1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
|
#!/usr/bin/env python3
# emacs: -*- mode: python; py-indent-offset: 4; indent-tabs-mode: nil -*-
# vi: set ft=python sts=4 ts=4 sw=4 et:
__doc__ = """
This script contains a quick demo on a multi-subject parcellation on a toy 2D
example.
Note how the middle parcels adapt to the individual configuration.
Needs matplotlib
"""
print(__doc__)
import numpy as np
try:
import matplotlib.pyplot as plt
except ImportError:
raise RuntimeError("This script needs the matplotlib library")
import nipy.labs.spatial_models.discrete_domain as dom
import nipy.labs.spatial_models.hierarchical_parcellation as hp
import nipy.labs.utils.simul_multisubject_fmri_dataset as simul
# step 1: generate some synthetic data
n_subj = 10
shape = (60, 60)
pos = 3 * np.array([[6, 7],
[10, 10],
[15, 10]])
ampli = np.array([5, 7, 6])
sjitter = 6.0
dataset = simul.surrogate_2d_dataset(n_subj=n_subj, shape=shape, pos=pos,
ampli=ampli, width=10.0)
# dataset represents 2D activation images from n_subj subjects,
# step 2 : prepare all the information for the parcellation
nbparcel = 10
ldata = np.reshape(dataset, (n_subj, np.prod(shape), 1))
domain = dom.grid_domain_from_shape(shape)
# step 3 : run the algorithm
Pa = hp.hparcel(domain, ldata, nbparcel, mu=3.0)
# note: play with mu to change the 'stiffness of the parcellation'
# step 4: look at the results
Label = np.array([np.reshape(Pa.individual_labels[:, s], shape)
for s in range(n_subj)])
plt.figure(figsize=(8, 4))
plt.title('Input data')
for s in range(n_subj):
plt.subplot(2, 5, s + 1)
plt.imshow(dataset[s], interpolation='nearest')
plt.axis('off')
plt.figure(figsize=(8, 4))
plt.title('Resulting parcels')
for s in range(n_subj):
plt.subplot(2, 5, s+1)
plt.imshow(Label[s], interpolation='nearest', vmin=-1, vmax=nbparcel)
plt.axis('off')
plt.show()
|
