Compressed Sensing & Sparsity
Click on
to view the abstract pdf and click on
to view the video presentation.
Monday May 9th
Room 710A 
11:00  13:00 
Moderators: 
Michael Lustig and Nicole Seiberlich 
11:00 
64. 
Introduction

11:12 
65. 
ESPIRiT (Efficient
EigenvectorBased L1SPIRiT) for Compressed Sensing Parallel
Imaging  Theoretical Interpretation and Improved Robustness
for Overlapped FOV Prescription
Peng Lai^{1}, Michael Lustig^{2,3},
Shreyas S Vasanawala^{4}, and Anja C.S Brau^{1}
^{1}Global Applied Science Laboratory, GE
Healthcare, Menlo Park, CA, United States, ^{2}Electrical
Engineering, Stanford University, Stanford, CA, United
States, ^{3}Electrical
Engineering and Computer Science, University of
California, Berkeley, CA, United States, ^{4}Radiology,
Stanford University, Stanford, CA, United States
Compressed sensing (CS) parallel imaging (PI) methods,
such as L1SPIRiT, provide better image quality than CS
or PI alone, but requires highly intensive iterative
computation. Efficient L1SPIRiT (ESPIRiT) greatly
reduces the computation intensity based on eigenvector
computations. This work provides a theoretical analysis
of similarities between these two approaches and
demonstrates that they should converge to the same
solution. Based on our analysis, we show the existence
of multiple dominant eigenvectors for overlapped FOV
acquisition, where original ESPIRiT generates
significant artifacts like mSENSE and identify a
solution. Our results based on invivo datasets showed
that the proposed modified ESPIRiT can provide
reconstruction very similar to L1SPIRiT regardless of
FOV overlap. The modified ESPIRiT algorithm is a robust
and computationally efficient solution to CSPI
reconstruction.

11:24 
66. 
Combination of Compressed
Sensing and Parallel Imaging with Respiratory Motion
Correction for HighlyAccelerated FirstPass Cardiac
Perfusion MRI
Ricardo Otazo^{1}, Daniel Kim^{1}, Leon
Axel^{1}, and Daniel K Sodickson^{1}
^{1}Department of Radiology, NYU School of
Medicine, New York, NY, United States
Firstpass cardiac perfusion MRI studies can be highly
accelerated using a combination of compressed sensing
and parallel imaging. However, this method is sensitive
to respiratory motion, which decreases sparsity in the
combined spatial and temporalfrequency domain and
produces temporal blurring in the reconstructed images.
In this work, we present a rigid respiratory motion
correction method for the combination of compressed
sensing and parallel imaging, to highly accelerate
firstpass cardiac perfusion MRI without the need of
strict breathholding.

11:36 
67. 
Entropy aided Kt Group
Sparse SENSE method for highly accelerated dynamic MRI
Muhammad Usman^{1}, Claudia Prieto^{1},
Tobias Schaeffter^{1}, and Philip G. Batchelor^{1}
^{1}Division of Imaging Sciences and Biomedical
Engineering, King's College London, London, United
Kingdom
Over the last few years, the combination of Compressed
sensing (CS) and parallel imaging have been of great
interest to accelerate MRI. For dynamic MRI, Kt sparse
SENSE (Kt SS) has been proposed for combining the CS
based Kt Sparse method with SENSE. Recently, Kt group
sparse method (Kt GS) has been shown to outperform Kt
Sparse for single coil reconstruction, by exploiting the
sparsity and the structure within the sparse
representation (xf space) of dynamic MRI. In this work,
we propose to extend Kt GS to parallel imaging
acquisition in order to achieve higher acceleration
factors by exploiting the spatial sensitive encoding
from multiple coils. This approach has been called Kt
group Sparse SENSE (Kt GSS). In contrast with the
previous singlecoil based Kt GS method for which a
performance parameter is manually optimized for every
frequency encode; we propose an entropy based scheme for
automatic selection of this parameter. Results from
retrospectively undersampled cardiac gated data show
that Kt GSS outperformed Kt sparse SENSE at high
acceleration factors (up to 16 fold).

11:48 
68. 
Improving Compressed
Sensing Parallel Imaging using Autocalibrating Parallel
Imaging Initialization with Variable Density Tiled Random
kspace Sampling
Peng Lai^{1}, Tao Zhang^{2}, Michael
Lustig^{2,3}, Shreyas S Vasanawala^{4},
and Anja C.S Brau^{1}
^{1}Global Applied Science Laboratory, GE
Healthcare, Menlo Park, CA, United States, ^{2}Electrical
Engineering, Stanford University, Stanford, CA, United
States, ^{3}Electrical
Engineering and Computer Science, University of
California, Berkeley, CA, United States, ^{4}Radiology,
Stanford University, Stanford, CA, United States
Compressed sensing (CS) parallel imaging (PI) is
computationally intensive due to its need for iterative
reconstruction. Autocalibrating PI can improve the
initial solution and largely reduce the number of
iterations needed. However, random sampling needed for
CS generates a huge number of synthesis patterns making
PI initialization extremely slow. Also, uniform density
kspace sampling currently used for CSPI is not optimal
in terms of reconstruction accuracy. The purpose of this
work was to develop a new tiledrandom kspace sampling
strategy with the desirable features of 1. incoherent
kspace sampling with a small number of synthesis
patterns and 2. variable density kspace sampling
providing more accurate center kspace reconstruction.
Based on our evaluations on 4 invivo datasets, the
proposed sampling scheme can improve image quality and
reconstruction accuracy compared to conventional
sampling schemes and meanwhile enables fast PI
initialization for CSPI.

12:00 
69. 
Kt Group Sparse using
Intensity Based Clustering
Claudia Prieto^{1}, Muhammad Usman^{1},
Eike Nagel^{1}, Philip Batchelor^{1},
and Tobias Schaeffter^{1}
^{1}Division of Imaging Sciences and Biomedical
Engineering, King's College London, London, United
Kingdom
Kt Group Sparse (kt GS) has been recently introduced
to achieve high acceleration factors in dynamicMRI.
KtGS exploits not just the sparsity of dynamicMRI but
also the spatial group structure of the xf space.
However, it presents two drawbacks: a) an additional
trainingscan is required for group assignment, and b)
the group assignment is based only on the connectivity
of neighbouring pixels using a timeconsuming hard
thresholding scheme. Here we propose to modify kt GS by
using the intensity order, estimated from the same
acquired data, for a more robust group assignment. This
approach has been tested in cine and perfusion cardiac
images with acceleration factors up to 9.

12:12 
70. 
HighFrequency Subband
Compressed Sensing with ARC Parallel Imaging
Kyunghyun Sung^{1}, Anderson N Nnewihe^{1,2},
Bruce L Daniel^{1}, and Brian A Hargreaves^{1}
^{1}Radiology, Stanford University, Stanford,
California, United States, ^{2}Bioengineering,
Stanford University, Stanford, California, United States
Compressed sensing (CS) is a technique that allows
accurate reconstruction of images from a reduced set of
acquired data. Here, we present a new method, which
efficiently combines CS and parallel imaging (PI) by
separating kspace sampling and reconstruction for high
and lowfrequency kspace data. This maximally utilizes
the waveletdomain sparsity and avoids possible CS
failure in low frequency region. This work has been
demonstrated for highresolution 3D breast imaging and
the reconstructed image successfully recovered
lowfrequency content and fine structures with a net
acceleration of 10.8.

12:24 
71. 
Joint Bayesian Compressed
Sensing for Multicontrast Reconstruction
Berkin Bilgic^{1}, Vivek K Goyal^{1},
and Elfar Adalsteinsson^{1,2}
^{1}EECS, MIT, Cambridge, MA, United States, ^{2}HarvardMIT
Division of Health Sciences and Technology, MIT,
Cambridge, MA, United States
Clinical MRI routinely relies on multiple acquisitions
of the same region of interest with several different
contrasts. We present a reconstruction algorithm based
on Bayesian compressed sensing to exploit such
multicontrast acquisitions for accelerated imaging by
jointly reconstructing a set of related images from
undersampled kspace. Our method offers better
performance than when the images are either
reconstructed individually with the algorithm by Lustig
et al., or jointly with a previously proposed method, MFOCUSS.

12:36 
72. 
Location Constrained
Approximate Message Passing (LCAMP) Algorithm for Compressed
Sensing
Kyunghyun Sung^{1}, Bruce L Daniel^{1},
and Brian A Hargreaves^{1}
^{1}Radiology, Stanford University, Stanford,
California, United States
Iterative thresholding methods have been extensively
studied as faster alternatives to convex optimization
for largesized problems in compressed sensing (CS). A
common largesized problem is dynamic contrast enhanced
(DCE) MRI, and the dynamic measurements possess data
redundancies, which can be used to estimate nonzero
signal locations. In this work, we present a novel
iterative thresholding method called LCAMP (Location
Constrained Approximate Message Passing) by adding the
nonzero location assumption and an approximate message
passing term. The method can reduce computational
complexity and improve reconstruction accuracy.

12:48 
73. 
On the Quality Evaluation
for Images reconstructed by Compressed Sensing
Tobias Wech^{1,2}, Daniel Stäb^{1},
André Fischer^{1}, Dietbert Hahn^{1},
and Herbert Köstler^{1}
^{1}Institute of Radiology, University of
Wuerzburg, Wuerzburg, Bavaria, Germany, ^{2}Center
for Applied Medical Imaging, Siemens Corporate Research,
Baltimore, Maryland, United States
Compressed Sensing reconstructions are characterized by
a nonlinear and nonstationary nature of the dedicated
algorithms. Therefore image quality estimation as used
for regular Fourier Imaging is not feasible. The aim of
this work was to develop a workaround that provides a
linear PSFapproximation as well as a validitytest to
control its quality. The workflow was tested on the
example of a sparse temporal difference image of the
human heart and showed a positive result for the
validitytest.

