High Field Susceptibility-Weighted Imaging

Julien Milles
Martijn Steenwijk


This project started officially in November 2008. It stems from the common interest from both the high-field MR (Dr. Ir. M.J.P. van Osch) and small animal imaging (Dr. L. van der Weerd) sections in the Neuro-Radiology department.

Several recent studies have shown that the use of MRI signal phase can improve contrast in specific human brain structures, including veins and iron-rich regions. Preliminary studies have also shown a substantial contrast between WM and cortical GM. This contrast originates, at least in part, from the magnetic susceptibility differences between tissues. Phase imaging allows increased dynamic range in detection of these susceptibility effects by directly measuring the change in frequency offset. Using high field MRI scanners for phase imaging presents some clear advantages over more commonly used low fields, such as the requirement for very homogeneous transmit fields that are hard to obtain at high fields or high power deposition that can occur. In addition, phase contrast is predicted to increase at higher magnetic fields.

Involvement of the Neuro-Image Processing section in that project is in the development of post-processing methods to extract relevant phase information from MRI raw data and to compute the SWI maps.


We have developed a number of applications that, combined in a pipeline, allow for computation and analysis of SWI data. Those applications are:

  1. Phase unwrapping algorithm. We have implemented and tested several different methods aimed at unwrapping the phase information. Those include unwrapping methods based on spatial information (e.g. region-growing) to Fourier-based unwrapping. Since ground truth is relatively unknown, evaluation has to be performed in a comparative manner.
  2. Phase deramping algorithm. Some phase unwrapping methods need to be used in combination with a deramping step that aims at removing the slow spatial variations created by the unwrapping. Similarly to phase unwrapping methods, several choices are possible, from polynomial fitting to spectral deramping.
  3. SWI mask computation algorithm. Once the relevant phase information is extracted, a SWI phase mask needs to be computed in order to enhance, or reduce, certain regions in the data. A simple way of understanding that step is to think of phase as a weighting mask applied to the magnitude data.




Figure 1.Example of phase unwrapping. From left to right and top to bottom: Magnitude image, corresponding wrapped phase image, unwrapped phase image and unwrapped phase image with enhanced contrast.


For further information, please contact:
Dr. JR Milles, PhD
Division of Image Processing
Department of Radiology, 1-C2S
Leiden University Medical Center
P.O. Box 9600
2300 RC Leiden
The Netherlands
Tel. +31 (0)71 526 5342
Fax. +31 (0)71 526 6801
mailto: J.R.Milles@lumc.nl


Conference abstracts

  • Milles J, van Osch MJ, van der Weerd L, Nabuurs RJ, Teeuwisse WM, van der Grond J, van Buchem MA, Reiber JH. “Local feature-preserving selection of kernel size for unwrapping of high-resolution phase images”. In: Proceedings ISMRM 2008.
  • Nabuurs RJ, Milles JR, van Rooden S, van der Grond J, Reiber JH, van Buchem MA, van der Weerd L. “High resolution T2* and phase contrast of human AD brain tissue at 9.4T: a structural comparison”. In: Proceedings ISMRM 2008.