algorithms.misc¶
AddCSVColumn¶
Short interface to add an extra column and field to a text file
Example¶
>>> from nipype.algorithms import misc
>>> addcol = misc.AddCSVColumn()
>>> addcol.inputs.in_file = 'degree.csv'
>>> addcol.inputs.extra_column_heading = 'group'
>>> addcol.inputs.extra_field = 'male'
>>> addcol.run()
Inputs:
[Mandatory]
in_file: (an existing file name)
Input comma-separated value (CSV) files
[Optional]
extra_column_heading: (a unicode string)
New heading to add for the added field.
extra_field: (a unicode string)
New field to add to each row. This is useful for saving the group or
subject ID in the file.
out_file: (a file name, nipype default value: extra_heading.csv)
Output filename for merged CSV file
Outputs:
csv_file: (a file name)
Output CSV file containing columns
AddCSVRow¶
Simple interface to add an extra row to a csv file
Note
Requires pandas
Warning
Multi-platform thread-safe execution is possible with lockfile. Please recall that (1) this module is alpha software; and (2) it should be installed for thread-safe writing. If lockfile is not installed, then the interface is not thread-safe.
Example¶
>>> from nipype.algorithms import misc
>>> addrow = misc.AddCSVRow()
>>> addrow.inputs.in_file = 'scores.csv'
>>> addrow.inputs.si = 0.74
>>> addrow.inputs.di = 0.93
>>> addrow.inputs.subject_id = 'S400'
>>> addrow.inputs.list_of_values = [ 0.4, 0.7, 0.3 ]
>>> addrow.run()
Inputs:
[Mandatory]
in_file: (a file name)
Input comma-separated value (CSV) files
[Optional]
_outputs: (a dictionary with keys which are any value and with values
which are any value, nipype default value: {})
ignore_exception: (a boolean, nipype default value: False)
Print an error message instead of throwing an exception in case the
interface fails to run
Outputs:
csv_file: (a file name)
Output CSV file containing rows
AddNoise¶
Corrupts with noise the input image
Example¶
>>> from nipype.algorithms.misc import AddNoise
>>> noise = AddNoise()
>>> noise.inputs.in_file = 'T1.nii'
>>> noise.inputs.in_mask = 'mask.nii'
>>> noise.snr = 30.0
>>> noise.run()
Inputs:
[Mandatory]
bg_dist: (u'normal' or u'rayleigh', nipype default value: normal)
desired noise distribution, currently only normal is implemented
dist: (u'normal' or u'rician', nipype default value: normal)
desired noise distribution
in_file: (an existing file name)
input image that will be corrupted with noise
[Optional]
in_mask: (an existing file name)
input mask, voxels outside this mask will be considered background
out_file: (a file name)
desired output filename
snr: (a float, nipype default value: 10.0)
desired output SNR in dB
Outputs:
out_file: (an existing file name)
corrupted image
CalculateNormalizedMoments¶
Calculates moments of timeseries.
Example¶
>>> from nipype.algorithms import misc
>>> skew = misc.CalculateNormalizedMoments()
>>> skew.inputs.moment = 3
>>> skew.inputs.timeseries_file = 'timeseries.txt'
>>> skew.run()
Inputs:
[Mandatory]
moment: (an integer (int or long))
Define which moment should be calculated, 3 for skewness, 4 for
kurtosis.
timeseries_file: (an existing file name)
Text file with timeseries in columns and timepoints in rows,
whitespace separated
[Optional]
Outputs:
moments: (a list of items which are a float)
Moments
CreateNifti¶
Creates a nifti volume
Inputs:
[Mandatory]
data_file: (an existing file name)
ANALYZE img file
header_file: (an existing file name)
corresponding ANALYZE hdr file
[Optional]
affine: (an array)
affine transformation array
ignore_exception: (a boolean, nipype default value: False)
Print an error message instead of throwing an exception in case the
interface fails to run
Outputs:
nifti_file: (an existing file name)
Distance¶
Calculates distance between two volumes.
Deprecated since version 0.10.0: Use nipype.algorithms.metrics.Distance instead.
Inputs:
[Mandatory]
volume1: (an existing file name)
Has to have the same dimensions as volume2.
volume2: (an existing file name)
Has to have the same dimensions as volume1.
[Optional]
ignore_exception: (a boolean, nipype default value: False)
Print an error message instead of throwing an exception in case the
interface fails to run
mask_volume: (an existing file name)
calculate overlap only within this mask.
method: (u'eucl_min' or u'eucl_cog' or u'eucl_mean' or u'eucl_wmean'
or u'eucl_max', nipype default value: eucl_min)
""eucl_min": Euclidean distance between two closest points
"eucl_cog": mean Euclidian distance between the Center of Gravity of
volume1 and CoGs of volume2 "eucl_mean": mean Euclidian minimum
distance of all volume2 voxels to volume1 "eucl_wmean": mean
Euclidian minimum distance of all volume2 voxels to volume1 weighted
by their values "eucl_max": maximum over minimum Euclidian distances
of all volume2 voxels to volume1 (also known as the Hausdorff
distance)
Outputs:
distance: (a float)
histogram: (a file name)
point1: (an array with shape (3,))
point2: (an array with shape (3,))
FuzzyOverlap¶
Calculates various overlap measures between two maps, using a fuzzy definition.
Deprecated since version 0.10.0: Use nipype.algorithms.metrics.FuzzyOverlap instead.
Inputs:
[Mandatory]
in_ref: (a list of items which are an existing file name)
Reference image. Requires the same dimensions as in_tst.
in_tst: (a list of items which are an existing file name)
Test image. Requires the same dimensions as in_ref.
[Optional]
ignore_exception: (a boolean, nipype default value: False)
Print an error message instead of throwing an exception in case the
interface fails to run
out_file: (a file name, nipype default value: diff.nii)
alternative name for resulting difference-map
weighting: (u'none' or u'volume' or u'squared_vol', nipype default
value: none)
'none': no class-overlap weighting is performed. 'volume': computed
class-overlaps are weighted by class volume 'squared_vol': computed
class-overlaps are weighted by the squared volume of the class
Outputs:
class_fdi: (a list of items which are a float)
Array containing the fDIs of each computed class
class_fji: (a list of items which are a float)
Array containing the fJIs of each computed class
dice: (a float)
Fuzzy Dice Index (fDI), all the classes
diff_file: (an existing file name)
resulting difference-map of all classes, using the chosen weighting
jaccard: (a float)
Fuzzy Jaccard Index (fJI), all the classes
Gunzip¶
Gunzip wrapper
Inputs:
[Mandatory]
in_file: (an existing file name)
[Optional]
ignore_exception: (a boolean, nipype default value: False)
Print an error message instead of throwing an exception in case the
interface fails to run
Outputs:
out_file: (an existing file name)
Matlab2CSV¶
Simple interface to save the components of a MATLAB .mat file as a text file with comma-separated values (CSVs).
CSV files are easily loaded in R, for use in statistical processing. For further information, see cran.r-project.org/doc/manuals/R-data.pdf
Example¶
>>> from nipype.algorithms import misc
>>> mat2csv = misc.Matlab2CSV()
>>> mat2csv.inputs.in_file = 'cmatrix.mat'
>>> mat2csv.run()
Inputs:
[Mandatory]
in_file: (an existing file name)
Input MATLAB .mat file
[Optional]
reshape_matrix: (a boolean, nipype default value: True)
The output of this interface is meant for R, so matrices will be
reshaped to vectors by default.
Outputs:
csv_files: (a list of items which are a file name)
MergeCSVFiles¶
This interface is designed to facilitate data loading in the R environment. It takes input CSV files and merges them into a single CSV file. If provided, it will also incorporate column heading names into the resulting CSV file.
CSV files are easily loaded in R, for use in statistical processing. For further information, see cran.r-project.org/doc/manuals/R-data.pdf
Example¶
>>> from nipype.algorithms import misc
>>> mat2csv = misc.MergeCSVFiles()
>>> mat2csv.inputs.in_files = ['degree.mat','clustering.mat']
>>> mat2csv.inputs.column_headings = ['degree','clustering']
>>> mat2csv.run()
Inputs:
[Mandatory]
in_files: (a list of items which are an existing file name)
Input comma-separated value (CSV) files
[Optional]
column_headings: (a list of items which are a unicode string)
List of column headings to save in merged CSV file (must be equal to
number of input files). If left undefined, these will be pulled from
the input filenames.
extra_column_heading: (a unicode string)
New heading to add for the added field.
extra_field: (a unicode string)
New field to add to each row. This is useful for saving the group or
subject ID in the file.
out_file: (a file name, nipype default value: merged.csv)
Output filename for merged CSV file
row_heading_title: (a unicode string, nipype default value: label)
Column heading for the row headings added
row_headings: (a list of items which are a unicode string)
List of row headings to save in merged CSV file (must be equal to
number of rows in the input files).
Outputs:
csv_file: (a file name)
Output CSV file containing columns
MergeROIs¶
Splits a 3D image in small chunks to enable parallel processing. ROIs keep time series structure in 4D images.
Example¶
>>> from nipype.algorithms import misc
>>> rois = misc.MergeROIs()
>>> rois.inputs.in_files = ['roi%02d.nii' % i for i in range(1, 6)]
>>> rois.inputs.in_reference = 'mask.nii'
>>> rois.inputs.in_index = ['roi%02d_idx.npz' % i for i in range(1, 6)]
>>> rois.run()
Inputs:
[Mandatory]
[Optional]
in_files: (a list of items which are an existing file name)
in_index: (a list of items which are an existing file name)
array keeping original locations
in_reference: (an existing file name)
reference file
Outputs:
merged_file: (an existing file name)
the recomposed file
ModifyAffine¶
Left multiplies the affine matrix with a specified values. Saves the volume as a nifti file.
Inputs:
[Mandatory]
volumes: (a list of items which are an existing file name)
volumes which affine matrices will be modified
[Optional]
ignore_exception: (a boolean, nipype default value: False)
Print an error message instead of throwing an exception in case the
interface fails to run
transformation_matrix: (an array with shape (4, 4), nipype default
value: (<bound method Array.copy_default_value of
<traits.trait_numeric.Array object at 0x7f9e85beb1d0>>, (array([[
1., 0., 0., 0.], [ 0., 1., 0., 0.], [ 0., 0.,
1., 0.], [ 0., 0., 0., 1.]]),), None))
transformation matrix that will be left multiplied by the affine
matrix
Outputs:
transformed_volumes: (a list of items which are a file name)
NormalizeProbabilityMapSet¶
Returns the input tissue probability maps (tpms, aka volume fractions) normalized to sum up 1.0 at each voxel within the mask.
Note
Please recall this is not a spatial normalization algorithm
Example¶
>>> from nipype.algorithms import misc
>>> normalize = misc.NormalizeProbabilityMapSet()
>>> normalize.inputs.in_files = [ 'tpm_00.nii.gz', 'tpm_01.nii.gz', 'tpm_02.nii.gz' ]
>>> normalize.inputs.in_mask = 'tpms_msk.nii.gz'
>>> normalize.run()
Inputs:
[Mandatory]
[Optional]
in_files: (a list of items which are an existing file name)
in_mask: (an existing file name)
Masked voxels must sum up 1.0, 0.0 otherwise.
Outputs:
out_files: (a list of items which are an existing file name)
normalized maps
Overlap¶
Calculates various overlap measures between two maps.
Deprecated since version 0.10.0: Use nipype.algorithms.metrics.Overlap instead.
Inputs:
[Mandatory]
bg_overlap: (a boolean, nipype default value: False)
consider zeros as a label
vol_units: (u'voxel' or u'mm', nipype default value: voxel)
units for volumes
volume1: (an existing file name)
Has to have the same dimensions as volume2.
volume2: (an existing file name)
Has to have the same dimensions as volume1.
[Optional]
ignore_exception: (a boolean, nipype default value: False)
Print an error message instead of throwing an exception in case the
interface fails to run
mask_volume: (an existing file name)
calculate overlap only within this mask.
out_file: (a file name, nipype default value: diff.nii)
weighting: (u'none' or u'volume' or u'squared_vol', nipype default
value: none)
'none': no class-overlap weighting is performed. 'volume': computed
class-overlaps are weighted by class volume 'squared_vol': computed
class-overlaps are weighted by the squared volume of the class
Outputs:
dice: (a float)
averaged dice index
diff_file: (an existing file name)
error map of differences
jaccard: (a float)
averaged jaccard index
labels: (a list of items which are an integer (int or long))
detected labels
roi_di: (a list of items which are a float)
the Dice index (DI) per ROI
roi_ji: (a list of items which are a float)
the Jaccard index (JI) per ROI
roi_voldiff: (a list of items which are a float)
volume differences of ROIs
volume_difference: (a float)
averaged volume difference
PickAtlas¶
Returns ROI masks given an atlas and a list of labels. Supports dilation and left right masking (assuming the atlas is properly aligned).
Inputs:
[Mandatory]
atlas: (an existing file name)
Location of the atlas that will be used.
labels: (an integer (int or long) or a list of items which are an
integer (int or long))
Labels of regions that will be included in the mask. Must be
compatible with the atlas used.
[Optional]
dilation_size: (an integer (int or long), nipype default value: 0)
Defines how much the mask will be dilated (expanded in 3D).
hemi: (u'both' or u'left' or u'right', nipype default value: both)
Restrict the mask to only one hemisphere: left or right
ignore_exception: (a boolean, nipype default value: False)
Print an error message instead of throwing an exception in case the
interface fails to run
output_file: (a file name)
Where to store the output mask.
Outputs:
mask_file: (an existing file name)
output mask file
SimpleThreshold¶
Applies a threshold to input volumes
Inputs:
[Mandatory]
threshold: (a float)
volumes to be thresholdedeverything below this value will be set to
zero
volumes: (a list of items which are an existing file name)
volumes to be thresholded
[Optional]
ignore_exception: (a boolean, nipype default value: False)
Print an error message instead of throwing an exception in case the
interface fails to run
Outputs:
thresholded_volumes: (a list of items which are an existing file
name)
thresholded volumes
SplitROIs¶
Splits a 3D image in small chunks to enable parallel processing. ROIs keep time series structure in 4D images.
Example¶
>>> from nipype.algorithms import misc
>>> rois = misc.SplitROIs()
>>> rois.inputs.in_file = 'diffusion.nii'
>>> rois.inputs.in_mask = 'mask.nii'
>>> rois.run()
Inputs:
[Mandatory]
in_file: (an existing file name)
file to be splitted
[Optional]
in_mask: (an existing file name)
only process files inside mask
roi_size: (a tuple of the form: (an integer (int or long), an integer
(int or long), an integer (int or long)))
desired ROI size
Outputs:
out_files: (a list of items which are an existing file name)
the resulting ROIs
out_index: (a list of items which are an existing file name)
arrays keeping original locations
out_masks: (a list of items which are an existing file name)
a mask indicating valid values
TSNR¶
Deprecated since version 0.12.1: Use nipype.algorithms.confounds.TSNR instead
Inputs:
[Mandatory]
in_file: (a list of items which are an existing file name)
realigned 4D file or a list of 3D files
[Optional]
detrended_file: (a file name, nipype default value: detrend.nii.gz)
input file after detrending
ignore_exception: (a boolean, nipype default value: False)
Print an error message instead of throwing an exception in case the
interface fails to run
mean_file: (a file name, nipype default value: mean.nii.gz)
output mean file
regress_poly: (an integer >= 1)
Remove polynomials
stddev_file: (a file name, nipype default value: stdev.nii.gz)
output tSNR file
tsnr_file: (a file name, nipype default value: tsnr.nii.gz)
output tSNR file
Outputs:
detrended_file: (a file name)
detrended input file
mean_file: (an existing file name)
mean image file
stddev_file: (an existing file name)
std dev image file
tsnr_file: (an existing file name)
tsnr image file
calc_moments()¶
Returns nth moment (3 for skewness, 4 for kurtosis) of timeseries (list of values; one per timeseries).
Keyword arguments: timeseries_file – text file with white space separated timepoints in rows
makefmtlist()¶
maketypelist()¶
matlab2csv()¶
merge_csvs()¶
normalize_tpms()¶
Returns the input tissue probability maps (tpms, aka volume fractions) normalized to sum up 1.0 at each voxel within the mask.