def topup(blip_up_file,
blip_down_file,
blip_up_phase_enc_dir,
blip_down_phase_enc_dir,
outdir,
fsl_sh,
apply_to=None,
unwarp_direction=None,
dwell_time=None,
verbose=0):
""" Bias field correction using topup.
Note: we do not set the total readout times for both acquisitons, and we
expect that the readout time is identical for all acquisitions. This
parameter simply serves to ensure that the estimated field is correctly
scaled in Hz, but it doesn't affect the result in terms of undistorted
images. We set it to 1, meaning that the field will be scaled as 'voxels
displacement' instead.
Parameters
----------
blip_up_file: str
path to EPI file acquired in opposite phase enc. directions: blip up.
blip_down_file: str
path to EPI file acquired in opposite phase enc. directions: blip down.
blip_up_phase_enc_dir: str
the phase enc. direction for the blip up.
blip_down_phase_enc_dir: str
the phase enc. direction for the blip down.
phase_enc_dirs: list of str
the phase enc. directions.
outdir: str
directory for outputs.
fsl_sh: str
path to fsl setup sh file.
apply_to: str, default None
apply the topup correction to the EPI volume.
unwarp_direction: str, default None
apply topup correction in this direction.
dwell_time: float, default None
set the EPI dwell time per phase-encode line, same as echo spacing (ms).
verbose: int, default 0
control the verbosity level.
Returns
-------
fieldmap_hz_file: str
the fieldmap in Hz
unwarped_epi_file: str
unwarped EPI file.
"""
# Call topup
fieldmap_hz_file, _, _ = _topup(
b0s=[blip_up_file, blip_down_file],
phase_enc_dirs=[blip_up_phase_enc_dir, blip_down_phase_enc_dir],
readout_time=1,
outroot=outdir,
apply_to=None,
fsl_sh=fsl_sh)
# Convert the fieldmap in rad/s to Hz
fieldmap_file = fieldmap_hz_file.replace(".nii.gz", "_rads.nii.gz")
cmd = ["fslmaths", fieldmap_hz_file, "-mul", "6.28", fieldmap_file]
fslprocess = FSLWrapper(shfile=fsl_sh)
fslprocess(cmd=cmd)
# Update env
fslenv = environment(fsl_sh)
if (fslenv["FSLDIR"] != os.environ.get("FSLDIR", "")):
os.environ = concat_environment(os.environ, fslenv)
# Apply topup
# Unwarping an input image (fieldmap in rad/s is known)
unwarp_direction = DIR_MAP[unwarp_direction]
unwarped_epi_file = os.path.join(outdir, "unwarped_epi_file.nii.gz")
fugue = FUGUE()
fugue.inputs.in_file = apply_to
fugue.inputs.fmap_in_file = fieldmap_file
fugue.inputs.unwarp_direction = unwarp_direction
fugue.inputs.unwarped_file = unwarped_epi_file
fugue.inputs.dwell_time = dwell_time * 1e-3
fugue.inputs.output_type = "NIFTI_GZ"
if verbose > 0:
print(fugue.cmdline)
returncode = fugue.run()
return fieldmap_hz_file, unwarped_epi_file
def fugue(epi_file,
phase_file,
magnitude_file,
delta_te,
dwell_time,
unwarp_direction,
manufacturer,
outdir,
fsl_sh,
verbose=0):
""" Unwarping of an EPI image based on fieldmap data using fugue.
Note: Brain extraction of the EPI image is very important and must be
tight - that is, it must exclude all non-brain voxels and any voxels with
only a small partial volume contribution. The reason for this is that
these areas are normally very noisy in the phase image. The exclusion
of brain voxels is actually fine and will have no repercussions, since the
fieldmap is extrapolated beyond this mask.
Note: If parallel acceleration is used in the EPI acquisition then the
*effective* echo spacing (dwell time) is the actual echo spacing between
acquired lines in k-space divided by the acceleration factor.
Parameters
----------
epi_file: str
the EPI file to unwarp.
phase_file: str
the phase image in the EPI space.
magnitude_file: str
the magnitude image in the EPI space.
delta_te: float
the echo time difference of the fieldmap sequence - find this out form
the operator (defaults are *usually* 2.46ms on SIEMENS).
dwell_time: float
set the EPI dwell time per phase-encode line, same as echo spacing (ms).
unwarp_direction: str
specifies direction of warping: ('x' or 'y' or 'z' or 'x-' or 'y-'
or 'z-')
manufacturer: str
must be SIEMENS.
outdir: str
directory for outputs.
fsl_sh: str
path to fsl setup sh file.
verbose: int, default 0
control the verbosity level.
Returns
-------
magnitude_brain_mask_file: str
the brain mask.
vsm_file: str
voxel shift map file.
unwarped_epi_file: str
unwarped EPI file.
"""
# Extract brain
avgmagnitude_file = os.path.join(outdir, "avgmag.nii.gz")
average_timeserie(magnitude_file, avgmagnitude_file, fslconfig=fsl_sh)
outputs = bet2(avgmagnitude_file, outdir, mask=True, f=0.35, shfile=fsl_sh)
#outputs = bet2(outputs[0], outdir, mask=True, f=0.6, shfile=fsl_sh)
magnitude_brain_file, magnitude_brain_mask_file = outputs[:2]
# Update env
fslenv = environment(fsl_sh)
if (fslenv["FSLDIR"] != os.environ.get("FSLDIR", "")):
os.environ = concat_environment(os.environ, fslenv)
# Prepare a rad/s fieldmap
fieldmap_file = os.path.join(outdir, "fieldmap.nii.gz")
fieldmap_file, fieldmap_hz_file = fsl_prepare_fieldmap(
manufacturer=manufacturer.upper(),
phase_file=phase_file,
brain_magnitude_file=magnitude_brain_file,
output_file=fieldmap_file,
delta_te=str(delta_te),
fsl_sh=fsl_sh)
# Computing the vsm (unwrapped phase map is known)
unwarp_direction = DIR_MAP[unwarp_direction]
vsm_file = os.path.join(outdir, "vsm.nii.gz")
fugue = FUGUE()
fugue.inputs.fmap_in_file = fieldmap_file
fugue.inputs.mask_file = magnitude_brain_mask_file
fugue.inputs.dwell_time = dwell_time * 1e-3
fugue.inputs.asym_se_time = delta_te * 1e-3
fugue.inputs.unwarp_direction = unwarp_direction
fugue.inputs.save_unmasked_shift = True
fugue.inputs.shift_out_file = vsm_file
fugue.inputs.smooth2d = 2
fugue.inputs.output_type = "NIFTI_GZ"
fugue.environ = {}
if verbose > 0:
print(fugue.cmdline)
returncode = fugue.run()
#outputs = returncode.outputs.get()
#vsm_file = outputs["shift_out_file"]
# Unwarping an input image (shift map is known)
unwarped_epi_file = os.path.join(outdir, "unwarped_epi_file.nii.gz")
fugue = FUGUE()
fugue.inputs.in_file = epi_file
fugue.inputs.mask_file = magnitude_brain_mask_file
fugue.inputs.shift_in_file = vsm_file
fugue.inputs.unwarp_direction = unwarp_direction
fugue.inputs.unwarped_file = unwarped_epi_file
fugue.inputs.output_type = "NIFTI_GZ"
if verbose > 0:
print(fugue.cmdline)
returncode = fugue.run()
#outputs = returncode.outputs.get()
#unwarped_epi_file = outputs["unwarped_file"]
return magnitude_brain_mask_file, vsm_file, unwarped_epi_file