def check_images(img_spec1, img_spec2, bkground_thresh=0.05):
"""Reads the two images and assers identical shape."""
img1 = read_image(img_spec1, bkground_thresh)
img2 = read_image(img_spec2, bkground_thresh)
if img1.shape != img2.shape:
raise ValueError(
'size mismatch! First image: {} Second image: {}\n'
'Two images to be compared must be of the same size in all dimensions.'
.format(img1.shape, img2.shape))
return img1, img2
def test_collage_class():
img_path = pjoin(base_dir, '3569_bl_PPMI.nii')
img = read_image(img_path, None)
scaled = scale_0to1(img)
c = Collage(num_slices=15, view_set=(0, 1), num_rows=3)
try:
c.attach(scaled)
except:
raise ValueError('Attach does not work')
try:
c.transform_and_attach(scaled, np.square)
except:
raise ValueError('transform_and_attach does not work')
try:
print(c)
except:
raise ValueError('repr implementation failed')
def test_slice_picker():
img_path = pjoin(base_dir, '3569_bl_PPMI.nii')
img = read_image(img_path, None)
sp = SlicePicker(img, num_slices=15, view_set=(0, 1))
try:
for dim, sl_num, data in sp.get_slices(extended=True):
print(dim, sl_num, data.shape)
except:
raise ValueError('get_slices() does not work')
try:
for d1, d2 in sp.get_slices_multi((img, img)):
assert np.allclose(d1, d2)
except:
raise ValueError('get_slices_multi() does not work')
try:
print(sp)
except:
raise ValueError('repr implementation failed')
def density_over(img2d, min_density=0.65):
return (np.count_nonzero(img2d.flatten()) / img2d.size) <= min_density
print('testing different sampling strategies .. ')
for sname, sampler in zip(('linear', 'percent', 'callable'),
('linear', (5, 50, 95), density_over)):
sp = SlicePicker(img, sampler=sampler)
print(sname)
print(repr(sp))
print('testing linear sampling')
for ns in np.random.randint(0, min(img.shape), 10):
sp_linear = SlicePicker(img, sampler='linear', num_slices=ns)
print(repr(sp_linear))
if 3 * ns != len(sp_linear.get_slice_indices()):
raise ValueError('error in linear sampling')
print('testing percentage sampling')
perc_list = [5, 10, 45, 60, 87]
sp_perc = SlicePicker(img, sampler=perc_list)
print(repr(sp_perc))
if 3 * len(perc_list) != len(sp_perc.get_slice_indices()):
raise ValueError('error in percentage sampling')
print('testing ability to save to gif')
import tempfile
gif_path = tempfile.NamedTemporaryFile(suffix='.gif').name
print(gif_path)
sp.save_as_gif(gif_path)
if not pexists(gif_path):
raise IOError('Saving to GIF failed')
try:
import imageio
gif = imageio.mimread(gif_path, format='gif')
except:
raise ValueError('Saved GIF file could not be read properly!')
print()
print()
img_format)))
for idx in range(5)
]
if all(scr_shot_exist):
continue
try:
t1_path = pjoin(in_dir, subject_id,
'{}_{}.nii'.format(subject_id, 't1'))
lesion_path = pjoin(in_dir, subject_id,
'{}_{}.nii'.format(subject_id, 'AutoSeg_linda'))
# img = nib.load(t1_path)
# # print("image shape", img.shape)
t1w = read_image(t1_path, None)
max_shape = max(t1w.shape)
lesion = read_image(lesion_path, None)
lesion = lesion.astype('int')
# casting to remove any negligible differences due to interpolation
# removing background - 0 stays 0
unique_labels = np.setdiff1d(np.unique(lesion.flatten()), 0)
LesionSlicePicker = SlicePicker(lesion,
view_set=[
2,
],
num_slices=5,
min_density=0.0001)
idx = 0
def aseg_on_mri(mri_spec,
aseg_spec,
alpha_mri=1.0,
alpha_seg=1.0,
num_rows=2,
num_cols=6,
rescale_method='global',
aseg_cmap='freesurfer',
sub_cortical=False,
annot=None,
padding=5,
bkground_thresh=0.05,
output_path=None,
figsize=None,
**kwargs):
"Produces a collage of various slices from different orientations in the given 3D image"
num_rows, num_cols, padding = check_params(num_rows, num_cols, padding)
mri = read_image(mri_spec, bkground_thresh=bkground_thresh)
seg = read_image(aseg_spec, bkground_thresh=0)
mri, seg = crop_to_seg_extents(mri, seg, padding)
num_slices_per_view = num_rows * num_cols
slices = pick_slices(seg, num_slices_per_view)
plt.style.use('dark_background')
num_axes = 3
if figsize is None:
figsize = [5 * num_axes * num_rows, 5 * num_cols]
fig, ax = plt.subplots(num_axes * num_rows, num_cols, figsize=figsize)
# displaying some annotation text if provided
if annot is not None:
fig.suptitle(annot, backgroundcolor='black', color='g')
display_params_mri = dict(interpolation='none',
aspect='equal',
origin='lower',
cmap='gray',
alpha=alpha_mri,
vmin=mri.min(),
vmax=mri.max())
display_params_seg = dict(interpolation='none',
aspect='equal',
origin='lower',
alpha=alpha_seg)
normalize_labels = colors.Normalize(vmin=seg.min(),
vmax=seg.max(),
clip=True)
fs_cmap = get_freesurfer_cmap(sub_cortical)
label_mapper = cm.ScalarMappable(norm=normalize_labels, cmap=fs_cmap)
ax = ax.flatten()
ax_counter = 0
for dim_index in range(3):
for slice_num in slices[dim_index]:
plt.sca(ax[ax_counter])
ax_counter = ax_counter + 1
slice_mri = get_axis(mri, dim_index, slice_num)
slice_seg = get_axis(seg, dim_index, slice_num)
# # masking data to set no-value pixels to transparent
# seg_background = np.isclose(slice_seg, 0.0)
# slice_seg = np.ma.masked_where(seg_background, slice_seg)
# slice_mri = np.ma.masked_where(np.logical_not(seg_background), slice_mri)
seg_rgb = label_mapper.to_rgba(slice_seg)
plt.imshow(seg_rgb, **display_params_seg)
plt.imshow(slice_mri, **display_params_mri)
plt.axis('off')
# plt.subplots_adjust(wspace=0.0, hspace=0.0)
plt.subplots_adjust(left=0.01,
right=0.99,
bottom=0.01,
top=0.99,
wspace=0.05,
hspace=0.02)
# fig.tight_layout()
if output_path is not None:
output_path = output_path.replace(' ', '_')
fig.savefig(output_path + '.png', bbox_inches='tight')
# plt.close()
return fig
def collage(img_spec,
num_rows=2,
num_cols=6,
rescale_method='global',
cmap='gray',
annot=None,
padding=5,
bkground_thresh=None,
output_path=None,
figsize=None,
**kwargs):
"Produces a collage of various slices from different orientations in the given 3D image"
num_rows, num_cols, padding = check_params(num_rows, num_cols, padding)
img = read_image(img_spec, bkground_thresh=bkground_thresh)
img = crop_image(img, padding)
img, (min_value, max_value) = check_rescaling_collage(img,
rescale_method,
return_extrema=True)
num_slices_per_view = num_rows * num_cols
slices = pick_slices(img, num_slices_per_view)
plt.style.use('dark_background')
num_axes = 3
if figsize is None:
figsize = [3 * num_axes * num_rows, 3 * num_cols]
fig, ax = plt.subplots(num_axes * num_rows, num_cols, figsize=figsize)
# displaying some annotation text if provided
if annot is not None:
fig.suptitle(annot, backgroundcolor='black', color='g')
display_params = dict(interpolation='none',
cmap=cmap,
aspect='equal',
origin='lower',
vmin=min_value,
vmax=max_value)
ax = ax.flatten()
ax_counter = 0
for dim_index in range(3):
for slice_num in slices[dim_index]:
plt.sca(ax[ax_counter])
ax_counter = ax_counter + 1
slice1 = get_axis(img, dim_index, slice_num)
# slice1 = crop_image(slice1, padding)
plt.imshow(slice1, **display_params)
plt.axis('off')
fig.tight_layout()
if output_path is not None:
output_path = output_path.replace(' ', '_')
fig.savefig(output_path + '.png', bbox_inches='tight')
# plt.close()
return fig