[MINC-users] functional connectivity problem with alternating slice pattern (Keith's way)

[Andrew Janke]

Hi Chris,

On Tue, Jul 28, 2009 at 03:47, Penhune Lab<penhunelab at gmail.com> wrote:
> problem: the mag_t.mnc files resulting from the analysis appear to have an
> alternating slice pattern.
>
> I have already conducted simple model-based fmrilm analyses on this data set
> and do not see this same alternating slice pattern in them. I acquired at a
> resolution of 4mm^3 (interleaved ascending*) *on the BIC 3T. From what I can
> tell, the issue arises because of the interleaved ascending slicetimes. If I
> change the slicetimes from interleaved ascending to ascending the
> alternating pattern disappears.

Sounds to me like you are being bitten by slice scaling issues. MINC
will scale each slice separately (and maintain an internal list of
scaling values) to ensure the best range is stored for each slice as
it is converted/processed. You seem to be using fmristat? which means
you are using EMMA and I not sure as to how well EMMA handled slice
scaling. My first approach to this problem would be to convert your
input volume to float. When volumes are in float/double format slice
scaling is not used.

   mincreshape -float input.mnc input_float.mnc

And then re-run the analysis.


--
Andrew Janke
(a.janke at gmail.com || http://a.janke.googlepages.com/)
Canberra->Australia    +61 (402) 700 883