1) Differentiate between CT and MR data types, errors, and typical image artifacts. 2) Detect effect of k-space data re-gridding in final image. 3) Describe propeller data corrections and identify benefits upon image quality.
We begin by exploring the differences between MR & CT data – and the differences in final reconstructed images. Next, we examine artifacts suffered by all MR scans, such as Gibbs ringing which results from sampling over only a finite region of k-space. k-space apodization reduces ringing artifacts.
Next we consider spiral scanning, one of several non-Cartesian data acquisition schemes, and examine the process by which k-space data sampled on a non-Cartesian set of points are resampled onto a Cartesian lattice suitable for inversion via Fas Fourier Transform. Why this process works especially well for MRI data and how it can fail are discussed.
Finally, we consider Propeller, a relatively new hybrid technique. Propeller fills Fourier space by sampling multiple rotated Cartesian data sets using fast spin echo (FSE), a standard - and slow - acquisition scheme. Acquisition is further slowed because by resampling the center of k-space many times over. Propeller’s effective temporal resolution is improved by exploiting relationships amongst the redundant measurements to minimize motion artifacts.
Patch, S,
k-Space Data Preprocessing for Artifact Reduction in MR Imaging. Radiological Society of North America 2007 Scientific Assembly and Annual Meeting, November 25 - November 30, 2007 ,Chicago IL.
http://archive.rsna.org/2007/4405284.html
