Basic Atlas plotting

Plot the regions of reference atlases.

Check the list of atlases to know which ones are shipped with Nilearn.

Retrieving the atlas data

from nilearn import datasets

dataset_ho = datasets.fetch_atlas_harvard_oxford("cort-maxprob-thr25-2mm")
atlas_ho_filename = dataset_ho.filename
print(f"Atlas ROIs are located at: {atlas_ho_filename}")

dataset_ju = datasets.fetch_atlas_juelich("maxprob-thr0-1mm")
atlas_ju_filename = dataset_ju.filename
print(f"Atlas ROIs are located at: {atlas_ju_filename}")

[fetch_atlas_harvard_oxford] Dataset created in /home/runner/nilearn_data/fsl
[fetch_atlas_harvard_oxford] Downloading data from
https://www.nitrc.org/frs/download.php/9902/HarvardOxford.tgz ...
[fetch_atlas_harvard_oxford] Downloaded 23560192 of 25716861 bytes (91.6%%,
0.1s remaining)
[fetch_atlas_harvard_oxford]  ...done. (1 seconds, 0 min)

[fetch_atlas_harvard_oxford] Extracting data from
/home/runner/nilearn_data/fsl/5c734f16e50cc772ef593cab9bb3137b/HarvardOxford.tgz
...
[fetch_atlas_harvard_oxford] .. done.

Atlas ROIs are located at: /home/runner/nilearn_data/fsl/data/atlases/HarvardOxford/HarvardOxford-cort-maxprob-thr25-2mm.nii.gz
[fetch_atlas_juelich] Dataset found in /home/runner/nilearn_data/fsl
[fetch_atlas_juelich] Downloading data from
https://www.nitrc.org/frs/download.php/12096/Juelich.tgz ...
[fetch_atlas_juelich]  ...done. (1 seconds, 0 min)

[fetch_atlas_juelich] Extracting data from
/home/runner/nilearn_data/fsl/7e62e7e7fcc4d6e1428b6b2cb48f7a7c/Juelich.tgz...
[fetch_atlas_juelich] .. done.

Atlas ROIs are located at: /home/runner/nilearn_data/fsl/data/atlases/Juelich/Juelich-maxprob-thr0-1mm.nii.gz

Visualizing the Harvard-Oxford atlas

from nilearn.plotting import plot_roi, show

plot_roi(atlas_ho_filename, title="Harvard Oxford atlas")

<nilearn.plotting.displays._slicers.OrthoSlicer object at 0x7f0f57f54df0>

Visualizing the Juelich atlas

plot_roi(atlas_ju_filename, title="Juelich atlas")

<nilearn.plotting.displays._slicers.OrthoSlicer object at 0x7f0f338a2f20>

Visualizing the Harvard-Oxford atlas with contours

plot_roi(
    atlas_ho_filename,
    view_type="contours",
    title="Harvard Oxford atlas in contours",
)
show()

Visualizing the Juelich atlas with contours

plot_roi(
    atlas_ju_filename, view_type="contours", title="Juelich atlas in contours"
)
show()

Visualizing an atlas with its own colormap

Some atlases come with a look-up table that determines the color to use to represent each of its regions.

You can pass this look-up table as a pandas dataframe to the cmap argument to use its colormap.

Look-up table format

The look-up table must be formatted according to the BIDS standard. and that the colors must be in color column using hexadecimal values.

If an invalid look-up table is passed, a warning will be thrown and the plot_roi function will fall back to using its default colormap.

Here we are using the Yeo atlas that comes with a predefined colormap.

dataset_yeo = datasets.fetch_atlas_yeo_2011(n_networks=17)

print(dataset_yeo.lut)

[fetch_atlas_yeo_2011] Dataset created in /home/runner/nilearn_data/yeo_2011
[fetch_atlas_yeo_2011] Downloading data from
ftp://surfer.nmr.mgh.harvard.edu/pub/data/Yeo_JNeurophysiol11_MNI152.zip ...
[fetch_atlas_yeo_2011] Downloaded 745472 of ? bytes.
[fetch_atlas_yeo_2011] Downloaded 1589248 of ? bytes.
[fetch_atlas_yeo_2011] Downloaded 2433024 of ? bytes.
[fetch_atlas_yeo_2011]  ...done. (4 seconds, 0 min)

[fetch_atlas_yeo_2011] Extracting data from
/home/runner/nilearn_data/yeo_2011/622858b56913b19ae0865d9fa8ad47cc/Yeo_JNeuroph
ysiol11_MNI152.zip...
[fetch_atlas_yeo_2011] .. done.

    index           name    color
0       0     Background  #000000
1       1   17Networks_1  #781286
2       2   17Networks_2  #ff0000
3       3   17Networks_3  #4682b4
4       4   17Networks_4  #2acca4
5       5   17Networks_5  #4a9b3c
6       6   17Networks_6  #00760e
7       7   17Networks_7  #c43afa
8       8   17Networks_8  #ff98d5
9       9   17Networks_9  #dcf8a4
10     10  17Networks_10  #7a8732
11     11  17Networks_11  #778cb0
12     12  17Networks_12  #e69422
13     13  17Networks_13  #87324a
14     14  17Networks_14  #0c30ff
15     15  17Networks_15  #000082
16     16  17Networks_16  #ffff00
17     17  17Networks_17  #cd3e4e

Let’s compare the atlas with the default colormap and its own colormap.

plot_roi(
    dataset_yeo.maps,
    title="Yeo atlas",
    colorbar=True,
)

plot_roi(
    dataset_yeo.maps,
    title="Yeo atlas with its own colors",
    cmap=dataset_yeo.lut,
    colorbar=True,
)

show()

• 
• 

Estimated memory usage: 656 MB