates motion-fields and time-varying contrast images. Different sources of intensity variations are represented with L+S decompositions and motion-fields, that can naturally provide the information required for downstream tasks such as perfusion (contrast dynamics) and myocardial strain analysis (tissue dynamics). Results indicate the feasibility of disentangling the different sources of temporal  intensity variations such as respiratory motion or contrast enhancement. Moreover, no breath holds nor ECG triggering/sorting are required since the time-resolved framework can simultaneously resolve the many overlapping dynamics from a continuous data-stream. 

View Presentation Video

Keywords: Signal Representations, Parallel Imaging

We develop a novel parallel imaging reconstruction method by extracting null-subspace bases of calibration data/matrix to calculate image-domain spatial nulling maps that contain both coil sensitivity and finite image support information. Images are reconstructed by solving a nulling system formed by multi-channel spatial nulling maps without any masking-related procedure (i.e., in existing SENSE/ESPIRiT for minimizing noise propagation). We demonstrate this method with 2D brain, knee and cardiac data under various conditions, yielding results highly comparable to ESPIRiT with optimal manual masking. Our proposed hybrid-domain reconstruction method is efficient, and more robust than existing ESPIRiT for parallel imaging in practice.

View Presentation Video

Felix Glang¹, Praveen Iyyappan Valsala¹, Anton V. Nikulin^1,2, Nikolai Avdievich¹, Theodor Steffen¹, and Klaus Scheffler^1,2
1High-field Magnetic Resonance Center, Max Planck Institute for Biological Cybernetics, Tübingen, Germany, ²Department of Biomedical Magnetic Resonance, Eberhard Karls University Tübingen, Tübingen, Germany

Keywords: Parallel Imaging, High-Field MRI

Non-Cartesian trajectories have several advantageous properties, including favorable parallel imaging performance. Recently, a novel concept of improving parallel imaging by electronically modulated time-varying receive sensitivities has been introduced. In the present work, we investigate if these two concepts can be combined, i.e., how dynamic sensitivity modulation impacts non-Cartesian parallel imaging reconstruction. To that end, numerical experiments are performed based on data from the novel reconfigurable coil array. We find improvement in convergence, reconstruction error and noise amplification due to rapid sensitivity modulation for radial, spiral, and Cartesian trajectories, implying the potential of this method for advanced encoding and reconstruction schemes.

Malte Riedel¹, Thomas Ulrich², and Klaas Pruessmann^2
1ETH Zurich, Zurich, Switzerland, ²Institute for Biomedical Engineering, ETH Zurich and University of Zurich, Zurich, Switzerland

Keywords: Image Reconstruction, Brain

Head motion is accompanied by a multitude of second order motion effects like changing coil sensitivity maps, background field inhomogeneities, and susceptibility-induced fields. While scan geometries and shims can be corrected in real-time, the scanner has no means to counteract changes of the coil sensitivity maps or the susceptibility-induced fields requiring data-driven retrospective motion correction algorithms for these issues. Randomized sampling can further be exploited to improve the problem conditioning of the parameter estimations on temporal sub-segments of the scan. In this work, we evaluate the combination prospective motion navigation with randomized sampling in a pose-dependent field correction algorithm.

View Presentation Video

Xi Chen¹, Wenchuan Wu¹, and Mark chiew^1,2,3
1Wellcome Centre for Integrative Neuroimaging, FMRIB, Nuffield Department of Clinical Neurosciences, University of Oxford, Oxford, United Kingdom, ²Department of Medical Biophysics, University of Toronto, Toronto, ON, Canada, ³Physical Sciences, Sunnybrook Research Institute, Toronto, ON, Canada

Keywords: Sparse & Low-Rank Models, Motion Correction, fMRI; 3D EPI

Structured low-rank (SLR) reconstruction has been successfully used in 3D multi-shot EPI for fMRI to improve its robustness to inter-shot phase variations. This work proposed a motion compensated structured low-rank (mcSLR) reconstruction, which further improves the robustness of 3D multi-shot EPI by joint modelling of both inter-shot motion and phase variations.

View Presentation Video

Tianle Cao^1,2, Yibin Xie¹, Debiao Li¹, and Anthony G. Christodoulou^1
1Cedars Sinai Medical Center, Los Angeles, CA, United States, ²University of California, Los Angeles, Los Angeles, CA, United States

Keywords: Sparse & Low-Rank Models, Sparse & Low-Rank Models

Low-rank tensor modelling is promising for multi-dimensional MR imaging. In this work, we developed a new low-rank tensor reconstruction approach using alternating minimization of spatial and temporal bases from the whole k-t space data instead of from split subsets of data. The approach was evaluated for 2D motion-resolved myocardial T1/T2/T2*/fat-fraction mapping and  could potentially be used for imporving reconstruction quality and/or further reducing scan time.

View Presentation Video

Yiming Dong¹, Kirsten Koolstra², Ziyu Li³, Matthias J.P. van Osch¹, and Peter Börnert^1,4
1C.J. Gorter MRI Center, Department of Radiology, LUMC, Leiden, Netherlands, ²Philips, Best, Netherlands, ³Wellcome Centre for Integrative Neuroimaging, FMRIB, Nuffield Department of Clinical Neurosciences, University of Oxford, Oxford, United Kingdom, ⁴Philips Research, Hamburg, Germany

Keywords: Image Reconstruction, Diffusion/other diffusion imaging techniques

Multi-shot EPI readout-approaches provide high spatial resolution at reduced geometric distortions and improved SNR in diffusion weighted imaging (DWI). As a specific challenge, physiological motion induces shot-to-shot phase variations and needs specific handling, e.g., using additionally measured phase navigators, data-driven phase estimation and/or low-rank regularizations. Furthermore, good fat-suppression is also needed in DWI, making the use of chemical-shift encoding interesting. In this work, a structured low-rank-based water/fat separation pipeline is proposed to jointly estimate water/fat images while correcting motion-induced phase variations with improved time efficiency. In-vivo examples from different anatomies demonstrate improved water/fat separation compared to conventional approaches.

View Presentation Video

Shuai Huang¹, James J. Lah¹, Jason W. Allen¹, and Deqiang Qiu^1
1Emory University, Atlanta, GA, United States

Keywords: Image Reconstruction, Quantitative Susceptibility mapping

We propose a robust Bayesian approach with built-in parameter estimation for quantitative susceptibility mapping (QSM). From a Bayesian perspective, wavelet coefficients of the susceptibility map are modeled by Laplace distribution. Noise is modeled by a two-component Gaussian-mixture distribution, where the second component is reserved to model the noise outliers. The susceptibility map and distribution parameters are jointly recovered using approximate message passing (AMP). The proposed approach achieves better performance in challenging cases of brain hemorrhage and calcification. It automatically estimates the parameters, which avoids subjective bias from the usual visual-tuning step of in vivo reconstruction.

View Presentation Video

Shen Zhao¹, Junyu Wang¹, and Michael Salerno^1
1Cardiovascular Medicine, Stanford University, Stanford, CA, United States

Keywords: Image Reconstruction, Image Reconstruction, SMS

Slice leakage is a significant issue in simultaneous multislice (SMS) imaging. In this work, we extend POMP and develop a new efficient accelerated SMS acquisition technique: Simultaneous Multislice Imaging via Linear phase modulated Extended field of view (SMILE). SMILE transforms the SMS problem into a 2D imaging task and enables direct implementation of 2D non-SMS-only reconstruction algorithms. SMILE increases the sampling degree of freedom by a factor of the number of slices and could theoretically avoid the significant "slice-leakage" issue.       

View Presentation Video

Yahang Li^1,2, Zepeng Wang ^1,2, Aaron Anderson ^2,3, Ruiyang Zhao ^2,4, Paul Arnold ^2,3, Graham Huesmann ^2,3,5, and Fan Lam ^1,2,4
1Department of Bioengineering, University of Illinois Urbana-Champaign, Urbana, IL, United States, ²Beckman Institute for Advanced Science and Technology, Urbaba, IL, United States, ³Neuroscience Institute, Carle Foundation Hospital, Urbaba, IL, United States, ⁴Department of Electrical and Computer Engineering, University of Illinois Urbana-Champaign, Urbana, IL, United States, ⁵School of Molecular and Cellular Biology, University of Illinois Urbana-Champaign, Urbana, IL, United States

Keywords: Image Reconstruction, Spectroscopy

A computationally efficient MRSI reconstruction method is presented. The proposed problem formulation integrates a subspace model of the high-dimensional spatiotemporal function (SPICE) and a network-based learned projector on to a low-dimensional manifold of generic spectroscopic signals. The subspace representation allows for more flexible spatiotemporal sampling designs than using nonlinear manifold constraint alone, while the manifold constraint effectively regularizes the subspace fitting, especially at higher orders. An efficient algorithm is designed to solve the optimization problem. The benefits of the proposed synergy have been demonstrated using simulations as well as experimental ³¹P and ¹H-MRSI data. 

View Presentation Video

Ziwen Ke¹, Yibo Zhao^2,3, Yudu Li^2,4, Rong Guo^2,5, Zhi-Pei Liang^2,3, and Yao Li^1
1School of Biomedical Engineering, Shanghai Jiao Tong University, Shanghai, China, ²Beckman Institute for Advanced Science and Technology, University of Illinois at Urbana-Champaign, Urbana, IL, United States, ³Department of Electrical and Computer Engineering, University of Illinois at Urbana-Champaign, Urbana, IL, United States, ⁴National Center for Supercomputing Applications, University of Illinois at Urbana-Champaign, Urbana, IL, United States, ⁵Siemens Medical Solutions USA, Urbana, IL, United States

Keywords: Image Reconstruction, Quantitative Susceptibility mapping

The SPICE technique has demonstrated a unique capability of simultaneous QSM/MRSI. To achieve fast high-resolution QSM, highly sparse sampling of (k,t)-space is used in SPICE data acquisition, which poses a significant challenge in image reconstruction. In this work, we solved this problem using a subspace-assisted parallel imaging technique with learned image priors. The proposed method has been validated using experimental data, producing high-quality QSM maps from the unsuppressed water signals of ¹H-MRSI scans.

Liyuan Liang^1,2, Mei-Lan Chu³, Nan-Kuei Chen^4,5, and Hing-Chiu Chang^1,2
1Department of Biomedical Engineering, The Chinese University of Hong Kong, Shatin, Hong Kong, ²Multi-Scale Medical Robotics Center, Shatin, Hong Kong, ³Graduate Institute of Biomedical Electronics and Bioinformatics, National Taiwan University, Taipei, Taiwan, ⁴Department of Biomedical Engineering, University of Arizona, Tucson, AZ, United States, ⁵Brain Imaging and Analysis Center, Duke University Medical Center, Durham, NC, United States

Keywords: Image Reconstruction, Diffusion Tensor Imaging

Recently, a novel self-navigated multi-shot EPI data acquisition scheme called iblocks-EPI was proposed to achieve high-resolution DWI images with further reduction of distortion compared to the conventional interleaved EPI. However, the total acquisition time of iblocks-EPI is several times longer than the conventional interleaved EPI because it requires multiple patch patterns for covering a complete k space. Consequently, it is time-consuming and impractical for routine acquisition of DTI data. In this work we propose a view-sharing iblocks-DTI scheme which can substantially reduce the scan time for iblocks-DTI acquisition while providing accurate DTI tensor calculation.

View Presentation Video

Philip Schaten¹ and Martin Uecker^1
1Institute of Biomedical Imaging, Graz University of Technology, Graz, Austria

Keywords: Image Reconstruction, Software Tools

The open-source software tool BART provides various state-of-the-art image reconstruction algorithms for MRI. We add an interface for streaming image and k-space data to BART, which should facilitate the application of modern image reconstruction to real-time MRI. Our implementation matches the requirements of interventional MRI in terms of frame rate and latency.

Furthermore, we present an intuitive way of interactive slice positioning for MR-guided Interventions using a game controller featuring an inertial measurement unit.