Anisotropic Field of Views (FOVs) for Radial Imaging

2D PR Point Spread Functions

We have developed new methods for design of acquisition schemes in radial imaging, used in both Computed Tomography (CT) and Magnetic Resonance Imaging (MRI), that will support a desired imaging shape, or field-of-view (FOV). The projection spacing is varied so any desired, convex FOV shape can be supported. This allows for FOVs that are tailored to non-circular objects or regions-of-interest in 2D and 3D imaging. Tailoring the FOV allows for scan time reductions without introducing aliasing artifacts and/or reduction of these artifacts.

On this page is MATLAB code for designing 2D and 3D imaging trajectories, including the PROPELLER trajectory. Also shown are movies demonstrating how the algorithms work. These movies illustrate the evolution of the point spread function (PSF), which defines the FOV and resolution for a given trajectory.


Larson PEZ, Gurney PT, Nishimura DG. "Anisotropic Field-of-Views in Radial Imaging." IEEE Transactions on Medical Imaging2008; 27(1): 47-57. PDF

Larson PE, Lustig MS, Nishimura DG. Anisotropic field-of-view shapes for improved PROPELLER imaging. Magn Reson Imaging. 2009 May; 27(4):470-9. PDF


GitHub (You can also become a contributor to the project and add any code of your own or improve/fix the current software.)
(Alternative download:

This MATLAB package provides design functions as well as some simple FOV shapes. Use 'help radial_fovs' in MATLAB, and view the README file for more information.


The following movies show how the PSF, which is proportional to the FOV, evolves as the anisotropic FOV algorithm progresses.


The left side in the movies show the gridded projections, while the right side shows the computed PSF for that set of projections. The movies show how the variable angular spacing defines the perpendicular FOV, and how varying this spacing varies the FOV.


In these movies, the bottom-left corner shows a top-down view of the sampling pattern, while the other three images are views of the PSF. They show planes at x=0, y=0, and z=0, illustrating how the PSF evolves in the different sampling directions as the projections are designed.

3D projection sample locations on the surface of a sphere
Movie of evolving projection angles

PROPELLER sampling patterns and point spread functions