import os
from nipype.interfaces.base import (CommandLine, CommandLineInputSpec,
TraitedSpec)
from nipype.interfaces.base import traits, File
# T1xT2BiasFieldCorrection
class T1xT2BiasFieldCorrectionInputSpec(CommandLineInputSpec):
t1_file = File(
exists=True,
desc='Whole-head T1w image',
mandatory=True, position=0, argstr="-t1 %s")
t2_file = File(
exists=True,
desc='Whole-head T2w image (use -aT2 if T2w image is not in the T1w \
space)',
mandatory=True, position=1, argstr="-t2 %s")
os = traits.String(
"_debiased", usedefault=True,
desc="Suffix for the bias field corrected images (default is \
\"_debiased\")",
argstr="-os %s", mandatory=False)
aT2 = traits.Bool(
False, usedefault=True,
argstr="-aT2",
desc="Will coregrister T2w to T1w using flirt. Output will have the\
suffix provided. Will only work for spatially close images.",
mandatory=False)
# as -> opt_as, as is already a part of python keywords...
opt_as = traits.String(
"-in-T1w", usedefault=True, argstr="-as %s",
desc="Suffix for T2w to T1w registration \
(\"-in-T1w\" if not specified)",
mandatory=False)
s = traits.Int(
4, usedefault=True,
desc='size of gauss kernel in mm when performing mean filtering \
(default=4)',
argstr="-s %d", mandatory=False)
# exists = True ???
b = traits.File(
argstr="-b %s",
desc="Brain mask file. Will also output bias corrected brain files \
with the format \"output_prefix_brain.nii.gz\"",
mandatory=False)
bet = traits.Int(
0, usedefault=True,
desc='Will try to "smart" BET the anat files to get a brain mask: n =\
the number of iterations BET will be run to find center of gravity\
(default=0, will not BET if option -b has been specified).',
argstr="-bet %d", mandatory=False)
bs = traits.String(
"_BET", usedefault=True,
desc="Suffix for the BET masked images (default is \"_BET\")",
argstr="-bs %s", mandatory=False)
f = traits.Float(
0.5, usedefault=True,
desc='-f options of BET: fractional intensity threshold (0->1); \
default=0.5; smaller values give larger brain outline estimates',
argstr="-f %f", mandatory=False)
g = traits.Float(
0.0, usedefault=True, desc='-g options of BET:\
vertical gradient in fractional intensity threshold (-1->1);\
default=0; positive values give larger brain outline at\
bottom, smaller at top',
argstr="-g %f", mandatory=False)
k = traits.Bool(
False, usedefault=True, argstr="-k",
desc="Will keep temporary files",
mandatory=False)
p = traits.String(
desc="Prefix for running FSL functions\
(can be a path or just a prefix)",
argstr="-p %s", mandatory=False)
class T1xT2BiasFieldCorrectionOutputSpec(TraitedSpec):
t1_debiased_file = File(
exists=True,
desc="debiased T1")
t2_debiased_file = File(
exists=True,
desc="debiased T2")
t2_coreg_file = File(
desc="T2 on T1")
t1_debiased_brain_file = File(
desc="debiased bet T1")
t2_debiased_brain_file = File(
desc="debiased bet T2")
debiased_mask_file = File(
desc="debiased bet mask")
[docs]class T1xT2BiasFieldCorrection(CommandLine):
"""
Description: Bias field correction using T1w & T2w images.
Provides an attempt of brain extration if wanted.
Inputs:
Mandatory:
t1_file
Whole-head T1w image
t2_file
Whole-head T2w image (use -aT2 if T2w image is not in the T1w
space)
Optional:
os
Suffix for the bias field corrected images (default is
"_debiased")
aT2
Will coregrister T2w to T1w using flirt. Output will have the\
suffix provided. Will only work for spatially close images.
opt_as
Suffix for T2w to T1w registration ("-in-T1w" if not specified)
s
size of gauss kernel in mm when performing mean filtering \
(default=4)argstr="-s %d", mandatory=False)
b
Brain mask file. Will also output bias corrected brain files \
with the format "output_prefix_brain.nii.gz"
bet
Will try to "smart" BET the anat files to get a brain mask:
n = the number of iterations BET will be run to find center of
gravity (default=0, will not BET if option -b has been
specified).
bs
Suffix for the BET masked images (default is "_BET")
f
-f options of BET: fractional intensity threshold (0->1); \
default=0.5; smaller values give larger brain outline
estimates
g
-g options of BET:\
vertical gradient in fractional intensity threshold
(-1->1); default=0; positive values give larger brain
outline at bottom, smaller at top
k
Will keep temporary files
p
Prefix for running FSL functions\
(can be a path or just a prefix)
Outputs:
t1_debiased_file
debiased T1
t2_debiased_file
debiased T2
t2_coreg_file
T2 on T1
t1_debiased_brain_file
debiased bet T1
t2_debiased_brain_file
debiased bet T2
debiased_mask_file
debiased bet mask
"""
input_spec = T1xT2BiasFieldCorrectionInputSpec
output_spec = T1xT2BiasFieldCorrectionOutputSpec
package_directory = os.path.dirname(os.path.abspath(__file__))
_cmd = 'bash {}/../bash/T1xT2BiasFieldCorrection.sh'.format(
package_directory)
def _list_outputs(self):
import os
from nipype.utils.filemanip import split_filename as split_f
outputs = self._outputs().get()
t1_path, t1_fname, ext = split_f(self.inputs.t1_file)
t2_path, t2_fname, ext = split_f(self.inputs.t2_file)
# !!!!warning, in Regis bash, only .nii.gz are handled
outputs["t1_debiased_file"] = os.path.abspath(
t1_fname + self.inputs.os + ".nii.gz")
outputs["t2_debiased_file"] = os.path.abspath(
t2_fname + self.inputs.os + ".nii.gz")
if self.inputs.aT2:
outputs["t2_coreg_file"] = os.path.abspath(
t2_fname + self.inputs.opt_as + ".nii.gz")
if self.inputs.bet:
outputs["t1_debiased_brain_file"] = os.path.abspath(
t1_fname + self.inputs.os + self.inputs.bs + ".nii.gz")
outputs["t2_debiased_brain_file"] = os.path.abspath(
t2_fname + self.inputs.os + self.inputs.bs + ".nii.gz")
outputs["debiased_mask_file"] = os.path.abspath(
t1_fname + self.inputs.os + self.inputs.bs + "_mask.nii.gz")
elif self.inputs.b:
outputs["t1_debiased_brain_file"] = os.path.abspath(
t1_fname + self.inputs.os + "_brain.nii.gz")
outputs["t2_debiased_brain_file"] = os.path.abspath(
t2_fname + self.inputs.os + "_brain.nii.gz")
return outputs
