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
|
#!/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:
"""Example running a parcel-based second-level analysis from a set of
first-level effect images.
This script takes as input a directory path that contains first-level
images in nifti format, as well as a group mask image and a
parcellation image (such as the AAL atlas, 'ROI_MNI_V4.nii', see
http://www.gin.cnrs.fr/spip.php?article217). All images are assumed to
be in a common reference space, e.g. the MNI/Talairach space.
It outputs three images:
* tmap.nii.gz, a `t-statistic` image similar to a SPM-like second-level
t-map, except it is derived under an assumption of localization
uncertainty in reference space.
* parcel_mu.nii.gz, an image that maps each voxel to the estimated
population effect in the parcel it belongs to.
* parcel_prob.nii.gz, an image that maps each voxel to the probability
that the population effect in the parcel it belongs to is
positive-valued.
See the `nipy.algorithms.group.ParcelAnalysis` class for more general
usage information.
"""
from argparse import ArgumentParser
from glob import glob
from os.path import join
from nipy import load_image
from nipy.algorithms.group import parcel_analysis
# Parse command line
description = 'Run a parcel-based second-level analysis from a set of\
first-level effect images.'
parser = ArgumentParser(description=description)
parser.add_argument('con_path', metavar='con_path',
help='directory where 1st-level images are to be found')
parser.add_argument('msk_file', metavar='msk_file',
help='group mask file')
parser.add_argument('parcel_file', metavar='parcel_file',
help='parcellation image file')
args = parser.parse_args()
# Load first-level images
con_files = glob(join(args.con_path, '*.nii'))
con_imgs = [load_image(f) for f in con_files]
# Load group mask
msk_img = load_image(args.msk_file)
# Load parcellation
parcel_img = load_image(args.parcel_file)
# Run parcel analysis and write output images in the current directory
effect_img, proba_img = parcel_analysis(con_imgs, parcel_img,
msk_img=msk_img, fwhm=8,
res_path='.')
|
