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
64
65
66
67
68
69
70
71
|
# emacs: -*- mode: python; py-indent-offset: 4; indent-tabs-mode: nil -*-
# vi: set ft=python sts=4 ts=4 sw=4 et:
__doc__ = \
"""
Example of a script that crates a 'hierarchical roi' structure
from the blob model of an image
Author: Bertrand Thirion, 2008-2009
"""
print __doc__
import numpy as np
import nipy.neurospin.spatial_models.hroi as hroi
import nipy.neurospin.utils.simul_multisubject_fmri_dataset as simul
import nipy.neurospin.graph.field as ff
##############################################################################
# simulate the data
dimx = 60
dimy = 60
pos = 2*np.array([[6,7],[10,10],[15,10]])
ampli = np.array([3,4,4])
dataset = simul.surrogate_2d_dataset(nbsubj=1, dimx=dimx, dimy=dimy, pos=pos,
ampli=ampli, width=10.0).squeeze()
dataset = np.reshape(dataset, (dimx, dimy,1))
ref_dim = (dimx,dimy,1)
xyz = np.array(np.where(dataset)).T
nbvox = np.size(xyz, 0)
affine = np.eye(4)
shape = (dimx,dimy,1)
# create the field strcture that encodes image topology
Fbeta = ff.Field(nbvox)
Fbeta.from_3d_grid(xyz.astype(np.int), 18)
beta = np.reshape(dataset,(nbvox,1))
Fbeta.set_field(beta)
nroi = hroi.NROI_from_field(Fbeta, affine, shape, xyz, th=2.0, smin = 5)
if nroi != None:
n1 = nroi.copy()
n2 = nroi.reduce_to_leaves()
td = n1.depth_from_leaves()
a = np.argmax(td)
lv = n1.rooted_subtree(a)
u = nroi.cc()
u = np.nonzero(u == u[0])[0]
nroi.set_discrete_feature_from_index('activation',beta)
nroi.discrete_to_roi_features('activation')
label = -np.ones((dimx,dimy))
for k in range(nroi.k):
label[nroi.xyz[k][:,0],nroi.xyz[k][:,1]]=k
import matplotlib.pylab as mp
mp.figure()
mp.subplot(1,2,1)
mp.imshow(np.squeeze(dataset))
mp.title('Input map')
mp.axis('off')
mp.subplot(1,2,2)
mp.title('Nested Rois')
mp.imshow(label,interpolation='Nearest')
mp.axis('off')
mp.show()
|
