Thresholded Surface Plot with nilearn

Susanne_Eisenhauer · December 7, 2018, 3:04pm

Hello,

I am trying to plot functional data on a surface with nilearn like in this example: https://nilearn.github.io/auto_examples/01_plotting/plot_3d_map_to_surface_projection.html#sphx-glr-auto-examples-01-plotting-plot-3d-map-to-surface-projection-py

This part works fine (although I am wondering about the uncolored part of the mesh):

from nilearn import image
from nilearn import surface
from nilearn import plotting
%matplotlib inline

data_img = image.smooth_img(’/data/hippocampus/PWlearn_MEG/SourceCustomCode/Volwrite_Familiarity1_PWL-W.nii’, fwhm=0)

texture = surface.vol_to_surf(data_img,
‘/data/common/SPM8current/spm8/canonical/cortex_5124.surf.gii’)

fig = plotting.plot_surf_stat_map(stat_map = texture, surf_mesh = ‘/data/common/SPM8current/spm8/canonical/cortex_5124.surf.gii’, hemi=‘left’,
title=‘Surface left hemisphere’, colorbar=True)

However, when I use a threshold only the functional data is plotted:

fig = plotting.plot_surf_stat_map(stat_map = texture, surf_mesh = ‘/data/common/SPM8current/spm8/canonical/cortex_5124.surf.gii’, hemi=‘left’,
colorbar=True, threshold = 0.6*max(texture))

grafik

I also tried to provide the surface image as bg_map but this produces an error:

ValueError Traceback (most recent call last)
in ()
1 fig = plotting.plot_surf_stat_map(stat_map = texture, surf_mesh = ‘/data/common/SPM8current/spm8/canonical/cortex_5124.surf.gii’, hemi=‘left’,
----> 2 colorbar=True, threshold = 0.6*max(texture), bg_map = ‘/data/common/SPM8current/spm8/canonical/cortex_5124.surf.gii’)

/data/home1/seisenha/.local/lib/python3.5/site-packages/nilearn/plotting/surf_plotting.py in plot_surf_stat_map(surf_mesh, stat_map, bg_map, hemi, view, threshold, alpha, vmax, cmap, colorbar, symmetric_cbar, bg_on_data, darkness, title, output_file, axes, figure, **kwargs)
417 alpha=alpha, bg_on_data=bg_on_data, darkness=1, vmax=vmax, vmin=vmin,
418 title=title, output_file=output_file, axes=axes, figure=figure,
–> 419 cbar_vmin=cbar_vmin, cbar_vmax=cbar_vmax, **kwargs)
420
421 return display

/data/home1/seisenha/.local/lib/python3.5/site-packages/nilearn/plotting/surf_plotting.py in plot_surf(surf_mesh, surf_map, bg_map, hemi, view, cmap, colorbar, avg_method, threshold, alpha, bg_on_data, darkness, vmin, vmax, cbar_vmin, cbar_vmax, title, output_file, axes, figure, **kwargs)
212 bg_data = load_surf_data(bg_map)
213 if bg_data.shape[0] != coords.shape[0]:
–> 214 raise ValueError('The bg_map does not have the same number ’
215 ‘of vertices as the mesh.’)
216 bg_faces = np.mean(bg_data[faces], axis=1)

ValueError: The bg_map does not have the same number of vertices as the mesh.

Also, view_surf is not working when I add the surf_map variable:

TypeError Traceback (most recent call last)
/data/home1/seisenha/Tools/anaconda3/lib/python3.5/site-packages/matplotlib/colors.py in to_rgba(c, alpha)
165 try:
–> 166 rgba = _colors_full_map.cache[c, alpha]
167 except (KeyError, TypeError): # Not in cache, or unhashable.

TypeError: unhashable type: ‘numpy.ndarray’

During handling of the above exception, another exception occurred:

ValueError Traceback (most recent call last)
in ()
----> 1 plotting.view_surf(surf_map = texture, surf_mesh=’/data/common/SPM8current/spm8/canonical/cortex_5124.surf.gii’)

/data/home1/seisenha/.local/lib/python3.5/site-packages/nilearn/plotting/html_surface.py in view_surf(surf_mesh, surf_map, bg_map, threshold, cmap, black_bg, vmax, symmetric_cmap)
261 surf_map=surf_map, surf_mesh=surf_mesh, threshold=threshold,
262 cmap=cmap, black_bg=black_bg, bg_map=bg_map,
–> 263 symmetric_cmap=symmetric_cmap, vmax=vmax)
264 return _fill_html_template(info, embed_js=True)

/data/home1/seisenha/.local/lib/python3.5/site-packages/nilearn/plotting/html_surface.py in one_mesh_info(surf_map, surf_mesh, threshold, cmap, black_bg, bg_map, symmetric_cmap, vmax)
54 info[‘vertexcolor_left’] = _get_vertexcolor(
55 surf_map, colors[‘cmap’], colors[‘norm’],
—> 56 colors[‘abs_threshold’], bg_map)
57 info[“cmin”], info[“cmax”] = float(colors[‘vmin’]), float(colors[‘vmax’])
58 info[‘black_bg’] = black_bg

/data/home1/seisenha/.local/lib/python3.5/site-packages/nilearn/plotting/html_surface.py in _get_vertexcolor(surf_map, cmap, norm, absolute_threshold, bg_map)
22 vertexcolor = cmap(norm(surf_map).data)
23 if absolute_threshold is None:
—> 24 return to_color_strings(vertexcolor)
25 if bg_map is None:
26 bg_map = np.ones(len(surf_map)) * .5

/data/home1/seisenha/.local/lib/python3.5/site-packages/nilearn/plotting/js_plotting_utils.py in to_color_strings(colors)
270 def to_color_strings(colors):
271 cmap = mpl.colors.ListedColormap(colors)
–> 272 colors = cmap(np.arange(cmap.N))[:, :3]
273 colors = np.asarray(colors * 255, dtype=‘uint8’)
274 colors = [’#{:02x}{:02x}{:02x}’.format(*row) for row in colors]

/data/home1/seisenha/Tools/anaconda3/lib/python3.5/site-packages/matplotlib/colors.py in call(self, X, alpha, bytes)
480 # See class docstring for arg/kwarg documentation.
481 if not self._isinit:
–> 482 self._init()
483 mask_bad = None
484 if not cbook.iterable(X):

/data/home1/seisenha/Tools/anaconda3/lib/python3.5/site-packages/matplotlib/colors.py in _init(self)
831
832 def _init(self):
–> 833 rgba = colorConverter.to_rgba_array(self.colors)
834 self._lut = np.zeros((self.N + 3, 4), float)
835 self._lut[:-3] = rgba

/data/home1/seisenha/Tools/anaconda3/lib/python3.5/site-packages/matplotlib/colors.py in to_rgba_array(arg, alpha)
351 then an empty array will be returned. Same for an empty list.
352 “”"
–> 353 return to_rgba_array(arg, alpha)
354
355

/data/home1/seisenha/Tools/anaconda3/lib/python3.5/site-packages/matplotlib/colors.py in to_rgba_array(c, alpha)
265 result = np.empty((len©, 4), float)
266 for i, cc in enumerate©:
–> 267 result[i] = to_rgba(cc, alpha)
268 return result
269

/data/home1/seisenha/Tools/anaconda3/lib/python3.5/site-packages/matplotlib/colors.py in to_rgba(c, alpha)
166 rgba = _colors_full_map.cache[c, alpha]
167 except (KeyError, TypeError): # Not in cache, or unhashable.
–> 168 rgba = _to_rgba_no_colorcycle(c, alpha)
169 try:
170 _colors_full_map.cache[c, alpha] = rgba

/data/home1/seisenha/Tools/anaconda3/lib/python3.5/site-packages/matplotlib/colors.py in _to_rgba_no_colorcycle(c, alpha)
217 # float)andnp.array(…).astype(float)` all convert “0.5” to 0.5.
218 # Test dimensionality to reject single floats.
–> 219 raise ValueError(“Invalid RGBA argument: {!r}”.format(orig_c))
220 # Return a tuple to prevent the cached value from being modified.
221 c = tuple(c.astype(float))

ValueError: Invalid RGBA argument: array([[0.05189484, 0. , 0. , 1. ]])

Thank you very much!
Susanne

jeromedockes · January 18, 2019, 7:15pm

Hello,

“(although I am wondering about the uncolored part of the mesh)”:

I’m guessing the map appears to be thresholded wherever it is 0, regardless of
wether you provided a threshold or not. This is an issue with this plotting
function that should be fixed in a future version of nilearn:

“However, when I use a threshold only the functional data is plotted”:

Again, this is an issue with this plotting function. If you provide no background,
the mesh appears transparent where the image is thresholded:
https://github.com/nilearn/nilearn/issues/1738. should be fixed in a future
version of nilearn

“I also tried to provide the surface image as bg_map but this produces an error”:

from the traceback you provided the error is:

ValueError: The bg_map does not have the same number of vertices as the mesh.

have you checked that the background map is the right one for the mesh you are using?
can you provide the shape of the mesh and of the background map?

could you provide the shape of surf_mesh, surf_map and bg_map?

thanks!

Susanne_Eisenhauer · January 26, 2019, 4:41pm

Hello,
thanks for the reply!

I provide the same input as surf_mesh and bg_map:
fig = plotting.plot_surf_stat_map(stat_map = texture, surf_mesh = ‘/data/common/SPM8current/spm8/canonical/cortex_5124.surf.gii’, hemi=‘left’,
colorbar=True, threshold = 0.6*max(texture), bg_map = ‘/data/common/SPM8current/spm8/canonical/cortex_5124.surf.gii’)

When I load them first, using surface.load_surf_mesh and surface.load_surf_data, both have the shape (2,).
surf_mesh[0].shape = (5124, 3)
surf_mesh[1].shape = (10240, 3)
bg_map[0].shape = (10240, 3)
bg_map[1].shape = (5124, 3)
stat_map.shape = (5124, 1)

Best, Susanne

jeromedockes · January 28, 2019, 9:00am

bg_map sould not be a mesh (i.e. coordinates), but a value for each node of the mesh,
which gets translated into a color for nodes where the stat map is thresholded.
typically, it can be the sulcal depth. so the bg_map should have the same shape
as the stat map, i.e. (number of nodes, )

best, Jérôme

Sebastian · February 25, 2020, 4:46pm

Hi all,

I am trying to plot a tstat map that shows activation in a ROI and is zero everywhere else.

plotting.view_img_on_surf(edge1_bin, threshold=‘0%’, surf_mesh=‘fsaverage’)

Is there a way to plot it without the surface being black?

Thanks for your help,

Sebastian

jeromedockes · March 18, 2020, 1:50pm

can you try threshold=1e-6? you may also want to change the colormap and use colorbar=False

Sebastian · January 14, 2021, 12:00pm

Thanks, that worked.