nipype.interfaces.camino.convert module

AnalyzeHeader

Link to code

Bases: StdOutCommandLine

Wrapped executable: analyzeheader.

Create or read an Analyze 7.5 header file.

Analyze image header, provides support for the most common header fields. Some fields, such as patient_id, are not currently supported. The program allows three nonstandard options: the field image_dimension.funused1 is the image scale. The intensity of each pixel in the associated .img file is (image value from file) * scale. Also, the origin of the Talairach coordinates (midline of the anterior commisure) are encoded in the field data_history.originator. These changes are included for compatibility with SPM.

All headers written with this program are big endian by default.

Example

>>> import nipype.interfaces.camino as cmon
>>> hdr = cmon.AnalyzeHeader()
>>> hdr.inputs.in_file = 'tensor_fitted_data.Bdouble'
>>> hdr.inputs.scheme_file = 'A.scheme'
>>> hdr.inputs.data_dims = [256,256,256]
>>> hdr.inputs.voxel_dims = [1,1,1]
>>> hdr.run()                  
Mandatory Inputs:
  • datatype (‘byte’ or ‘char’ or ‘[u]short’ or ‘[u]int’ or ‘float’ or ‘complex’ or ‘double’) – The char datatype is 8 bit (not the 16 bit char of Java), as specified by the Analyze 7.5 standard. The byte, ushort and uint types are not part of the Analyze specification but are supported by SPM. Maps to a command-line argument: -datatype %s.

  • in_file (a pathlike object or string representing an existing file) – Tensor-fitted data filename. Maps to a command-line argument: < %s (position: 1).

Optional Inputs:
  • args (a string) – Additional parameters to the command. Maps to a command-line argument: %s.

  • centre (a list of from 3 to 3 items which are an integer) – Voxel specifying origin of Talairach coordinate system for SPM, default [0 0 0]. Maps to a command-line argument: -centre %s.

  • data_dims (a list of from 3 to 3 items which are an integer) – Data dimensions in voxels. Maps to a command-line argument: -datadims %s.

  • description (a string) – Short description - No spaces, max length 79 bytes. Will be null terminated automatically. Maps to a command-line argument: -description %s.

  • environ (a dictionary with keys which are a bytes or None or a value of class ‘str’ and with values which are a bytes or None or a value of class ‘str’) – Environment variables. (Nipype default value: {})

  • greylevels (a list of from 2 to 2 items which are an integer) – Minimum and maximum greylevels. Stored as shorts in the header. Maps to a command-line argument: -gl %s.

  • initfromheader (a pathlike object or string representing an existing file) – Reads header information from file and initializes a new header with the values read from the file. You may replace any combination of fields in the new header by specifying subsequent options. Maps to a command-line argument: -initfromheader %s (position: 3).

  • intelbyteorder (a boolean) – Write header in intel byte order (little-endian). Maps to a command-line argument: -intelbyteorder.

  • networkbyteorder (a boolean) – Write header in network byte order (big-endian). This is the default for new headers. Maps to a command-line argument: -networkbyteorder.

  • nimages (an integer) – Number of images in the img file. Default 1. Maps to a command-line argument: -nimages %d.

  • offset (an integer) – According to the Analyze 7.5 standard, this is the byte offset in the .img file at which voxels start. This value can be negative to specify that the absolute value is applied for every image in the file. Maps to a command-line argument: -offset %d.

  • out_file (a pathlike object or string representing a file) – Maps to a command-line argument: > %s (position: -1).

  • picoseed (a list of from 3 to 3 items which are an integer) – Voxel specifying the seed (for PICo maps), default [0 0 0]. Maps to a command-line argument: -picoseed %s.

  • printbigendian (a pathlike object or string representing an existing file) – Prints 1 if the header is big-endian, 0 otherwise. Maps to a command-line argument: -printbigendian %s (position: 3).

  • printimagedims (a pathlike object or string representing an existing file) – Prints image data and voxel dimensions as Camino arguments and exits. Maps to a command-line argument: -printimagedims %s (position: 3).

  • printintelbyteorder (a pathlike object or string representing an existing file) – Prints 1 if the header is little-endian, 0 otherwise. Maps to a command-line argument: -printintelbyteorder %s (position: 3).

  • printprogargs (a pathlike object or string representing an existing file) – Prints data dimension (and type, if relevant) arguments for a specific Camino program, where prog is one of shredder, scanner2voxel, vcthreshselect, pdview, track. Maps to a command-line argument: -printprogargs %s (position: 3).

  • readheader (a pathlike object or string representing an existing file) – Reads header information from file and prints to stdout. If this option is not specified, then the program writes a header based on the other arguments. Maps to a command-line argument: -readheader %s (position: 3).

  • scaleinter (a float) – Constant to add to the image intensities. Used by SPM and MRIcro. Maps to a command-line argument: -scaleinter %d.

  • scaleslope (a float) – Intensities in the image are scaled by this factor by SPM and MRICro. Default is 1.0. Maps to a command-line argument: -scaleslope %d.

  • scheme_file (a pathlike object or string representing an existing file) – Camino scheme file (b values / vectors, see camino.fsl2scheme). Maps to a command-line argument: %s (position: 2).

  • voxel_dims (a list of from 3 to 3 items which are a float) – Voxel dimensions in mm. Maps to a command-line argument: -voxeldims %s.

Outputs:

header (a pathlike object or string representing an existing file) – Analyze header.

DT2NIfTI

Link to code

Bases: CommandLine

Wrapped executable: dt2nii.

Converts camino tensor data to NIfTI format

Reads Camino diffusion tensors, and converts them to NIFTI format as three .nii files.

Mandatory Inputs:
  • header_file (a pathlike object or string representing an existing file) – A Nifti .nii or .hdr file containing the header information. Maps to a command-line argument: -header %s (position: 3).

  • in_file (a pathlike object or string representing an existing file) – Tract file. Maps to a command-line argument: -inputfile %s (position: 1).

Optional Inputs:
  • args (a string) – Additional parameters to the command. Maps to a command-line argument: %s.

  • environ (a dictionary with keys which are a bytes or None or a value of class ‘str’ and with values which are a bytes or None or a value of class ‘str’) – Environment variables. (Nipype default value: {})

  • output_root (a pathlike object or string representing a file) – Filename root prepended onto the names of three output files. Maps to a command-line argument: -outputroot %s (position: 2).

Outputs:
  • dt (a pathlike object or string representing an existing file) – Diffusion tensors in NIfTI format.

  • exitcode (a pathlike object or string representing an existing file) – Exit codes from Camino reconstruction in NIfTI format.

  • lns0 (a pathlike object or string representing an existing file) – Estimated lns0 from Camino reconstruction in NIfTI format.

FSL2Scheme

Link to code

Bases: StdOutCommandLine

Wrapped executable: fsl2scheme.

Converts b-vectors and b-values from FSL format to a Camino scheme file.

Examples

>>> import nipype.interfaces.camino as cmon
>>> makescheme = cmon.FSL2Scheme()
>>> makescheme.inputs.bvec_file = 'bvecs'
>>> makescheme.inputs.bvec_file = 'bvals'
>>> makescheme.run()                  
Mandatory Inputs:
  • bval_file (a pathlike object or string representing an existing file) – B value file. Maps to a command-line argument: -bvalfile %s (position: 2).

  • bvec_file (a pathlike object or string representing an existing file) – B vector file. Maps to a command-line argument: -bvecfile %s (position: 1).

Optional Inputs:
  • args (a string) – Additional parameters to the command. Maps to a command-line argument: %s.

  • bscale (a float) – Scaling factor to convert the b-values into different units. Default is 10^6. Maps to a command-line argument: -bscale %d.

  • diffusiontime (a float) – Diffusion time. Maps to a command-line argument: -diffusiontime %f.

  • environ (a dictionary with keys which are a bytes or None or a value of class ‘str’ and with values which are a bytes or None or a value of class ‘str’) – Environment variables. (Nipype default value: {})

  • flipx (a boolean) – Negate the x component of all the vectors. Maps to a command-line argument: -flipx.

  • flipy (a boolean) – Negate the y component of all the vectors. Maps to a command-line argument: -flipy.

  • flipz (a boolean) – Negate the z component of all the vectors. Maps to a command-line argument: -flipz.

  • interleave (a boolean) – Interleave repeated scans. Only used with -numscans. Maps to a command-line argument: -interleave.

  • numscans (an integer) – Output all measurements numerous (n) times, used when combining multiple scans from the same imaging session. Maps to a command-line argument: -numscans %d.

  • out_file (a pathlike object or string representing a file) – Maps to a command-line argument: > %s (position: -1).

  • usegradmod (a boolean) – Use the gradient magnitude to scale b. This option has no effect if your gradient directions have unit magnitude. Maps to a command-line argument: -usegradmod.

Outputs:

scheme (a pathlike object or string representing an existing file) – Scheme file.

Image2Voxel

Link to code

Bases: StdOutCommandLine

Wrapped executable: image2voxel.

Converts Analyze / NIFTI / MHA files to voxel order.

Converts scanner-order data in a supported image format to voxel-order data. Either takes a 4D file (all measurements in single image) or a list of 3D images.

Examples

>>> import nipype.interfaces.camino as cmon
>>> img2vox = cmon.Image2Voxel()
>>> img2vox.inputs.in_file = '4d_dwi.nii'
>>> img2vox.run()                  
Mandatory Inputs:

in_file (a pathlike object or string representing an existing file) – 4d image file. Maps to a command-line argument: -4dimage %s (position: 1).

Optional Inputs:
  • args (a string) – Additional parameters to the command. Maps to a command-line argument: %s.

  • environ (a dictionary with keys which are a bytes or None or a value of class ‘str’ and with values which are a bytes or None or a value of class ‘str’) – Environment variables. (Nipype default value: {})

  • out_file (a pathlike object or string representing a file) – Maps to a command-line argument: > %s (position: -1).

  • out_type (‘float’ or ‘char’ or ‘short’ or ‘int’ or ‘long’ or ‘double’) – “i.e. Bfloat”. Can be “char”, “short”, “int”, “long”, “float” or “double”. Maps to a command-line argument: -outputdatatype %s (position: 2). (Nipype default value: float)

Outputs:

voxel_order (a pathlike object or string representing an existing file) – Path/name of 4D volume in voxel order.

NIfTIDT2Camino

Link to code

Bases: CommandLine

Wrapped executable: niftidt2camino.

Converts NIFTI-1 diffusion tensors to Camino format. The program reads the NIFTI header but does not apply any spatial transformations to the data. The NIFTI intensity scaling parameters are applied.

The output is the tensors in Camino voxel ordering: [exit, ln(S0), dxx, dxy, dxz, dyy, dyz, dzz].

The exit code is set to 0 unless a background mask is supplied, in which case the code is 0 in brain voxels and -1 in background voxels.

The value of ln(S0) in the output is taken from a file if one is supplied, otherwise it is set to 0.

NOTE FOR FSL USERS - FSL’s dtifit can output NIFTI tensors, but they are not stored in the usual way (which is using NIFTI_INTENT_SYMMATRIX). FSL’s tensors follow the ITK / VTK “upper-triangular” convention, so you will need to use the -uppertriangular option to convert these correctly.

Mandatory Inputs:

in_file (a pathlike object or string representing an existing file) – A NIFTI-1 dataset containing diffusion tensors. The tensors are assumed to be in lower-triangular order as specified by the NIFTI standard for the storage of symmetric matrices. This file should be either a .nii or a .hdr file. Maps to a command-line argument: -inputfile %s (position: 1).

Optional Inputs:
  • args (a string) – Additional parameters to the command. Maps to a command-line argument: %s.

  • bgmask (a pathlike object or string representing an existing file) – Binary valued brain / background segmentation, may be a raw binary file (specify type with -maskdatatype) or a supported image file. Maps to a command-line argument: -bgmask %s.

  • environ (a dictionary with keys which are a bytes or None or a value of class ‘str’ and with values which are a bytes or None or a value of class ‘str’) – Environment variables. (Nipype default value: {})

  • lns0_file (a pathlike object or string representing an existing file) – File containing the log of the unweighted signal for each voxel, may be a raw binary file (specify type with -inputdatatype) or a supported image file. Maps to a command-line argument: -lns0 %s.

  • out_file (a pathlike object or string representing a file) – Maps to a command-line argument: > %s (position: -1).

  • s0_file (a pathlike object or string representing an existing file) – File containing the unweighted signal for each voxel, may be a raw binary file (specify type with -inputdatatype) or a supported image file. Maps to a command-line argument: -s0 %s.

  • scaleinter (a float) – A value v in the diffusion tensor is scaled to v * s + i. This is applied after any scaling specified by the input image. Default is 0.0. Maps to a command-line argument: -scaleinter %s.

  • scaleslope (a float) – A value v in the diffusion tensor is scaled to v * s + i. This is applied after any scaling specified by the input image. Default is 1.0. Maps to a command-line argument: -scaleslope %s.

  • uppertriangular (a boolean) – Specifies input in upper-triangular (VTK style) order. Maps to a command-line argument: -uppertriangular %s.

Outputs:

out_file (a pathlike object or string representing a file) – Diffusion tensors data in Camino format.

ProcStreamlines

Link to code

Bases: StdOutCommandLine

Wrapped executable: procstreamlines.

Process streamline data

This program does post-processing of streamline output from track. It can either output streamlines or connection probability maps.

Examples

>>> import nipype.interfaces.camino as cmon
>>> proc = cmon.ProcStreamlines()
>>> proc.inputs.in_file = 'tract_data.Bfloat'
>>> proc.run()                  
Mandatory Inputs:

in_file (a pathlike object or string representing an existing file) – Data file. Maps to a command-line argument: -inputfile %s (position: 1).

Optional Inputs:
  • allowmultitargets (a boolean) – Allows streamlines to connect to multiple target volumes. Maps to a command-line argument: -allowmultitargets.

  • args (a string) – Additional parameters to the command. Maps to a command-line argument: %s.

  • datadims (a list of from 3 to 3 items which are an integer) – Data dimensions in voxels. Maps to a command-line argument: -datadims %s.

  • directional (a list of from 3 to 3 items which are an integer) – Splits the streamlines at the seed point and computes separate connection probabilities for each segment. Streamline segments are grouped according to their dot product with the vector (X, Y, Z). The ideal vector will be tangential to the streamline trajectory at the seed, such that the streamline projects from the seed along (X, Y, Z) and -(X, Y, Z). However, it is only necessary for the streamline trajectory to not be orthogonal to (X, Y, Z). Maps to a command-line argument: -directional %s.

  • discardloops (a boolean) – This option allows streamlines to enter a waypoint exactly once. After the streamline leaves the waypoint, the entire streamline is discarded upon a second entry to the waypoint. Maps to a command-line argument: -discardloops.

  • endpointfile (a pathlike object or string representing a file) – Image containing endpoint ROIs. This should be an Analyze 7.5 header / image file.hdr and file.img. Maps to a command-line argument: -endpointfile %s.

  • environ (a dictionary with keys which are a bytes or None or a value of class ‘str’ and with values which are a bytes or None or a value of class ‘str’) – Environment variables. (Nipype default value: {})

  • exclusionfile (a pathlike object or string representing a file) – Image containing exclusion ROIs. This should be an Analyze 7.5 header / image file.hdr and file.img. Maps to a command-line argument: -exclusionfile %s.

  • gzip (a boolean) – Save the output image in gzip format. Maps to a command-line argument: -gzip.

  • inputmodel (‘raw’ or ‘voxels’) – Input model type (raw or voxels). Maps to a command-line argument: -inputmodel %s. (Nipype default value: raw)

  • iterations (a float) – Number of streamlines generated for each seed. Not required when outputting streamlines, but needed to create PICo images. The default is 1 if the output is streamlines, and 5000 if the output is connection probability images. Maps to a command-line argument: -iterations %d.

  • maxtractlength (an integer) – Maximum length of tracts. Maps to a command-line argument: -maxtractlength %d.

  • maxtractpoints (an integer) – Maximum number of tract points. Maps to a command-line argument: -maxtractpoints %d.

  • mintractlength (an integer) – Minimum length of tracts. Maps to a command-line argument: -mintractlength %d.

  • mintractpoints (an integer) – Minimum number of tract points. Maps to a command-line argument: -mintractpoints %d.

  • noresample (a boolean) – Disables resampling of input streamlines. Resampling is automatically disabled if the input model is voxels. Maps to a command-line argument: -noresample.

  • out_file (a pathlike object or string representing a file) – Maps to a command-line argument: > %s (position: -1).

  • outputacm (a boolean) – Output all tracts in a single connection probability map (Analyze image). Maps to a command-line argument: -outputacm. Requires inputs: outputroot, seedfile.

  • outputcbs (a boolean) – Outputs connectivity-based segmentation maps; requires target outputfile. Maps to a command-line argument: -outputcbs. Requires inputs: outputroot, targetfile, seedfile.

  • outputcp (a boolean) – Output the connection probability map (Analyze image, float). Maps to a command-line argument: -outputcp. Requires inputs: outputroot, seedfile.

  • outputroot (a pathlike object or string representing a file) – Prepended onto all output file names. Maps to a command-line argument: -outputroot %s.

  • outputsc (a boolean) – Output the connection probability map (raw streamlines, int). Maps to a command-line argument: -outputsc. Requires inputs: outputroot, seedfile.

  • outputtracts (a boolean) – Output streamlines in raw binary format. Maps to a command-line argument: -outputtracts.

  • regionindex (an integer) – Index of specific region to process. Maps to a command-line argument: -regionindex %d.

  • resamplestepsize (a float) – Each point on a streamline is tested for entry into target, exclusion or waypoint volumes. If the length between points on a tract is not much smaller than the voxel length, then streamlines may pass through part of a voxel without being counted. To avoid this, the program resamples streamlines such that the step size is one tenth of the smallest voxel dimension in the image. This increases the size of raw or oogl streamline output and incurs some performance penalty. The resample resolution can be controlled with this option or disabled altogether by passing a negative step size or by passing the -noresample option. Maps to a command-line argument: -resamplestepsize %d.

  • seedfile (a pathlike object or string representing a file) – Image Containing Seed Points. Maps to a command-line argument: -seedfile %s.

  • seedpointmm (a list of from 3 to 3 items which are an integer) – The coordinates of a single seed point for tractography in mm. Maps to a command-line argument: -seedpointmm %s.

  • seedpointvox (a list of from 3 to 3 items which are an integer) – The coordinates of a single seed point for tractography in voxels. Maps to a command-line argument: -seedpointvox %s.

  • targetfile (a pathlike object or string representing a file) – Image containing target volumes. Maps to a command-line argument: -targetfile %s.

  • truncateinexclusion (a boolean) – Retain segments of a streamline before entry to an exclusion ROI. Maps to a command-line argument: -truncateinexclusion.

  • truncateloops (a boolean) – This option allows streamlines to enter a waypoint exactly once. After the streamline leaves the waypoint, it is truncated upon a second entry to the waypoint. Maps to a command-line argument: -truncateloops.

  • voxeldims (a list of from 3 to 3 items which are an integer) – Voxel dimensions in mm. Maps to a command-line argument: -voxeldims %s.

  • waypointfile (a pathlike object or string representing a file) – Image containing waypoints. Waypoints are defined as regions of the image with the same intensity, where 0 is background and any value > 0 is a waypoint. Maps to a command-line argument: -waypointfile %s.

Outputs:
  • outputroot_files (a list of items which are a pathlike object or string representing an existing file)

  • proc (a pathlike object or string representing an existing file) – Processed Streamlines.

Shredder

Link to code

Bases: StdOutCommandLine

Wrapped executable: shredder.

Extracts periodic chunks from a data stream.

Shredder makes an initial offset of offset bytes. It then reads and outputs chunksize bytes, skips space bytes, and repeats until there is no more input.

If the chunksize is negative, chunks of size chunksize are read and the byte ordering of each chunk is reversed. The whole chunk will be reversed, so the chunk must be the same size as the data type, otherwise the order of the values in the chunk, as well as their endianness, will be reversed.

Examples

>>> import nipype.interfaces.camino as cam
>>> shred = cam.Shredder()
>>> shred.inputs.in_file = 'SubjectA.Bfloat'
>>> shred.inputs.offset = 0
>>> shred.inputs.chunksize = 1
>>> shred.inputs.space = 2
>>> shred.run()                  
Mandatory Inputs:

in_file (a pathlike object or string representing an existing file) – Raw binary data file. Maps to a command-line argument: < %s (position: -2).

Optional Inputs:
  • args (a string) – Additional parameters to the command. Maps to a command-line argument: %s.

  • chunksize (an integer) – Reads and outputs a chunk of chunksize bytes. Maps to a command-line argument: %d (position: 2).

  • environ (a dictionary with keys which are a bytes or None or a value of class ‘str’ and with values which are a bytes or None or a value of class ‘str’) – Environment variables. (Nipype default value: {})

  • offset (an integer) – Initial offset of offset bytes. Maps to a command-line argument: %d (position: 1).

  • out_file (a pathlike object or string representing a file) – Maps to a command-line argument: > %s (position: -1).

  • space (an integer) – Skips space bytes. Maps to a command-line argument: %d (position: 3).

Outputs:

shredded (a pathlike object or string representing an existing file) – Shredded binary data file.

TractShredder

Link to code

Bases: StdOutCommandLine

Wrapped executable: tractshredder.

Extracts bunches of streamlines.

tractshredder works in a similar way to shredder, but processes streamlines instead of scalar data. The input is raw streamlines, in the format produced by track or procstreamlines.

The program first makes an initial offset of offset tracts. It then reads and outputs a group of bunchsize tracts, skips space tracts, and repeats until there is no more input.

Examples

>>> import nipype.interfaces.camino as cmon
>>> shred = cmon.TractShredder()
>>> shred.inputs.in_file = 'tract_data.Bfloat'
>>> shred.inputs.offset = 0
>>> shred.inputs.bunchsize = 1
>>> shred.inputs.space = 2
>>> shred.run()                  
Mandatory Inputs:

in_file (a pathlike object or string representing an existing file) – Tract file. Maps to a command-line argument: < %s (position: -2).

Optional Inputs:
  • args (a string) – Additional parameters to the command. Maps to a command-line argument: %s.

  • bunchsize (an integer) – Reads and outputs a group of bunchsize tracts. Maps to a command-line argument: %d (position: 2).

  • environ (a dictionary with keys which are a bytes or None or a value of class ‘str’ and with values which are a bytes or None or a value of class ‘str’) – Environment variables. (Nipype default value: {})

  • offset (an integer) – Initial offset of offset tracts. Maps to a command-line argument: %d (position: 1).

  • out_file (a pathlike object or string representing a file) – Maps to a command-line argument: > %s (position: -1).

  • space (an integer) – Skips space tracts. Maps to a command-line argument: %d (position: 3).

Outputs:

shredded (a pathlike object or string representing an existing file) – Shredded tract file.

VtkStreamlines

Link to code

Bases: StdOutCommandLine

Wrapped executable: vtkstreamlines.

Use vtkstreamlines to convert raw or voxel format streamlines to VTK polydata

Examples

>>> import nipype.interfaces.camino as cmon
>>> vtk = cmon.VtkStreamlines()
>>> vtk.inputs.in_file = 'tract_data.Bfloat'
>>> vtk.inputs.voxeldims = [1,1,1]
>>> vtk.run()                  
Mandatory Inputs:

in_file (a pathlike object or string representing an existing file) – Data file. Maps to a command-line argument: < %s (position: -2).

Optional Inputs:
  • args (a string) – Additional parameters to the command. Maps to a command-line argument: %s.

  • colourorient (a boolean) – Each point on the streamline is coloured by the local orientation. Maps to a command-line argument: -colourorient.

  • environ (a dictionary with keys which are a bytes or None or a value of class ‘str’ and with values which are a bytes or None or a value of class ‘str’) – Environment variables. (Nipype default value: {})

  • inputmodel (‘raw’ or ‘voxels’) – Input model type (raw or voxels). Maps to a command-line argument: -inputmodel %s. (Nipype default value: raw)

  • interpolate (a boolean) – The scalar value at each point on the streamline is calculated by trilinear interpolation. Maps to a command-line argument: -interpolate.

  • interpolatescalars (a boolean) – The scalar value at each point on the streamline is calculated by trilinear interpolation. Maps to a command-line argument: -interpolatescalars.

  • out_file (a pathlike object or string representing a file) – Maps to a command-line argument: > %s (position: -1).

  • scalar_file (a pathlike object or string representing a file) – Image that is in the same physical space as the tracts. Maps to a command-line argument: -scalarfile %s (position: 3).

  • seed_file (a pathlike object or string representing a file) – Image containing seed points. Maps to a command-line argument: -seedfile %s (position: 1).

  • target_file (a pathlike object or string representing a file) – Image containing integer-valued target regions. Maps to a command-line argument: -targetfile %s (position: 2).

  • voxeldims (a list of from 3 to 3 items which are an integer) – Voxel dimensions in mm. Maps to a command-line argument: -voxeldims %s (position: 4).

Outputs:

vtk (a pathlike object or string representing an existing file) – Streamlines in VTK format.