Russ-
Thanks very much for your scripts. They are really useful. Leonardo Bonilha
and I have updated the scripts to work on Windows as well as Linux and
written a tutorial that includes a sample dataset:
http://www.psychology.nottingham.ac.uk/staff/cr1/fsl/
I have tried to mark all the major changes in the scripts with '%CRorden'. A
few of the changes include:
1.) Includes a parameter that allows the user to point to a brain image in
stereotaxic space. Your original code required modifying FSL so that a
reg_standard brain would not be deleted.
2.) Changed '/' symbols to 'filesep', so paths work in Windows and Unix.
3.) Removed transpose in selective_avg. Not sure why your original code
transposed the data, as both Windows and Linux would not run with the
transposition:
b_est=pinv(SCM)*roidata';
to read
b_est=pinv(SCM)*roidata;
4.) Selective_avg attempts to find the directory of applyxfm4D (UNIX
default: /usr/local/fsl/bin/; Windows default: C:\cygwin\usr\local\fsl\bin\
and warns the user that the file must be in the user path if not explicitly
found.
5.) Modified save_avw_hdr to find the directory of avwcreatehd (UNIX
default: /usr/local/fsl/bin/; Windows default: C:\cygwin\usr\local\fsl\bin\
and warns the user that the file must be in the user path if not explicitly
found.
6.) Modified save_avw_hdr to store image origin information
7.) Modified read_avw_hdr to read image origin information
8.) The designconvert.sed file is very sensitive to line-ending symbols:
UNIX and Windows require different versions of this file.
9.) Created batch file to set parameters and run scripts.
-chris
|