[MINC-users] Power spectrum across time

Andrew Janke a.janke at gmail.com
Sun Nov 29 21:15:10 EST 2015


For others with an interest in this, I have included most of the
changes needed to do this into mincfft:

   https://github.com/BIC-MNI/mincfft

Note that to perform my original request you will have to chop the
volume up into spatial + time slices as mincfft will only handle input
volumes with 3 dimensions + the vector real/imaginary dimensions if
appropriate.

Turns out I had to do this anyhow as the files were simply too large
to work on as a whole for most tools I could get my hands on. More
details in the README here:

   https://github.com/BIC-MNI/mincfft/blob/master/README

Still have to integrate Vladimirs MINC2 version into it yet...


a


On 30 November 2015 at 10:33, Andrew Janke <a.janke at gmail.com> wrote:
> Thanks Vladimir,
>
> This looks pretty much akin to what I have been doing with mincfft but
> more straight forward. I should integrate the MINC2 bits from this
> into mincfft, I'm curious to know if it will be faster. I know that
> most of the execution time of mincfft is taken up by copying arrays to
> and forth due to how I was using volume_io.
>
> a
>
>
> On 30 November 2015 at 07:51,  <vladimir.fonov at gmail.com> wrote:
>> Hello,
>>
>> I extracted the tool from my internal repository, see https://github.com/vfonov/simple_fmri_tools
>> the tool that you need is called time_fft  ( https://github.com/vfonov/simple_fmri_tools/blob/master/src/time_fft.cpp ) - it will only work if the time_dimension is fastest varying.
>>
>>
>>
>>> On Nov 26, 2015, at 18:29, Andrew Janke <a.janke at gmail.com> wrote:
>>>
>>> Has anyone got anything for power spectrum of the time domain for
>>> large MINC volumes? I don't think pyminc or octave are going to cut
>>> it. The files in question are ~20GB+ in size. I could chop them up but
>>> this seems counter-productive.
>>>
>>> I'm thinking mincfft + a bit of minccalc but if someone has something
>>> specific that uses voxel_loop or something similar I'm all ears.
>>>
>>> Thanks
>>>
>>>
>>> a
>>> _______________________________________________
>>> MINC-users at bic.mni.mcgill.ca
>>> http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users
>>
>> Best regards,
>> Vladimir S. FONOV ~ vladimir.fonov <at> gmail.com
>>
>>
>>
>>
>> _______________________________________________
>> MINC-users at bic.mni.mcgill.ca
>> http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users


More information about the MINC-users mailing list