ISMRM 24th Annual Meeting & Exhibition • 07-13 May 2016 • Singapore

Weekend Educational Course: Image Acquisition & Reconstruction

Skill Level: Intermediate to Advanced

Organizers: Thomas Foo, Ph.D. & N. Jon Shah, Ph.D.

Sunday 08 May 2016

This course will describe the basis for pulse sequence design, including systems calibration, RF pulse design, and motion compensation. Acquisition methods other than cartesian methods will be discussed. Image reconstruction methods that include parallel imaging and compressed sensing will be described. In addition, methods for generation of different MR image contrast from a limited acquisition set will also be described.

Target Audience
Physicists and engineers who wish to acquire an understanding of aspects of MR imaging, including non-cartersian k-space acquisition methods and undersampled image acquisition and reconstruction.

Educational Objectives
Upon completion of this course, participants should be able to:

  • Develop an understanding of the building blocks of an MRI pulse sequence and the necessary system calibration methods;
  • Develop an understanding of motion compensation methods for MRI; and
  • Develop and understand how image acquisition can be speeded up by undersampling and the different image reconstruction approaches.

Moderators: Houchun Hu, Desmond Yeo
      Pulse Sequence and Building Blocks  
RF Pulse Design
Douglas C Noll1
1Biomedical Engineering, University of Michgan, Ann Arbor, MI, United States
Excitation is a necessary process for MRI in order to create observable magnetization to image.  In this work, we develop the basic principles of excitation using the Bloch Equations, written for the rotating frame, which makes it easier to visualize the effects of applied rotating magnetic field used for excitation.  The small tip angle approximation is shown to be useful for understanding slice profile and multidimensional excitation.  Large tip-angles requires different approaches, such as the Shinnar-LeRoux algorithm. Lastly, excitation k-space, similar to k-space for image acquisition, is a concept that can be used to design complicated patterns of excitation.

Systems Calibrations (Bo, B1, Flip Angle Mapping, Shimming)
Lawrence Wald
Prescan: Transmit/Receive Gain Settings, Frequency Calibration
Kevin King
RF transmit/receive gain and transmit/receive frequency must be adjusted for each patient exam. The problem is similar to parameter mapping but over a smaller volume.  Transmit gain accuracy ideally produces the desired flip angle, however B1 field non-uniformity prevents this in practice. The receive gain is ideally set so that the maximum signal does not saturate the A/D converters which would produce shading, and the noise standard deviation is at least one bit to avoid quantization error.  Transmit/receive frequency accuracy is required for accurate localization, good EPI and spiral image quality and for fat suppression pulses to work optimally.

Break & Meet the Teachers
      Motion Control  
Motion Compensation Methods
Christopher J. Hardy1
1GE Global Research
MRI’s relatively long scan times can result in increased vulnerability to motion artifacts, producing degraded image quality, more complex patient workflow, and the need in some cases for patient sedation, restraint, or rescanning. Most commercial scanners employ a range of methods to ameliorate motion problems, including gating, triggering, and respiratory navigation techniques. In addition, a number of new technologies are under investigation. These include advanced two and three dimensional navigator methods, and self-navigation techniques, which correct for motion using the imaging data themselves, without the need for separate motion-tracking sequences. 

External Sensors & Real-Time Compensation
Maximilian Haeberlin1
1Electrical Engineering, Institute for Biomedical Engineering, University of Zurich and ETH Zurich
This talk will provide an overview on current methods in prospective motion correction for head MRI. It includes both optical motion correction methods as well as NMR-based methods. A selection of currently available technologies will be discussed, including moiré phase tracking, self-encoded optical markers, and gradient tones.
      Acquisition Methods  
Non-Cartesian Methods (Radial, Spiral) & Considerations
Xiaohong Joe Zhou1
1Center for MR Research and Department of Radiology University of Illinois at Chicago, Chicago, Illinois, USA
Non-Cartesian k-space sampling is increasingly used in recent years due to its high efficiency in k-space traversal, robustness against motion, reduced acoustic noise, and/or the ability to achieve ultra-short TE for visualizing tissues with short T2’s or detecting non-proton signals. A number of physics and engineering challenges, however, are present in implementing Non-Cartesian sampling strategies. This presentation will review the advantages and disadvantages of non-Cartesian k-space sampling by focusing on three pulse sequences: SPIRAL, RADIAL (or projection acquisition), and PROPELLER. For each method, the essential pulse sequence design will be described, followed by its variations and implementation. A strong emphasis is placed on practical issues, such as point-spread function, gradient waveform design, image artifacts, and image quality improvement. The presentation will conclude by highlighting a number of emerging applications enabled by non-Cartesian k-space sampling strategies.
Lunch & Meet the Teachers
      Image Reconstruction  
Reconstruction of Non-Cartesian k-Space Data
Gigi Galiana1
1Yale University, New Haven, CT, United States
The most common reconstruction strategy for non-Cartesian data is to interpolate the data onto a Cartesian k-space grid, followed by a Fast Fourier Transform to the image domain. However, interpolation has important consequences for the final image, so it must be properly chosen and compensated, though several packages yield good results using standard parameters. In addition, iterative techniques, both with and without regridding, can be used to incorporate an enormous range of imaging strategies.

Parallel Imaging & Multi-Coil Image Reconstruction
Vikas Gulani1
1Radiology, Case Western Reserve University, Cleveland, OH, United States
Parallel imaging reconstructions using multiple receiver coil data will be discussed, with a focus on Cartesian parallel imaging methods. SENSE and GRAPPA will be used as the representative techniques that are both widely used, and help understand a variety of other technologies.

Compressed Sensing Reconstruction
Manojkumar Saranathan1
1Dept. of Medical Imaging, University of Arizona, Tucson, AZ, United States
Many methods have been proposed to address the spatio-temporal resolution tradeoff in MRI. Compressed sensing (CS) is the latest among these and holds great promise. This talk covers the basics of compressed sensing reconstruction and also touches on more advanced CS methods that incorporate parallel imaging and redundant coil information.

Break & Meet the Teachers
      Doing More with Less  
MR Fingerprinting
Mark Griswold1
1CWRU, Cleveland, OH, United States
Synthetic MRI
Marcel Warntjes1
1Center for Medical Imaging Science and Visualization (CMIV), Linköping, Sweden
Synthetic MRI has been a long-standing dream in MRI, which recently gained more attention. Quantification techniques improve and access to clinical application becomes more facilitated. This lecture will explain the technique of synthetic MRI, its limitations and clinical impact.

Using MR Phase: Temperature Mapping & Phase-Sensitive Reconstruction
Nathan McDannold
Adjournment & Meet the Teachers

The International Society for Magnetic Resonance in Medicine is accredited by the Accreditation Council for
Continuing Medical Education to provide continuing medical education for physicians.