[Neuroimaging] Use of atlas to compute Z-Scores

[Gabriel Reynés]

Dear all,

thank you for all responses. My main problem was that the
function masker.fit_transform() needs a 4D image to work. I was trying to
perform the computation first with the single patient and this error was
raised:

"DimensionError: Input data has incompatible dimensionality: Expected
dimension is 4D and you provided a 3D image. See http://nilearn.github.io/
manipulating_images/input_output.html.
"

To solve that, based in your example, I saw that I need to perform the
fit_transform over the single subject and the normal database at the same
time, this is more cpu-consuming and prone to errors, but know seems to
work just fine.

Thanks!


Best regards,


Gabriel

On Wed, Feb 14, 2018 at 11:25 AM, Gabriel Reynés <greynell at gmail.com> wrote:

> Dear Christophe  ,
>
>
> Thanks for your answer! I spent many hours triyng to figure the behaviour
> of NiftiLabelsMasker
> I do not understand how to iterate over each mask ROI.
>
> How one can iterate over each label of a masker?
>
> `# Obtain AAL Atlas
> aal = datasets.fetch_atlas_aal('SPM12')
> aal_labels = aal.labels
>
> # Obtain mask
> masker = input_data.NiftiLabelsMasker(aal.maps)
>
> # Input image to compute the mean value for each
> patient_img = image.load_img(path_nii)
> patient_img  = patient_img.get_data()
>
> # Fit the mask to the image
> masker_fit(patient_img)
> mask_img = masker_fit.mask_img_
>
> Thanks in advance,
>
>
> Gabriel
>
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20180215/8f2fb9ed/attachment-0001.html>