Hi,

I have a set of ROIs from SPM in MNI space that I want to load into FSL.  When I take the standard FSL brain MNI152_T1_1mm_brain and load it in MRIcro and overlay the ROI, it looks fine, like the first screenshot: http://www.ma-tea.com/~Matt/MRIcroOnMNI.png  However, if I resample the 4X4X4mm ROI so it can be overlayed on this brain in FSLView, using:

flirt –in 4X4X4mmROI –ref MNI152_T1_1mm_brain –applyxfm –init eye.mat –out 1X1X1mmROI where eye.mat is the identity matrix:

1 0 0 0

0 1 0 0

0 0 1 0

0 0 0 1

The result is incorrect, shown by the second screenshot in FSLView: http://www.ma-tea.com/~Matt/FSLOnMNI.png

The ROI file header is the following:

<nifti_image

  nifti_type = 'ANALYZE-7.5'

  image_offset = '0'

  ndim = '4'

  nx = '41'

  ny = '48'

  nz = '35'

  nt = '1'

  dx = '4'

  dy = '4'

  dz = '4'

  dt = '1'

  datatype = '16'

  nbyper = '4'

  byteorder = 'LSB_FIRST'

  cal_min = '0'

  cal_max = '1'

  descrip = 'FSL4.0'

  sform_code = '2'

  sto_xyz_matrix = '-4 0 0 80 0 4 0 -112 0 0 4 -52 0 0 0 1'

  num_ext = '0'

/>

I would appreciate any assistance with this as I am unfamiliar with this type of issue if it has to do with the analyze file format or header.

Thanks,

Matt.