Not being Roland and al...

On 23. Mar 2007, at 18:49 Uhr, freesurfer-request@nmr.mgh.harvard.edu wrote:

Message: 2 Date: Thu, 22 Mar 2007 15:18:05 -0400 (EDT) From: Bruce Fischl fischl@nmr.mgh.harvard.edu Subject: Re: [Freesurfer] Re: Freesurfer Digest, Vol 37, Issue 16 To: Paul Aparicio aparicio@mit.edu Cc: freesurfer@nmr.mgh.harvard.edu Message-ID: Pine.LNX.4.62.0703221517330.3420@gate.nmr.mgh.harvard.edu Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed

no, definitely don't use COR. I think you can use mri_convert -i[ijk]s. Rudolph: can you elaborate on how to do this?

here is what I use to accomplish this (in a csh script): set targ_dir = /path/to/your/input_volumes set targ_file_name = example.mgz set conformed_vol = the_name_of_the_volume_with_faked_resolution.mgz mri_convert -i ${targ_dir}/${targ_file_name} -o ${targ_dir}/${conformed_vol} -iis 1 -ijs 1 -iks 1

CAVEAT this expects isometric input scans, otherwise it gets a bit more elaborate...: this is what I used in the days of non-isometric scanning. It will express all resolutions as multiples of the minmum so that resampling later (by running mri_convert -cm ...) will do something reasonable. You still have to make sure, that you have <= 256 voxels per dimension...

ahoi & hope it helps Sebastian

##### silly script to follow...###### #! /bin/csh! # run number from the dicom file zero padded to three digits set run = 008 # number of the first dicom file to contain data for this run (actually any dicom file for a run should do'?) set first_dicom = 34840259

# the session to work on set session_dir = 050610Lupo set dicom_dir = 06131042 set user_id = moeller # where does the dicom data live set src_dir = /space/data/raw_data/20${session_dir}/ set src_file = $src_dir/$first_dicom # where to put the cor files set targ_base_dir = /space/data/${user_id}/cooked/anatomicals/$session_dir/mri mkdir $targ_base_dir # the next one has to match the run number from the dicom information... set targ_dir = $targ_base_dir/orig

# get the dimensions (read from unpack.log) # let's do it like freesurfer does (no error checkin though) set first_two_dims = `mri_probedicom --i $src_file --t 28 30` # tag 28 30 is of form 0.453125\0.453125, so split the critter at the back slash #echo $first_two_dims set first_dim = `echo "$first_two_dims" | awk '{split($0, a, "\"); print a[1]}'` #echo $first_dim set second_dim = `echo "$first_two_dims" | awk '{split($0, a, "\"); print a[2]}'` #echo $second_dim # the slice resolution either in tag 18 50 or 18 88 set third_dim = `mri_probedicom --i $src_file --t 18 50` #echo $third_dim

# get the shortest dimension, all this requires gnu bc to be installed set min_dim = `echo "min = $first_dim; if (min > $second_dim) { min = $second_dim }; if (min > $third_dim) { min = $third_dim }; min" | bc -l` set iis = `echo "scale=6; $first_dim / $min_dim" | bc -l` set ijs = `echo "scale=6; $second_dim / $min_dim" | bc -l` set iks = `echo "scale=6; $third_dim / $min_dim" | bc -l`

#for unpacking at raw resolution mri_convert -i $src_file -o $targ_dir -it siemens_dicom -ot mgh #-iis $iis -ijs $ijs -iks $iks #end of script... ###################################################