Hi Mark,
I tried running fdt_rotate_bvec on a couple of machines, and it turned out the problem seems restricted to the set up:
The problem I initially encountered was on a MacBook Pro running FSL 5.0.4 under OSX 10.8.4
fdt_rotate_bvec ran fine on FSL 5.0.1 under Red Hat Linux 6.4 (I copied the script from a later version) and on FSL 5.0.4 on Ubuntu 12.04 LTS (I had to change the first line from #!/bin/sh to #!/bin/bash).
I still haven't been able to get fdt_rotate_bvec working on OSX. Do you have any suggestions for me?
Thanks,
Vincent
Op 17 aug. 2013, om 04:18 heeft Mark Jenkinson <[log in to unmask]> het volgende geschreven:
> Hi,
>
> This looks fine, so I've no idea why you are getting all zeros.
> Can you please upload the bvec and ecclog files to this site:
> https://oxfile.ox.ac.uk/oxfile/work/extBox?id=68312615463381F4C
> and we'll see if we can track down the problem at this end.
>
> All the best,
> Mark
>
>
> On 16 Aug 2013, at 12:42, Vincent Koppelmans <[log in to unmask]> wrote:
>
>> Hi Mark,
>>
>> Like I said, I have two datasets, so here are the outputs for one bvec matrix of each set:
>>
>> Data set 1)
>> file: ASCII text, with very long lines
>> wc: 3 186 3107 <name_oiginal_bvec_file>.bvec
>>
>> Data set 2)
>> file: ASCII text
>> wc: 3 48 430 <name_oiginal_bvec_file>.bvec
>>
>> Thanks for your help,
>>
>> Vincent
>>
>>
>>
>> Op 16 aug. 2013, om 04:00 heeft Mark Jenkinson <[log in to unmask]> het volgende geschreven:
>>
>>> Hi,
>>>
>>> It sounds like you are running it correctly.
>>> My only suggestion is that the bvec file might not have a standard unix format.
>>> Can you run the following:
>>> file original_bvec_file.bvec
>>> and
>>> wc original_bvec_file.bvec
>>> and let us know what the outputs are?
>>>
>>> All the best,
>>> Mark
>>>
>>>
>>>
>>>
>>> On 15 Aug 2013, at 19:55, Vincent <[log in to unmask]> wrote:
>>>
>>>> Hi FSL users,
>>>>
>>>> I have two different datasets of DWI images (different resoution, directions, NSA, etc.) that I corrected for motion and eddy current using eddy_current.
>>>>
>>>> I tried using fdt_rotate_bvecs (FSL 5.0.4) on subjects of both data sets, but the script always outputs a matrix of zeros.
>>>>
>>>> The syntax I use is:
>>>> fdt_rotate_bvec original_bvec_file.bvec output_name_new_bvec_file.txt ecc_log.ecclog
>>>>
>>>> An example of a bvec file is:
>>>> 0 -0.216002 0.444561 0.635186 -0.500549 0.22123 0.891941 0.169316 0.263966 -0.598524 0.899631 -0.985305 -0.627905 0.350119 -0.527048 -0.420617
>>>> 0 -0.898615 -0.418727 -0.770066 -0.192538 0.53117 0.079143 0.513429 -0.440254 0.00692 0.223429 -0.166405 0.62708 0.93647 0.773124 -0.804161
>>>> 0 0.381882 0.791854 -0.059474 0.844026 -0.817873 -0.445171 0.841262 -0.858195 -0.801075 0.375157 0.038505 0.460984 0.020968 -0.352844 -0.420008
>>>>
>>>> An example of the ecclog file is:
>>>> processing ecc_DWI_tmp0000
>>>>
>>>> Final result:
>>>> 1.000000 0.000000 0.000000 0.000000
>>>> 0.000000 1.000000 0.000000 0.000000
>>>> 0.000000 0.000000 1.000000 0.000000
>>>> 0.000000 0.000000 0.000000 1.000000
>>>>
>>>> processing ecc_DWI_tmp0001
>>>>
>>>> Final result:
>>>> 1.013296 0.001221 -0.000585 -1.589181
>>>> -0.000941 1.018853 -0.003931 -1.686168
>>>> 0.001744 -0.003690 1.030896 -1.585792
>>>> 0.000000 0.000000 0.000000 1.000000
>>>>
>>>> processing ecc_DWI_tmp0002
>>>>
>>>> Final result:
>>>> 1.012999 0.000073 -0.000476 -1.460119
>>>> 0.001104 1.017430 -0.010545 -1.262310
>>>> -0.000546 0.001471 1.029328 -1.436947
>>>> 0.000000 0.000000 0.000000 1.000000
>>>> (etc...)
>>>>
>>>> fdt_rotate_bvec does not output an error. It just shows me the count of all the slices/directions being processed.
>>>>
>>>> Where do I go wrong?
>>>>
>>>> Thanks, Vincent
|