Marialena,
I agree that it is complicated! In principle QUASIL can cope with data coming in in a number of different arrangements (it does a certain amount of internal re-arrangement to do the analysis). However, by default it assumes that you have all the pairs of measurements from a single repetition of one phase at a single TI together. Thus you would have a block of 6 repeats the phases at a single TI, and have 13 of these blocks in your file in order of increasing TI value.
Michael
> On 15 Nov 2017, at 13:14, Marialena Dounavi <[log in to unmask]> wrote:
>
> Dear FSL users,
> I have some QUASAR ASL data and I am trying to process them with quasil.
> This thread was particularly helpful but confusing as well in terms of the terminology (phases, volumes etc).
>
> The data I am processing have been acquired with the original QUASAR protocol (crushed, non crushed, low flip angle).
>
> I have tried several permutations of the nifti file but it seems that my data are not organised in a format appropriate for quasil processing.
>
> I created a 4D nifti file with dimensions 64*64*7*1092
> Dcm2niix seems to organise these 1092 as 13 timepoints * 14 (7 phases *2 for control-label) *6 repetitions of the scheme.
> However I am getting with this rather homogenous maps with no areas of activation.
> Is this how the data need to be organised? A 4D matrix that actually is structured as follows: 64,64,7,13,14,6?
>
> Thank you in advance for your help,
> Kind Regards,
> Marialena
|