def make_mask_from_points(pois, mask_file, new_mask_file):
m = xndarray.load(mask_file)
new_m = xndarray.xndarray_like(m)
for poi in pois:
i,j,k = poi['sagittal'], poi['coronal'], poi['axial']
new_m.data[i,j,k] = 1
new_m.save(new_mask_file)
def test_voronoi_with_seeds(self):
import os.path as op
from pyhrf.ndarray import xndarray
import pyhrf
fn = 'subj0_parcellation.nii.gz'
mask_file = pyhrf.get_data_file_name(fn)
orientation = ['axial', 'coronal', 'sagittal']
seeds = xndarray.xndarray_like(
xndarray.load(mask_file)).reorient(orientation)
seed_coords = np.array([[24, 35, 8], # axial, coronal, sagittal
[27, 35, 5],
[27, 29, 46],
[31, 28, 46]])
seeds.data[:] = 0
seeds.data[tuple(seed_coords.T)] = 1
seed_file = op.join(self.tmp_dir, 'voronoi_seeds.nii')
seeds.save(seed_file, set_MRI_orientation=True)
output_file = op.join(self.tmp_dir, 'voronoi_parcellation.nii')
cmd = 'pyhrf_parcellate_spatial %s -m voronoi -c %s -o %s -v %d' \
% (mask_file, seed_file, output_file, logger.getEffectiveLevel())
if os.system(cmd) != 0:
raise Exception('"' + cmd + '" did not execute correctly')
logger.info('cmd: %s', cmd)
assert op.exists(output_file)
parcellation = xndarray.load(output_file)
n_parcels = len(np.unique(parcellation.data)) - 1
self.assertEqual(n_parcels, len(seed_coords))
def make_parcellation(subject, dest_dir="parcellation", roi_mask_file=None):
"""
Perform a functional parcellation from input fmri data
Return: parcellation file name (str)
"""
# Loading names for folders and files
# - T maps (input)
# func_files = glob(op.join(op.join(op.join('./', subject), \
# 't_maps'), 'BOLD*nii'))
# func_files = glob(op.join('./', subject, 'ASLf', 'spm_analysis', \
# 'Tmaps*img'))
func_files = glob(op.join("./", subject, "ASLf", "spm_analysis", "spmT*img"))
print "Tmap files: ", func_files
# - Mask (input)
# spm_mask_file = op.join(spm_maps_dir, 'mask.img')
mask_dir = op.join("./", subject, "preprocessed_data")
if not op.exists(mask_dir):
os.makedirs(mask_dir)
mask_file = op.join(mask_dir, "mask.nii")
mask = op.join(mask_dir, "rcut_tissue_mask.nii")
volume = op.join("./", subject, "ASLf", "funct", "coregister", "mean" + subject + "_ASLf_correctionT1_0001.nii")
make_mask(mask, volume, mask_file)
# - parcellation (output)
parcellation_dir = op.join("./", subject, dest_dir)
if not op.exists(parcellation_dir):
os.makedirs(parcellation_dir)
pfile = op.join(parcellation_dir, "parcellation_func.nii")
# Parcellation
from pyhrf.parcellation import make_parcellation_from_files
# make_parcellation_from_files(func_files, mask_file, pfile,
# nparcels=200, method='ward_and_gkm')
# Masking with a ROI so we just consider parcels inside
# a certain area of the brain
# if roi_mask_file is not None:
if 0: # for ip in np.array([11, 51, 131, 194]):
# ip = 200
# print 'Masking parcellation with roi_mask_file: ', roi_mask_file
print "Masking ROI: ", ip
pfile_masked = op.join(parcellation_dir, "parcellation_func_masked_roi" + str(ip) + ".nii")
from pyhrf.ndarray import xndarray
parcellation = xndarray.load(pfile)
# m = xndarray.load(roi_mask_file)
# parcels_to_keep = np.unique(parcellation.data * m.data)
masked_parcellation = xndarray.xndarray_like(parcellation)
# for ip in parcels_to_keep:
# masked_parcellation.data[np.where(parcellation.data==ip)] = ip
masked_parcellation.data[np.where(parcellation.data == ip)] = ip
masked_parcellation.save(pfile_masked)
from pyhrf.ndarray import xndarray
for tmap in func_files:
func_file_i = xndarray.load(tmap)
func_data = func_file_i.data
# func_file_i.data[np.where(func_data<0.1*func_data.max())] = 0
parcellation = xndarray.load(pfile)
print func_data.max()
print func_data.argmax()
# ip = parcellation.data[func_data.argmax()]
ip = parcellation.data[np.unravel_index(func_data.argmax(), parcellation.data.shape)]
print ip.shape
masked_parcellation = xndarray.xndarray_like(parcellation)
print masked_parcellation.data.shape
print parcellation.data.shape
masked_parcellation.data[np.where(parcellation.data == ip)] = ip
masked_parcellation.save(tmap[:-4] + "_parcelmax.nii")
return pfile
def generate_noise(stim_induced_signal, noise_var):
noise = np.random.randn(*stim_induced_signal.data.shape) * noise_var**.5
return xndarray.xndarray_like(stim_induced_signal, data=noise)
def make_parcellation(subject, dest_dir='parcellation', roi_mask_file=None):
"""
Perform a functional parcellation from input fmri data
Return: parcellation file name (str)
"""
# Loading names for folders and files
# - T maps (input)
#func_files = glob(op.join(op.join(op.join('./', subject), \
# 't_maps'), 'BOLD*nii'))
#func_files = glob(op.join('./', subject, 'ASLf', 'spm_analysis', \
# 'Tmaps*img'))
func_files = glob(op.join('./', subject, 'ASLf', 'spm_analysis', \
'spmT*img'))
print 'Tmap files: ', func_files
# - Mask (input)
#spm_mask_file = op.join(spm_maps_dir, 'mask.img')
mask_dir = op.join('./', subject, 'preprocessed_data')
if not op.exists(mask_dir): os.makedirs(mask_dir)
mask_file = op.join(mask_dir, 'mask.nii')
mask = op.join(mask_dir, 'rcut_tissue_mask.nii')
volume = op.join('./', subject, 'ASLf', 'funct', 'coregister', \
'mean' + subject + '_ASLf_correctionT1_0001.nii')
make_mask(mask, volume, mask_file)
# - parcellation (output)
parcellation_dir = op.join('./', subject, dest_dir)
if not op.exists(parcellation_dir): os.makedirs(parcellation_dir)
pfile = op.join(parcellation_dir, 'parcellation_func.nii')
# Parcellation
from pyhrf.parcellation import make_parcellation_from_files
#make_parcellation_from_files(func_files, mask_file, pfile,
# nparcels=200, method='ward_and_gkm')
# Masking with a ROI so we just consider parcels inside
# a certain area of the brain
#if roi_mask_file is not None:
if 0: #for ip in np.array([11, 51, 131, 194]):
#ip = 200
#print 'Masking parcellation with roi_mask_file: ', roi_mask_file
print 'Masking ROI: ', ip
pfile_masked = op.join(parcellation_dir, 'parcellation_func_masked_roi' + str(ip) + '.nii')
from pyhrf.ndarray import xndarray
parcellation = xndarray.load(pfile)
#m = xndarray.load(roi_mask_file)
#parcels_to_keep = np.unique(parcellation.data * m.data)
masked_parcellation = xndarray.xndarray_like(parcellation)
#for ip in parcels_to_keep:
# masked_parcellation.data[np.where(parcellation.data==ip)] = ip
masked_parcellation.data[np.where(parcellation.data==ip)] = ip
masked_parcellation.save(pfile_masked)
from pyhrf.ndarray import xndarray
for tmap in func_files:
func_file_i = xndarray.load(tmap)
func_data = func_file_i.data
#func_file_i.data[np.where(func_data<0.1*func_data.max())] = 0
parcellation = xndarray.load(pfile)
print func_data.max()
print func_data.argmax()
#ip = parcellation.data[func_data.argmax()]
ip = parcellation.data[np.unravel_index(func_data.argmax(), parcellation.data.shape)]
print ip.shape
masked_parcellation = xndarray.xndarray_like(parcellation)
print masked_parcellation.data.shape
print parcellation.data.shape
masked_parcellation.data[np.where(parcellation.data==ip)] = ip
masked_parcellation.save(tmap[:-4] + '_parcelmax.nii')
return pfile