MR sequence information

  • T1-weighted (T1w) images: They are commonly used for segmentation of the brain. They are usually acquired at a rather low readout bandwidth to maximize SNR. However, using a low readout bandwidth comes with the disadvantage that the positions of the (fatty) spongy bone and subcutaneous fat will be displaced in the MR images due to the chemical shift artifact. This can result in the spongy bone touching brain gray matter (GM), rendering an accurate segmentation of the GM pial surface and the boundary between cerebrospinal fluid (CSF) and skull difficult. For this reason, we recommend to acquire the T1w images WITH A FAT SUPPRESSION METHOD such as selective water excitation.
  • T2-weighted (T2w) images: As both CSF and compact bone are dark in T1w images (also spongy bone will be dark when fat suppression is used), an accurate reconstruction of the skull is difficult when only a T1w image is available. For this reason, we recommend to additionally acquire a T2w image WITHOUT FAT SUPPRESSION. T2w images are usually acquired at a high readout bandwidth, minimizing the chemical shift artifact. As they also provide very good contrast between CSF (bright) and compact bone (dark), they are a good starting point for segmenting the skull.
  • Example for a pair of T1w and T2w images of good quality:

  • Example for a problematic T1w. The spongy bone is in part very bright (indicated by the red arrows) and a dark CSF region to separate GM and skull is missing at several positions (purple arrows indicate example positions):

  • Further information on MR sequence settings can be found in (Nielsen et al, NeuroImage, 2018)
 
 
version21/mri_sequences.txt · Last modified: 2018/07/27 18:04 by axel
Recent changes RSS feed Creative Commons License Valid XHTML 1.0 Valid CSS Driven by DokuWiki
Drupal Garland Theme for Dokuwiki