3 d_@sPdZddlZddljZddlmZmZmZm Z ddl m Z m Z m Z mZmZmZmZmZmZmZddl mZmZmZmZmZmZmZmZdd lmZejd Z Gd d d eZ!Gd ddeZ"Gddde Z#GdddeZ$GdddeZ%Gddde Z&GdddeZ'GdddeZ(GdddeZ)GdddeZ*Gdd d e Z+Gd!d"d"eZ,Gd#d$d$e Z-Gd%d&d&eZ.Gd'd(d(eZ/Gd)d*d*eZ0Gd+d,d,eZ1Gd-d.d.eZ2Gd/d0d0eZ3Gd1d2d2eZ4Gd3d4d4e Z5Gd5d6d6e Z6Gd7d8d8eZ7Gd9d:d:e!Z8Gd;d<dd>eZ:Gd?d@d@eZ;GdAdBdBeZGdGdHdHe!Z?GdIdJdJeZ@GdKdLdLeZAGdMdNdNeZBGdOdPdPeZCGdQdRdReZDGdSdTdTe ZEGdUdVdVe ZFGdWdXdXeZGGdYdZdZe!ZHGd[d\d\eZIGd]d^d^eZJGd_d`d`eZKGdadbdbeZLGdcddddeZMGdedfdfe ZNGdgdhdhe ZOGdidjdje ZPGdkdldle ZQGdmdndne ZRGdodpdpe ZSGdqdrdre ZTGdsdtdte ZUGdudvdveZVGdwdxdxeZWGdydzdzeZXGd{d|d|e ZYGd}d~d~eZZGdddeZ[GdddeZ\GdddeZ]Gddde Z^GdddeZ_GdddeZ`Gddde ZaGdddeZbGdddeZcGddde ZdGdddeZeGdddeZfGdddeZgGdddeZhGdddeZiGdddeZjGdddeZkGdddeZlGdddeZmGdddeZnGdddeZoGdddeZpGdddeZqGddde ZrGdddeZsGdddeZtGddde ZuGdddeZvGdddeZwGddde ZxGdddeZyGdddewZzGdddeyZ{dS)zAFNI preprocessing interfaces.N) load_json save_jsonsplit_filenamefname_presuffix) CommandLineInputSpec CommandLine TraitedSpectraits isdefinedFileInputMultiPath UndefinedStrInputMultiObject)AFNICommandBase AFNICommandAFNICommandInputSpecAFNICommandOutputSpecAFNIPythonCommandInputSpecAFNIPythonCommandInfono_afni)loggingznipype.interfacec@sVeZdZdZeddddZejdddZej d d dZ ej d d dZ ej d ddZ dS)CentralityInputSpecz;Common input spec class for all centrality-related commandszmask file to mask input dataz-mask %sT)descargstrexistsz5threshold to exclude connections where corr <= threshz -thresh %f)rrz -polort %dz-Clip off low-intensity regions in the datasetz -autoclipz,Mask the dataset to target brain-only voxelsz -automaskN)__name__ __module__ __qualname____doc__r maskr FloatZthreshIntpolortBoolautoclipautomaskr,r,C/tmp/pip-build-7vycvbft/nipype/nipype/interfaces/afni/preprocess.pyr's rc@seZdZeddddddZeddddddZejejdd ej d d d d dddZ ej dddZ ej dddZ ej dddZejdddddZej d(dddZej d d!dd"d#d$Zej d d!dd%d&d$Zd'S))AlignEpiAnatPyInputSpeczEPI dataset to alignz-epi %sTF)rr mandatoryrcopyfilezname of structural datasetz-anat %sr)lowmeanmedianmaxzLthe epi base used in alignmentshould be one of (0/mean/median/max/subbrick#)z -epi_base %s)rr/rz)align anatomical to EPI dataset (default)z -anat2epi)rrzalign EPI to anatomical datasetz -epi2anatz!save skull-stripped (not aligned)z-save_skullstripZ_alzfappend suffix to the original anat/epi dataset to usein the resulting dataset names (default is "_al")z -suffix %s)r usedefaultr 3dSkullStrip 3dAutomaskNonezOmethod to mask brain in EPI datashould be one of[3dSkullStrip]/3dAutomask/None)z -epi_strip %sonoffz_do volume registration on EPI dataset before alignmentshould be 'on' or 'off', defaults to 'on'z -volreg %s)r5rrzYdo time shifting of EPI dataset before alignmentshould be 'on' or 'off', defaults to 'on'z -tshift %sN)r6r7r8)r!r"r#r in_fileanatr EitherRangeEnumZepi_baser)anat2epiepi2anatsave_skullstriprsuffixZ epi_stripvolregtshiftr,r,r,r-r.8sX    r.c@speZdZeddZeddZeddZeddZeddZeddZ eddZ ed dZ ed dZ ed dZ d S) AlignEpiAnatPyOutputSpecz3A version of the anatomy that is aligned to the EPI)rz3A version of the EPI dataset aligned to the anatomyz;A version of the EPI dataset aligned to a standard templatez"matrix to align anatomy to the EPIzmatrix to align EPI to anatomyzmatrix to volume register EPIz2matrix to volume register and align epi to anatomyzMmatrix to volume register and align epito anatomy and put into standard spacezumotion parameters from EPI time-seriesregistration (tsh included in name if slicetiming correction is also included).z#skull-stripped (not aligned) volumeN)r!r"r#r anat_al_orig epi_al_origZ epi_tlrc_al anat_al_mat epi_al_mat epi_vr_al_matepi_reg_al_matZepi_al_tlrc_mat epi_vr_motion skullstripr,r,r,r-rFus      rFc@s$eZdZdZdZeZeZddZ dS)AlignEpiAnatPya Align EPI to anatomical datasets or vice versa. This Python script computes the alignment between two datasets, typically an EPI and an anatomical structural dataset, and applies the resulting transformation to one or the other to bring them into alignment. This script computes the transforms needed to align EPI and anatomical datasets using a cost function designed for this purpose. The script combines multiple transformations, thereby minimizing the amount of interpolation applied to the data. Basic Usage:: align_epi_anat.py -anat anat+orig -epi epi+orig -epi_base 5 The user must provide :abbr:`EPI (echo-planar imaging)` and anatomical datasets and specify the EPI sub-brick to use as a base in the alignment. Internally, the script always aligns the anatomical to the EPI dataset, and the resulting transformation is saved to a 1D file. As a user option, the inverse of this transformation may be applied to the EPI dataset in order to align it to the anatomical data instead. This program generates several kinds of output in the form of datasets and transformation matrices which can be applied to other datasets if needed. Time-series volume registration, oblique data transformations and Talairach (standard template) transformations will be combined as needed and requested (with options to turn on and off each of the steps) in order to create the aligned datasets. Examples -------- >>> from nipype.interfaces import afni >>> al_ea = afni.AlignEpiAnatPy() >>> al_ea.inputs.anat = "structural.nii" >>> al_ea.inputs.in_file = "functional.nii" >>> al_ea.inputs.epi_base = 0 >>> al_ea.inputs.epi_strip = '3dAutomask' >>> al_ea.inputs.volreg = 'off' >>> al_ea.inputs.tshift = 'off' >>> al_ea.inputs.save_skullstrip = True >>> al_ea.cmdline # doctest: +ELLIPSIS 'python2 ...align_epi_anat.py -anat structural.nii -epi_base 0 -epi_strip 3dAutomask -epi functional.nii -save_skullstrip -suffix _al -tshift off -volreg off' >>> res = allineate.run() # doctest: +SKIP See Also -------- For complete details, see the `align_epi_anat.py documentation. `__. zalign_epi_anat.pycCs|jj}|j|jj}|j|jj}d|krHdj|jddd}d|krhdj|jddd}|jj}|dkr~d}n t j |}d}|jj }|jj r|j||d|d|d <|j||d |d|d <|jj r|j||d|d|d <|j||d |d|d <|jjdkrv|j|d|d |d|d<|jjdkrT|j|d|d|d<n"|jjdkrv|j|d|d|d<|jjdkr|jj r|j|d|d |d|d<|jjr|j|d|d|_|S)N+r rAFNIz.HEADz.1Dz+orig)rCextrGz _mat.aff12rIrHrJr9Z_vrrKZ tsh_vr_motionrMr:Z vr_motionZ_regrLZ_nsrSz_ns+orig) output_specget _gen_fnameinputsr<r;joinsplit outputtyperoutput_type_to_extrCr@rArDrErBrN)selfoutputsZ anat_prefixZ epi_prefixrZrRZmatextrCr,r,r- _list_outputssN    zAlignEpiAnatPy._list_outputsN) r!r"r#r$_cmdr. input_specrFrTr^r,r,r,r-rOs 4rOc@seZdZeddddddZeddddZed d d d dd gdZedddd gdZeddddgdZedddd gdZ eddddgdZ e j dd d!Z ed"d#dd%ddd&gdZd'd(d)d*d+d,d-d.d/d0d1d2d3d4gZe jed5d6d7Zd8d9d:d;dd?d7Ze jed@dAd7Ze jdBdCd7Ze j dDdEd7Ze j dFdGd7Ze jdHdId7Ze j dJdKd7Ze je jedLdMd7Ze j dNdOd7Ze j dPdQd7Ze jdRdSd7Z e j dTdUd7Z!e jdVdWd7Z"e jdXdYd7Z#e$dZd[d7Z%e$d\d]d7Z&e jd^d_d7Z'e j d`dad7Z(e j dbdcd7Z)eddddedfdgdhZ*e j+eddie jdddjd7Z,edkdld gdZ-eddmdndZ.e jdodpd7Z/e jdqdrdsdtdudvd7Z0e j dwdxd7Z1e j dydzd7Z2e jed{d|d7Z3e j d}d~d7Z4e jddd7Z5e jddd7Z6e jddd7Z7e jddd7Z8eddddZ9e jddd7Z:dd:d;ddddddg Z;e je;ddd7Ze je je=ddd7Z?e j ddd7Z@e j ddd7ZAd=S)AllineateInputSpeczinput file to 3dAllineatez -source %sTF)rrr/rr0z-base %sz{file to be used as reference, the first volume will be used if not given the reference will be the first volume of in_file.)rrrzoutput file from 3dAllineatez -prefix %sz %s_allineater;allcostx)rr name_template name_source hash_filesxorz-1Dparam_save %sz/Save the warp parameters in ASCII (.1D) format. in_param_file)rrrfz-1Dparam_apply %sz^Read warp parameters from file and apply them to the source dataset, and produce a new datasetout_param_file)rrrrfz-1Dmatrix_save %sz/Save the transformation matrix for each volume. in_matrixzmatrix to align input filez-1Dmatrix_apply %sr out_matrix)rrpositionrfz*overwrite output file if it already existsz -overwrite)rrzCompute and print ALL available cost functionals for the un-warped inputsAND THEN QUIT. If you use this option none of the other expected outputs will be producedz-allcostx |& tee %srout_fileout_weight_fileZleastsqZlsZ mutualinfomiZ corratio_mulZcrMZnorm_mutualinfonmiZ hellingerhelZ corratio_addZcrAZ corratio_unsZcrUz-cost %szUDefines the 'cost' function that defines the matching between the source and the base)rrZnearestneighbourlinearcubicquinticZwsinc5Nz -interp %sz3Defines interpolation method to use during matchingz -final %sz>Defines interpolation method used to create the output datasetz -nmatch %dz5Use at most n scattered points to match the datasets.z-nopadz*Do not use zero-padding on the base image.z-zclipzHReplace negative values in the input datasets (source & base) with zero.z-conv %fz1Convergence test in millimeters (default 0.05mm).z-usetempztemporary file usez -check %sa%After cost functional optimization is done, start at the final parameters and RE-optimize using this new cost functions. If the results are too different, a warning message will be printed. However, the final parameters from the original optimization will be used to create the output dataset.z-onepasszUse only the refining pass -- do not try a coarse resolution pass first. Useful if you know that only small amounts of image alignment are needed.z-twopasszxUse a two pass alignment strategy for all volumes, searching for a large rotation+shift and then refining the alignment.z -twoblur %fz1Set the blurring radius for the first pass in mm.z -twofirstzUse -twopass on the first image to be registered, and then on all subsequent images from the source dataset, use results from the first image's coarse pass to start the fine pass.z -twobest %daIn the coarse pass, use the best 'bb' set of initialpoints to search for the starting point for the finepass. If bb==0, then no search is made for the beststarting point, and the identity transformation isused as the starting point. [Default=5; min=0 max=11]z -fineblur %faSet the blurring radius to use in the fine resolution pass to 'x' mm. A small amount (1-2 mm?) of blurring at the fine step may help with convergence, if there is some problem, especially if the base volume is very noisy. [Default == 0 mm = no blurring at the final alignment pass]z-cmass%sz9Use the center-of-mass calculation to bracket the shifts.z -autoweight%sz^Compute a weight function using the 3dAutomask algorithm plus some blurring of the base image.z -automask+%dz7Compute a mask function, set a value for dilation or 0.z-autoboxzYExpand the -automask function to enclose a rectangular box that holds the irregular mask.z-nomaskziDon't compute the autoweight/mask; if -weight is not also used, then every voxel will be counted equally.z -weight %sz1.0.0weightzSet the weighting for each voxel in the base dataset; larger weights mean that voxel count more in the cost function. Must be defined on the same grid as the base dataset)rr deprecatednew_namer)rzSet the weighting for each voxel in the base dataset; larger weights mean that voxel count more in the cost function. If an image file is given, the volume must be defined on the same grid as the base datasetz -wtprefix %sz,Write the weight volume to disk as a datasetz-source_mask %szmask the input datasetz-source_automask+%dz9Automatically mask the source dataset with dilation or 0.Z shift_onlyZ shift_rotateZshift_rotate_scaleZaffine_generalz-warp %szSet the warp type.z -warpfreezez8Freeze the non-rigid body parameters after first volume.z -replacebasez[If the source has more than one volume, then after the first volume is aligned to the base.z-replacemeth %sz\After first volume is aligned, switch method for later volumes. For use with '-replacebase'.z-EPIzTreat the source dataset as being composed of warped EPI slices, and the base as comprising anatomically 'true' images. Only phase-encoding direction image shearing and scaling will be allowed with this option.z -maxrot %fz$Maximum allowed rotation in degrees.z -maxshf %fzMaximum allowed shift in mm.z -maxscl %fzMaximum allowed scaling factor.z -maxshr %fz Maximum allowed shearing factor.z -master %sz7Write the output dataset on the same grid as this file.z -newgrid %fz`_ Examples -------- >>> from nipype.interfaces import afni >>> allineate = afni.Allineate() >>> allineate.inputs.in_file = 'functional.nii' >>> allineate.inputs.out_file = 'functional_allineate.nii' >>> allineate.inputs.in_matrix = 'cmatrix.mat' >>> allineate.cmdline '3dAllineate -source functional.nii -prefix functional_allineate.nii -1Dmatrix_apply cmatrix.mat' >>> res = allineate.run() # doctest: +SKIP >>> allineate = afni.Allineate() >>> allineate.inputs.in_file = 'functional.nii' >>> allineate.inputs.reference = 'structural.nii' >>> allineate.inputs.allcostx = 'out.allcostX.txt' >>> allineate.cmdline '3dAllineate -source functional.nii -base structural.nii -allcostx |& tee out.allcostX.txt' >>> res = allineate.run() # doctest: +SKIP >>> allineate = afni.Allineate() >>> allineate.inputs.in_file = 'functional.nii' >>> allineate.inputs.reference = 'structural.nii' >>> allineate.inputs.nwarp_fixmot = ['X', 'Y'] >>> allineate.cmdline '3dAllineate -source functional.nii -nwarp_fixmotX -nwarp_fixmotY -prefix functional_allineate -base structural.nii' >>> res = allineate.run() # doctest: +SKIP Z 3dAllineatecstt|j}|jjr(tj|jj|d<|jjrvt|jjd }|j d krd|j |jjdd|d<ntj|jj|d<|jj rt|jj d }|j dkr|j |jj dd|d <ntj|jj |d <|jj rt jj|jj |d <|S)Nrmr.1d.1Dz .aff12.1D)rCrjz .param.1DrhZallcostXrS)rrrS)rr)superrr^rWrmopabspathrjrlowerrVrhrbospath)r\r]rR) __class__r,r-r^Js$  zAllineate._list_outputs) r!r"r#r$r_rar`rrTr^ __classcell__r,r,)rr-r$s  rc@s~eZdZedddddddZejddd Zejd d d Z edd d dZ ejdddgdZ eddddgdZ edddddZ dS)AutoTcorrelateInputSpecz+timeseries x space (volume or surface) filez%srTF)rrrkr/rr0z;Remove polynomical trend of order m or -1 for no detrendingz -polort %d)rrzeta^2 similarityz-eta2zmask of voxelsz-mask %s)rrrzuse mask only on targets voxelsz-mask_only_targets mask_source)rrrfzmask for source voxelsz-mask_source %smask_only_targets)rrrrfz%s_similarity_matrix.1Dzoutput image file namez -prefix %sr;)rcrrrdNrS)r!r"r#r r;r r'r(r)Zeta2r%rrrlr,r,r,r-rgs4  rc@s&eZdZdZeZeZdZdddZ dS)AutoTcorrelateatComputes the correlation coefficient between the time series of each pair of voxels in the input dataset, and stores the output into a new anatomical bucket dataset [scaled to shorts to save memory space]. For complete details, see the `3dAutoTcorrelate Documentation. `_ Examples -------- >>> from nipype.interfaces import afni >>> corr = afni.AutoTcorrelate() >>> corr.inputs.in_file = 'functional.nii' >>> corr.inputs.polort = -1 >>> corr.inputs.eta2 = True >>> corr.inputs.mask = 'mask.nii' >>> corr.inputs.mask_only_targets = True >>> corr.cmdline # doctest: +ELLIPSIS '3dAutoTcorrelate -eta2 -mask mask.nii -mask_only_targets -prefix functional_similarity_matrix.1D -polort -1 functional.nii' >>> res = corr.run() # doctest: +SKIP Z3dAutoTcorrelateNcCs4t|\}}}|jdkr"|d}tjj|||S)N.1d.1D.nii.gz.nii)rrrr)rrrrrX)r\valuenamerbaserRr,r,r-_overload_extensions z"AutoTcorrelate._overload_extension)N) r!r"r#r$rr`rrTr_rr,r,r,r-rs rc@sjeZdZedddddddZeddd d d Zed d dd d ZejdddZ ej dddZ ej dddZ dS)AutomaskInputSpeczinput file to 3dAutomaskz%srTF)rrrkr/rr0z%s_maskzoutput image file namez -prefix %sr;)rcrrrdz %s_maskedzoutput file from 3dAutomaskz-apply_prefix %szpsets the clip level fraction (must be 0.1-0.9). A small value will tend to make the mask larger [default = 0.5].z -clfrac %s)rrzdilate the mask outwardsz -dilate %szerode the mask inwardsz -erode %sNrS) r!r"r#r r;rl brain_filer r&Zclfracr'ZdilateZeroder,r,r,r-rs,rc@s$eZdZedddZedddZdS)AutomaskOutputSpecz mask fileT)rrzbrain file (skull stripped)N)r!r"r#r rlrr,r,r,r-rs rc@seZdZdZdZeZeZdS)AutomaskaCreate a brain-only mask of the image using AFNI 3dAutomask command For complete details, see the `3dAutomask Documentation. `_ Examples -------- >>> from nipype.interfaces import afni >>> automask = afni.Automask() >>> automask.inputs.in_file = 'functional.nii' >>> automask.inputs.dilate = 1 >>> automask.inputs.outputtype = 'NIFTI' >>> automask.cmdline # doctest: +ELLIPSIS '3dAutomask -apply_prefix functional_masked.nii -dilate 1 -prefix functional_mask.nii functional.nii' >>> res = automask.run() # doctest: +SKIP r7N) r!r"r#r$r_rr`rrTr,r,r,r-rsrc@sVeZdZejdeejjddZ e ddddddZ ej d dd d Z ejd d dZdS)AutoTLRCInputSpecrQzAFNI output filetype)rzkOriginal anatomical volume (+orig).The skull is removed by this scriptunless instructed otherwise (-no_ss).z -input %sTF)rrr/rr0apReference anatomical volume. Usually this volume is in some standard space like TLRC or MNI space and with afni dataset view of (+tlrc). Preferably, this reference volume should have had the skull removed but that is not mandatory. AFNI's distribution contains several templates. For a longer list, use "whereami -show_templates" TT_N27+tlrc --> Single subject, skull stripped volume. This volume is also known as N27_SurfVol_NoSkull+tlrc elsewhere in AFNI and SUMA land. (www.loni.ucla.edu, www.bic.mni.mcgill.ca) This template has a full set of FreeSurfer (surfer.nmr.mgh.harvard.edu) surface models that can be used in SUMA. For details, see Talairach-related link: https://afni.nimh.nih.gov/afni/suma TT_icbm452+tlrc --> Average volume of 452 normal brains. Skull Stripped. (www.loni.ucla.edu) TT_avg152T1+tlrc --> Average volume of 152 normal brains. Skull Stripped.(www.bic.mni.mcgill.ca) TT_EPI+tlrc --> EPI template from spm2, masked as TT_avg152T1 TT_avg152 and TT_EPI volume sources are from SPM's distribution. (www.fil.ion.ucl.ac.uk/spm/) If you do not specify a path for the template, the script will attempt to locate the template AFNI's binaries directory. NOTE: These datasets have been slightly modified from their original size to match the standard TLRC dimensions (Jean Talairach and Pierre Tournoux Co-Planar Stereotaxic Atlas of the Human Brain Thieme Medical Publishers, New York, 1988). That was done for internal consistency in AFNI. You may use the original form of these volumes if you choose but your TLRC coordinates will not be consistent with AFNI's TLRC database (San Antonio Talairach Daemon database), for example.z-base %s)rr/raDo not strip skull of input data set (because skull has already been removed or because template still has the skull) NOTE: The ``-no_ss`` option is not all that optional. Here is a table of when you should and should not use ``-no_ss`` +------------------+------------+---------------+ | Dataset | Template | +==================+============+===============+ | | w/ skull | wo/ skull | +------------------+------------+---------------+ | WITH skull | ``-no_ss`` | xxx | +------------------+------------+---------------+ | WITHOUT skull | No Cigar | ``-no_ss`` | +------------------+------------+---------------+ Template means: Your template of choice Dset. means: Your anatomical dataset ``-no_ss`` means: Skull stripping should not be attempted on Dset xxx means: Don't put anything, the script will strip Dset No Cigar means: Don't try that combination, it makes no sense.z-no_ss)rrN)r!r"r#r r?listrZftypeskeysrZr r;rrr)Zno_ssr,r,r,r-rs&rc@s$eZdZdZdZeZeZddZ dS)AutoTLRCa\A minmal wrapper for the AutoTLRC script The only option currently supported is no_ss. For complete details, see the `3dQwarp Documentation. `_ Examples -------- >>> from nipype.interfaces import afni >>> autoTLRC = afni.AutoTLRC() >>> autoTLRC.inputs.in_file = 'structural.nii' >>> autoTLRC.inputs.no_ss = True >>> autoTLRC.inputs.base = "TT_N27+tlrc" >>> autoTLRC.cmdline '@auto_tlrc -base TT_N27+tlrc -input structural.nii -no_ss' >>> res = autoTLRC.run() # doctest: +SKIP z @auto_tlrccCs6|jj}d}tjj|j|jjdd||d<|S)Nz.HEADz+tlrc)rCrl)rTrUrrrrVrWr;)r\r]rRr,r,r-r^Ns  zAutoTLRC._list_outputsN) r!r"r#r$r_rr`rrTr^r,r,r,r-r7s rc@seZdZeddd/ddddZeddd dd d Zejd d d0ddZejdd d1ddZ edddddZ ej dddZ e edddddZeddddZej dddZejd d!dZejd"d#dZej d$d%dZej d&d'dZejd(d)dZejd*d+dZej d,d-dZd.S)2BandpassInputSpeczinput file to 3dBandpassz%srTF)rrrkr/rr0z%s_bpzoutput file from 3dBandpassz -prefix %sr;)rcrrrkrdlowpassz%fr)rrrkr/highpassrz mask filez-mask %s)rrkrrz-despikez}Despike each time series before other processing. Hopefully, you don't actually need to do this, which is why it is optional.)rr)rz-ort %szQAlso orthogonalize input to columns in f.1D. Multiple '-ort' options are allowed.z -dsort %szOrthogonalize each voxel to the corresponding voxel time series in dataset 'fset', which must have the same spatial and temporal grid structure as the main input dataset. At present, only one '-dsort' option is allowed.)rrrz -nodetrendzSkip the quadratic detrending of the input that occurs before the FFT-based bandpassing. You would only want to do this if the dataset had been detrended already in some other program.z-dt %fz8Set time step (TR) in sec [default=from dataset header].z-nfft %dz+Set the FFT length [must be a legal value].z-normzHMake all output time series have L2 norm = 1 (i.e., sum of squares = 1).z -automaskz%Create a mask from the input dataset.z-blur %fzLBlur (inside the mask only) with a filter width (FWHM) of 'fff' millimeters.z -localPV %fzReplace each vector by the local Principal Vector (AKA first singular vector) from a neighborhood of radius 'rrr' millimeters. Note that the PV time series is L2 normalized. This option is mostly for Bob Cox to have fun with.z-notranszDon't check for initial positive transients in the data. The test is a little slow, so skipping it is OK, if you KNOW the data time series are transient-free.NrSr~)r!r"r#r r;rlr r&rrr%r)ZdespikerZorthogonalize_fileZorthogonalize_dsetZ no_detrendtrr'Znfft normalizer+blurZlocalPVZnotransr,r,r,r-rWs`   rc@seZdZdZdZeZeZdS)BandpassaProgram to lowpass and/or highpass each voxel time series in a dataset, offering more/different options than Fourier For complete details, see the `3dBandpass Documentation. `_ Examples -------- >>> from nipype.interfaces import afni >>> from nipype.testing import example_data >>> bandpass = afni.Bandpass() >>> bandpass.inputs.in_file = 'functional.nii' >>> bandpass.inputs.highpass = 0.005 >>> bandpass.inputs.lowpass = 0.1 >>> bandpass.cmdline '3dBandpass -prefix functional_bp 0.005000 0.100000 functional.nii' >>> res = bandpass.run() # doctest: +SKIP Z 3dBandpassN) r!r"r#r$r_rr`rrTr,r,r,r-rsrc@seZdZedddddddZeddd d dd Zed d dZedddZej dddZ ej ddddZ ej dddZ ej dddZeddddZdS) BlurInMaskInputSpeczinput file to 3dSkullStripz -input %srTF)rrrkr/rr0z%s_blurzoutput to the filez -prefix %sr;)rcrrrdrkz~Mask dataset, if desired. Blurring will occur only within the mask. Voxels NOT in the mask will be set to zero in the output.z-mask %s)rrzvMulti-mask dataset -- each distinct nonzero value in dataset will be treated as a separate mask for blurring purposes.z -Mmask %sz*Create an automask from the input dataset.z -automaskzfwhm kernel sizez-FWHM %f)rrr/zNormally, voxels not in the mask will be set to zero in the output. If you want the original values in the dataset to be preserved in the output, use this option.z -preservez>Save dataset as floats, no matter what the input data type is.z-floatoptionsz%sr)rrrkNrS)r!r"r#r r;rlr%Z multimaskr r)r+r&fwhmZpreserveZ float_outrrr,r,r,r-rs: rc@seZdZdZdZeZeZdS) BlurInMaskabBlurs a dataset spatially inside a mask. That's all. Experimental. For complete details, see the `3dBlurInMask Documentation. `_ Examples -------- >>> from nipype.interfaces import afni >>> bim = afni.BlurInMask() >>> bim.inputs.in_file = 'functional.nii' >>> bim.inputs.mask = 'mask.nii' >>> bim.inputs.fwhm = 5.0 >>> bim.cmdline # doctest: +ELLIPSIS '3dBlurInMask -input functional.nii -FWHM 5.000000 -mask mask.nii -prefix functional_blur' >>> res = bim.run() # doctest: +SKIP Z 3dBlurInMaskN) r!r"r#r$r_rr`rrTr,r,r,r-rsrc@sbeZdZedddddZejdddZejdd dZ ejd d dZ ed d ddZ eddddZ dS)BlurToFWHMInputSpecz!The dataset that will be smoothedz -input %sT)rrr/rz*Create an automask from the input dataset.z -automask)rrz1Blur until the 3D FWHM reaches this value (in mm)z-FWHM %fz=Blur until the 2D (x,y)-plane FWHM reaches this value (in mm)z -FWHMxy %fz2The dataset whose smoothness controls the process.z-blurmaster %s)rrrzKMask dataset, if desired. Voxels NOT in mask will be set to zero in output.z-mask %sN) r!r"r#r r;r r)r+r&rZfwhmxyZ blurmasterr%r,r,r,r-rs(  rc@seZdZdZdZeZeZdS) BlurToFWHMa>Blurs a 'master' dataset until it reaches a specified FWHM smoothness (approximately). For complete details, see the `3dBlurToFWHM Documentation `_ Examples -------- >>> from nipype.interfaces import afni >>> blur = afni.preprocess.BlurToFWHM() >>> blur.inputs.in_file = 'epi.nii' >>> blur.inputs.fwhm = 2.5 >>> blur.cmdline # doctest: +ELLIPSIS '3dBlurToFWHM -FWHM 2.500000 -input epi.nii -prefix epi_afni' >>> res = blur.run() # doctest: +SKIP Z 3dBlurToFWHMN) r!r"r#r$r_rr`rrTr,r,r,r-r!src@sPeZdZeddddddZejdddd Zejd d d d dZ eddd ddZ dS)ClipLevelInputSpeczinput file to 3dClipLevelz%srT)rrrkr/rz2Use the number ff instead of 0.50 in the algorithmz -mfrac %sr)rrrkz1Apply the algorithm to each sub-brick separately.z-doallrgrad)rrrkrfzbAlso compute a 'gradual' clip level as a function of voxel position, and output that to a dataset.z-grad %sdoallNrS) r!r"r#r r;r r&Zmfracr)rrr,r,r,r-r9s(rc@seZdZejddZdS)ClipLevelOutputSpecoutput)rN)r!r"r#r r&clip_valr,r,r,r-rUsrc@s&eZdZdZdZeZeZdddZ dS) ClipLevelaEstimates the value at which to clip the anatomical dataset so that background regions are set to zero. For complete details, see the `3dClipLevel Documentation. `_ Examples -------- >>> from nipype.interfaces.afni import preprocess >>> cliplevel = preprocess.ClipLevel() >>> cliplevel.inputs.in_file = 'anatomical.nii' >>> cliplevel.cmdline '3dClipLevel anatomical.nii' >>> res = cliplevel.run() # doctest: +SKIP Z 3dClipLevelNc Cs|j}tjjtjd}|dkrPyt|d}Wqtk rL|jjSXng}xV|j j dD]F}|rb|j }t |dkr|j dd|Dqb|j dd|DqbWt |dkr|d}t|t|d ||_|S) Nzstat_result.jsonstat rcSsg|] }t|qSr,)float).0valr,r,r- sz/ClipLevel.aggregate_outputs..cSsg|] }t|qSr,)r)rrr,r,r-rsr)r)_outputsrrrXgetcwdrIOErrorrunr]stdoutrYlenappendextendrdictr)r\runtimeneeded_outputsr]outfilerlinevaluesr,r,r-aggregate_outputsos&  zClipLevel.aggregate_outputs)NN) r!r"r#r$r_rr`rrTrr,r,r,r-rYs rc@s>eZdZdZedddddddZejdd d Ze d d d Z d S)DegreeCentralityInputSpeczDegreeCentrality inputspecz input file to 3dDegreeCentralityz%srTF)rrrkr/rr0z(only take the top percent of connectionsz -sparsity %f)rrz2output filepath to text dump of correlation matrixz -out1D %sNrS) r!r"r#r$r r;r r&sparsityr oned_filer,r,r,r-rs rc@seZdZdZeddZdS)DegreeCentralityOutputSpeczDegreeCentrality outputspeczThe text output of the similarity matrix computed after thresholding with one-dimensional and ijk voxel indices, correlations, image extents, and affine matrix.)rN)r!r"r#r$r rr,r,r,r-rsrcs,eZdZdZdZeZeZfddZ Z S)DegreeCentralityaPerforms degree centrality on a dataset using a given maskfile via 3dDegreeCentrality For complete details, see the `3dDegreeCentrality Documentation. `_ Examples -------- >>> from nipype.interfaces import afni >>> degree = afni.DegreeCentrality() >>> degree.inputs.in_file = 'functional.nii' >>> degree.inputs.mask = 'mask.nii' >>> degree.inputs.sparsity = 1 # keep the top one percent of connections >>> degree.inputs.out_file = 'out.nii' >>> degree.cmdline '3dDegreeCentrality -mask mask.nii -prefix out.nii -sparsity 1.000000 functional.nii' >>> res = degree.run() # doctest: +SKIP Z3dDegreeCentralitycs.tt|j}|jjr*tjj|jj|d<|S)Nr)rrr^rWrrrr)r\r])rr,r-r^szDegreeCentrality._list_outputs) r!r"r#r$r_rr`rrTr^rr,r,)rr-rs rc@s0eZdZeddd ddddZeddd d d Zd S)DespikeInputSpeczinput file to 3dDespikez%srTF)rrrkr/rr0z %s_despikezoutput image file namez -prefix %sr;)rcrrrdNrS)r!r"r#r r;rlr,r,r,r-rsrc@seZdZdZdZeZeZdS)DespikeaRemoves 'spikes' from the 3D+time input dataset For complete details, see the `3dDespike Documentation. `_ Examples -------- >>> from nipype.interfaces import afni >>> despike = afni.Despike() >>> despike.inputs.in_file = 'functional.nii' >>> despike.cmdline '3dDespike -prefix functional_despike functional.nii' >>> res = despike.run() # doctest: +SKIP Z 3dDespikeN) r!r"r#r$r_rr`rrTr,r,r,r-rsrc@s0eZdZeddd ddddZeddd d d Zd S)DetrendInputSpeczinput file to 3dDetrendz%srTF)rrrkr/rr0z %s_detrendzoutput image file namez -prefix %sr;)rcrrrdNrS)r!r"r#r r;rlr,r,r,r-rsrc@seZdZdZdZeZeZdS)Detrenda[This program removes components from voxel time series using linear least squares For complete details, see the `3dDetrend Documentation. `_ Examples -------- >>> from nipype.interfaces import afni >>> detrend = afni.Detrend() >>> detrend.inputs.in_file = 'functional.nii' >>> detrend.inputs.args = '-polort 2' >>> detrend.inputs.outputtype = 'AFNI' >>> detrend.cmdline '3dDetrend -polort 2 -prefix functional_detrend functional.nii' >>> res = detrend.run() # doctest: +SKIP Z 3dDetrendN) r!r"r#r$r_rr`rrTr,r,r,r-rsrc@seZdZdZedddddddZejdd d Zej d d d Z ej d dd Z ejddd Z ejddd Z ejddd Zejddd Zejddd ZdS) ECMInputSpecz ECM inputspeczinput file to 3dECMz%srTF)rrrkr/rr0z(only take the top percent of connectionsz -sparsity %f)rrz_Full power method; enables thresholding; automatically selected if -thresh or -sparsity are setz-fullzFast centrality method; substantial speed increase but cannot accomodate thresholding; automatically selected if -thresh or -sparsity are not setz-fecmzshift correlation coefficients in similarity matrix to enforce non-negativity, s >= 0.0; default = 0.0 for -full, 1.0 for -fecmz -shift %fzwscale correlation coefficients in similarity matrix to after shifting, x >= 0.0; default = 1.0 for -full, 0.5 for -fecmz -scale %fzsets the stopping criterion for the power iteration; :math:`l2\|v_\text{old} - v_\text{new}\| < eps\|v_\text{old}\|`; default = 0.001z-eps %fzSsets the maximum number of iterations to use in the power iteration; default = 1000z -max_iter %dzgLimit memory consumption on system by setting the amount of GB to limit the algorithm to; default = 2GBz -memory %fNrS)r!r"r#r$r r;r r&rr)fullZfecmshiftscaleepsr'Zmax_iterZmemoryr,r,r,r-rs> rc@seZdZdZdZeZeZdS)ECMaPerforms degree centrality on a dataset using a given maskfile via the 3dECM command For complete details, see the `3dECM Documentation. `_ Examples -------- >>> from nipype.interfaces import afni >>> ecm = afni.ECM() >>> ecm.inputs.in_file = 'functional.nii' >>> ecm.inputs.mask = 'mask.nii' >>> ecm.inputs.sparsity = 0.1 # keep top 0.1% of connections >>> ecm.inputs.out_file = 'out.nii' >>> ecm.cmdline '3dECM -mask mask.nii -prefix out.nii -sparsity 0.100000 functional.nii' >>> res = ecm.run() # doctest: +SKIP Z3dECMN) r!r"r#r$r_rr`rrTr,r,r,r-rRsrc@s`eZdZedddddddZeddd d d Zed d ddddZejddddZ e ddddZ dS) FimInputSpeczinput file to 3dfim+z -input %srTF)rrrkr/rr0z%s_fimzoutput image file namez -bucket %sr;)rcrrrdzideal time series file namez-ideal_file %sr)rrrkr/rz!fim internal mask threshold valuez -fim_thr %fr)rrrkz&Flag to output the specified parameterz-out %sN) r!r"r#r r;rlZ ideal_filer r&Zfim_thrroutr,r,r,r-rls, rc@seZdZdZdZeZeZdS)FimaProgram to calculate the cross-correlation of an ideal reference waveform with the measured FMRI time series for each voxel. For complete details, see the `3dfim+ Documentation. `_ Examples -------- >>> from nipype.interfaces import afni >>> fim = afni.Fim() >>> fim.inputs.in_file = 'functional.nii' >>> fim.inputs.ideal_file= 'seed.1D' >>> fim.inputs.out_file = 'functional_corr.nii' >>> fim.inputs.out = 'Correlation' >>> fim.inputs.fim_thr = 0.0009 >>> fim.cmdline '3dfim+ -input functional.nii -ideal_file seed.1D -fim_thr 0.000900 -out Correlation -bucket functional_corr.nii' >>> res = fim.run() # doctest: +SKIP z3dfim+N) r!r"r#r$r_rr`rrTr,r,r,r-rsrc@s^eZdZedddddddZeddd d d Zejd d ddZejddddZ ej dddZ dS)FourierInputSpeczinput file to 3dFourierz%srTF)rrrkr/rr0z %s_fourierzoutput image file namez -prefix %sr;)rcrrrdrz -lowpass %f)rrr/rz -highpass %fzdAny mean and linear trend are removed before filtering. This will restore the trend after filtering.z-retrend)rrNrS) r!r"r#r r;rlr r&rrr)Zretrendr,r,r,r-rs"rc@seZdZdZdZeZeZdS)FourieraProgram to lowpass and/or highpass each voxel time series in a dataset, via the FFT For complete details, see the `3dFourier Documentation. `_ Examples -------- >>> from nipype.interfaces import afni >>> fourier = afni.Fourier() >>> fourier.inputs.in_file = 'functional.nii' >>> fourier.inputs.retrend = True >>> fourier.inputs.highpass = 0.005 >>> fourier.inputs.lowpass = 0.1 >>> fourier.cmdline '3dFourier -highpass 0.005000 -lowpass 0.100000 -prefix functional_fourier -retrend functional.nii' >>> res = fourier.run() # doctest: +SKIP Z 3dFourierN) r!r"r#r$r_rr`rrTr,r,r,r-rsrc@seZdZedddddddZedddd d gd Zejddd d dZeddddd d!dZ eddddZ ej dddZ ej dddZej dddZej dddZd S)" HistInputSpeczinput file to 3dHistz -input %srTF)rrrkr/rr0z-Write histogram to niml file with this prefixz%s_histz -prefix %sr;)rrckeep_extensionrrdzwrite a text visual histogramz -showhist)r5rrz %s_hist.outzoutput image file namez> %s)rcrrrrdrkzmatrix to align input filez-mask %s)rrrznumber of binsz-nbin %d)rrz-max %fzmaximum intensity value)rrz-min %fzminimum intensity valuez -binwidth %fz bin widthNrS)r!r"r#r r;rlr r)showhistout_showr%r'Znbinr&Z max_valueZ min_valueZ bin_widthr,r,r,r-rs6 rc@s"eZdZedddZeddZdS)HistOutputSpecz output fileT)rrzoutput visual histogram)rN)r!r"r#r rlrr,r,r,r-rs rcsJeZdZdZdZeZeZdZ fddZ d fdd Z fd d Z Z S) HistaComputes average of all voxels in the input dataset which satisfy the criterion in the options list For complete details, see the `3dHist Documentation. `_ Examples -------- >>> from nipype.interfaces import afni >>> hist = afni.Hist() >>> hist.inputs.in_file = 'functional.nii' >>> hist.cmdline '3dHist -input functional.nii -prefix functional_hist' >>> res = hist.run() # doctest: +SKIP Z3dHistTc s6tt|jf|ts2tj}|ddkr2d|_dS)NriF)rr__init__rrversion _redirect_x)r\rWr)rr,r-rs  z Hist.__init__Ncs0|jjs|dkrg}|dg7}tt|j|dS)Nr)skip)rWrrr _parse_inputs)r\r)rr,r-rs  zHist._parse_inputscs2tt|j}|dd7<|jjs.t|d<|S)Nrlz .niml.histr)rrr^rWrr)r\r])rr,r-r^%s zHist._list_outputs)N)r!r"r#r$r_rr`rrTrrrr^rr,r,)rr-rs rc@s$eZdZdZeddd ddddZdS) LFCDInputSpeczLFCD inputspeczinput file to 3dLFCDz%srTF)rrrkr/rr0NrS)r!r"r#r$r r;r,r,r,r-r-src@seZdZdZdZeZeZdS)LFCDaPerforms degree centrality on a dataset using a given maskfile via the 3dLFCD command For complete details, see the `3dLFCD Documentation. `_ Examples -------- >>> from nipype.interfaces import afni >>> lfcd = afni.LFCD() >>> lfcd.inputs.in_file = 'functional.nii' >>> lfcd.inputs.mask = 'mask.nii' >>> lfcd.inputs.thresh = 0.8 # keep all connections with corr >= 0.8 >>> lfcd.inputs.out_file = 'out.nii' >>> lfcd.cmdline '3dLFCD -mask mask.nii -prefix out.nii -thresh 0.800000 functional.nii' >>> res = lfcd.run() # doctest: +SKIP Z3dLFCDN) r!r"r#r$r_rr`rrTr,r,r,r-r:src@sTeZdZedddddddZedddd d dd Zed dd ddZejd dddZ dS)MaskaveInputSpeczinput file to 3dmaskavez%srTF)rrrkr/rr0z %s_maskave.1Dzoutput image file namez> %sr;r)rcrrrrdrkzmatrix to align input filez-mask %s)rrrkrz-quiet)rrrkNrrS) r!r"r#r r;rlr%r r)rr,r,r,r-rSs"rc@seZdZdZdZeZeZdS)MaskaveaComputes average of all voxels in the input dataset which satisfy the criterion in the options list For complete details, see the `3dmaskave Documentation. `_ Examples -------- >>> from nipype.interfaces import afni >>> maskave = afni.Maskave() >>> maskave.inputs.in_file = 'functional.nii' >>> maskave.inputs.mask= 'seed_mask.nii' >>> maskave.inputs.quiet= True >>> maskave.cmdline # doctest: +ELLIPSIS '3dmaskave -mask seed_mask.nii -quiet functional.nii > functional_maskave.1D' >>> res = maskave.run() # doctest: +SKIP Z 3dmaskaveN) r!r"r#r$r_rr`rrTr,r,r,r-rjsrc@seZdZeddd"dddZeddd#ddZejd d d Zed d dddZ eddd Z ej ddd Z ej ddd Z ej ddd Zej ddd Zej ddd Zej ddd Zej dd d Zd!S)$MeansInputSpeczinput file to 3dMeanz%srT)rrrkr/rzanother input file to 3dMeanr)rrrkrz(Sets the data type of the output datasetz -datum %s)rrz%s_meanzoutput image file namez -prefix %s in_file_a)rcrrrdzscaling of outputz-%sscalezuse only non-zero valuesz -non_zerozcalculate std devz-stdevzmean square instead of valuez-sqrztake sum, (not average)z-sumz compute count of non-zero voxelsz-countzcreate intersection maskz -mask_interzcreate union maskz -mask_unionNrrS)r!r"r#r rZ in_file_br rdatumrlrr)Znon_zeroZstd_devZsqrZsummcountZ mask_interZ mask_unionr,r,r,r-rs.  rc@seZdZdZdZeZeZdS)Meansa_Takes the voxel-by-voxel mean of all input datasets using 3dMean For complete details, see the `3dMean Documentation. `_ Examples -------- >>> from nipype.interfaces import afni >>> means = afni.Means() >>> means.inputs.in_file_a = 'im1.nii' >>> means.inputs.in_file_b = 'im2.nii' >>> means.inputs.out_file = 'output.nii' >>> means.cmdline '3dMean -prefix output.nii im1.nii im2.nii' >>> res = means.run() # doctest: +SKIP >>> from nipype.interfaces import afni >>> means = afni.Means() >>> means.inputs.in_file_a = 'im1.nii' >>> means.inputs.out_file = 'output.nii' >>> means.inputs.datum = 'short' >>> means.cmdline '3dMean -datum short -prefix output.nii im1.nii' >>> res = means.run() # doctest: +SKIP Z3dMeanN) r!r"r#r$r_rr`rrTr,r,r,r-rsrc@seZdZedddd,ddZeddddgd d Zejd d d ddddZej ddddgddZ ej ddddgddZ ej dddddZ ej dddddZ ej ddddZeddd gd!dd"d#Zejd$d%d&Zej ddd'd(dZedd gdd)d*Zd+S)-OutlierCountInputSpecz%sTrz input dataset)rr/rrkrz-mask %sr*r+z'only count voxels within the given mask)rrrfrgMbP?gg?z -qthr %.5fz'indicate a value for q to compute alpha)rr1highr5rrFz -autoclipr%zclip off small voxels)r5rrfrz -automaskz -fractionzLwrite out the fraction of masked voxels which are outliers at each timepoint)r5rrz-rangezCwrite out the median + 3.5 MAD of outlier count with each timepointzenables out_file option)r5rz %s_outliersz-save %sr; out_outlierszoutput image file name)rcrrdZ output_namerrz -polort %dz.detrend each voxel timeseries with polynomials)rrz -legendrezuse Legendre polynomialszcapture standard output)rcrdrrNr)r!r"r#r r;r%r r>Zqthrr)r*r+fractioninterval save_outliers outliers_filer'r(Zlegendrerlr,r,r,r-rsj rc@s"eZdZedddZeddZdS)OutlierCountOutputSpecTzoutput image file name)rrzcapture standard output)rN)r!r"r#r rrlr,r,r,r-rs rcsHeZdZdZdZeZeZdZ d fdd Z dfdd Z d d Z Z S) OutlierCountaCalculates number of 'outliers' at each time point of a a 3D+time dataset. For complete details, see the `3dToutcount Documentation `_ Examples -------- >>> from nipype.interfaces import afni >>> toutcount = afni.OutlierCount() >>> toutcount.inputs.in_file = 'functional.nii' >>> toutcount.cmdline # doctest: +ELLIPSIS '3dToutcount -qthr 0.00100 functional.nii' >>> res = toutcount.run() # doctest: +SKIP Z 3dToutcount file_splitNcs>|dkr g}|jdkrd|_|jjs.|dg7}tt|j|S)Nnonerr)Zterminal_outputrWrrrr)r\r)rr,r-r"s  zOutlierCount._parse_inputsrc sHtt|j||}ttj|jjd}|j|j p6|j WdQRX|S)Nw) rr_run_interfaceopenrrrWrlwriterZmerged)r\rcorrect_return_codesZoutfh)rr,r-r/s  zOutlierCount._run_interfacecCs<|jj}tj|jj|d<|jjr8tj|jj|d<|S)Nrlr)rTrUrrrWrlrr)r\r]r,r,r-r^9s  zOutlierCount._list_outputs)Nr)r )r!r"r#r$r_rr`rrT_terminal_outputrrr^rr,r,)rr-r s  rc@seZdZedddd"ddZeddddgd d Zejd dd d dZejd ddddZ ejd dddgddZ ejd dddgddZ ej dddZ ejd ddddZeddgdd d#dd Zd!S)$QualityIndexInputSpecz%sTrz input dataset)rr/rrkrz-mask %sr*r+z-compute correlation only across masked voxels)rrrfrFz -spearmanz|Quality index is 1 minus the Spearman (rank) correlation coefficient of each sub-brick with the median sub-brick. (default).)r5rrz -quadrantzbSimilar to -spearman, but using 1 minus the quadrant correlation coefficient as the quality index.z -autoclipr%zclip off small voxels)r5rrfrz -automaskz-clip %fzclip off values below)rrz-rangezCwrite out the median + 3.5 MAD of outlier count with each timepointz%s_tqualr;z> %srzcapture standard output)rcrdrrrkrNrrS)r!r"r#r r;r%r r)spearmanquadrantr*r+r&Zcliprrlr,r,r,r-r AsTr c@seZdZeddZdS)QualityIndexOutputSpecz,file containing the captured standard output)rN)r!r"r#r rlr,r,r,r-rysrc@seZdZdZdZeZeZdS) QualityIndexasComputes a quality index for each sub-brick in a 3D+time dataset. The output is a 1D time series with the index for each sub-brick. The results are written to stdout. Examples -------- >>> from nipype.interfaces import afni >>> tqual = afni.QualityIndex() >>> tqual.inputs.in_file = 'functional.nii' >>> tqual.cmdline # doctest: +ELLIPSIS '3dTqual functional.nii > functional_tqual' >>> res = tqual.run() # doctest: +SKIP See Also -------- For complete details, see the `3dTqual Documentation `_ Z3dTqualN) r!r"r#r$r_r r`rrTr,r,r,r-r}src @seZdZeddd>dddZeddddd d d Zedddd Zejd ddZ ej dddZ ej dgdddZ eddddZejdddZejdddZejdddZejdd dZejd!gd"d#d$Zejd%gd&d'd$Zd(d)d*d+d,d-d.d/d0d1d2d3g Zeejed4d5dZed6d7d8d9d:d?d<Zd=S)@ROIStatsInputSpecz input datasetz%srT)rrrkr/rz input maskz-mask %srz1.1.4 mask_file)rrrkrrurv)rrrzPTells the program to convert a float mask to short integers, by simple rounding.z -mask_f2short)rraForces the assumption that the mask dataset's ROIs are denoted by 1 to n inclusive. Normally, the program figures out the ROIs on its own. This option is useful if a) you are certain that the mask dataset has no values outside the range [0 n], b) there may be some ROIs missing between [1 n] in the mask data-set and c) you want those columns in the output any-way so the output lines up with the output from other invocations of 3dROIstats.z -numroi %snum_roiz{For ROI labels not found, use the provided string instead of a '0' in the output file. Only active if `num_roi` is enabled.z -zerofill %s)requiresrrzOnly considers ROIs denoted by values found in the specified file. Note that the order of the ROIs as specified in the file is not preserved. So an SEL.1D of '2 8 20' produces the same output as '8 20 2'z -roisel %s)rrrzprint debug informationz-debugzexecute quietlyz-quietz=Do not include the (zero-inclusive) mean among computed statsz -nomeanoutz2Do not print the sub-brick label next to its indexz -nobriklab format1DRz %sr;r)rcrrrrdrkNrrS)r!r"r#r r;r%rr r)Z mask_f2shortr'rrZzerofillZroiseldebugrZ nomeanoutZ nobriklabrrZ _stat_namesrr?rrlr,r,r,r-rs|  rc@seZdZedddZdS)ROIStatsOutputSpecz output tab-separated values fileT)rrN)r!r"r#r rlr,r,r,r-r"sr"cs0eZdZdZdZdZeZeZ fddZ Z S)ROIStatsaDisplay statistics over masked regions For complete details, see the `3dROIstats Documentation `_ Examples -------- >>> from nipype.interfaces import afni >>> roistats = afni.ROIStats() >>> roistats.inputs.in_file = 'functional.nii' >>> roistats.inputs.mask_file = 'skeleton_mask.nii.gz' >>> roistats.inputs.stat = ['mean', 'median', 'voxels'] >>> roistats.inputs.nomeanout = True >>> roistats.cmdline '3dROIstats -mask skeleton_mask.nii.gz -nomeanout -nzmean -nzmedian -nzvoxels functional.nii > functional_roistat.1D' >>> res = roistats.run() # doctest: +SKIP Z 3dROIstatsZ allatoncec sLddddddddd d d d d |dkr8fdd|D}tt|j|||S)Nz-nzmeanz -nzmedianz-nzmodez -nzminmaxz-nzsigmaz -nzvoxelsz-nzsumz-summaryz-minmaxz-medianz-sigmaz-mode) r2r3rrrrrrrrrr rcsg|] }|qSr,r,)rv) _stat_dictr,r-r=sz(ROIStats._format_arg..)rr# _format_arg)r\r trait_specr)r)r%r-r&-szROIStats._format_arg) r!r"r#r$r_r rr`r"rTr&rr,r,)rr-r#s r#c@seZdZeddd"ddddZeddgd d dd Zed d d#ddZeddd$ddZej ddd%dZ ej ddd&dZ eddd'ddZ eddd(ddZ d!S))RetroicorInputSpeczinput file to 3dretroicorz%srTF)rrrkr/rr0z %s_retroicorr;zoutput image file namez -prefix %s)rcrdrrrkz+1D cardiac data file for cardiac correctionz-card %sr)rrrkrz+1D respiratory waveform data for correctionz-resp %srzThreshold for detection of R-wave peaks in input (Make sure it is above the background noise level, Try 3/4 or 4/5 times range plus minimum)z -threshold %dr)rrrkz*The order of the correction (2 is typical)z -order %sz$Filename for 1D cardiac phase outputz -cardphase %s)rrrkrez!Filename for 1D resp phase outputz -respphase %sNrSrr~ii)r!r"r#r r;rlcardrespr r' thresholdorderZ cardphaseZ respphaser,r,r,r-r(AsRr(cs,eZdZdZdZeZeZfddZ Z S) RetroicoraEPerforms Retrospective Image Correction for physiological motion effects, using a slightly modified version of the RETROICOR algorithm The durations of the physiological inputs are assumed to equal the duration of the dataset. Any constant sampling rate may be used, but 40 Hz seems to be acceptable. This program's cardiac peak detection algorithm is rather simplistic, so you might try using the scanner's cardiac gating output (transform it to a spike wave if necessary). This program uses slice timing information embedded in the dataset to estimate the proper cardiac/respiratory phase for each slice. It makes sense to run this program before any program that may destroy the slice timings (e.g. 3dvolreg for motion correction). For complete details, see the `3dretroicor Documentation. `_ Examples -------- >>> from nipype.interfaces import afni >>> ret = afni.Retroicor() >>> ret.inputs.in_file = 'functional.nii' >>> ret.inputs.card = 'mask.1D' >>> ret.inputs.resp = 'resp.1D' >>> ret.inputs.outputtype = 'NIFTI' >>> ret.cmdline '3dretroicor -prefix functional_retroicor.nii -resp resp.1D -card mask.1D functional.nii' >>> res = ret.run() # doctest: +SKIP Z 3dretroicorcs<|dkr(t|jj r(t|jj r(dStt|j|||S)Nr;)r rWr.r/rr2r&)r\rr'r)rr,r-r&szRetroicor._format_arg) r!r"r#r$r_r(r`rrTr&rr,r,)rr-r2ws !r2c@seZdZeddd#ddddZejejdedddd d$dd Zejd d dddZ ej dddZ e dddZ ej dddZe dddZe dddZe dddZej dddZejd d!dZd"S)% SegInputSpeczANAT is the volume to segmentz-anat %srT)rrrkr/rr0ZAUTO)rzonly non-zero voxels in mask are analyzed. mask can either be a dataset or the string "AUTO" which would use AFNI's automask function to create the mask.z-mask %sr)rrrkr/ZBFTZBIMz -blur_meth %sz1set the blurring method for bias field estimation)rrzPThe amount of blurring used when estimating the field bias with the Wells methodz -bias_fwhm %f)rrzthe prefix for the output folder containing all output volumesz -prefix %sz}MIXFRAC sets up the volume-wide (within mask) tissue fractions while initializing the segmentation (see IGNORE for exception)z -mixfrac %sz5Set the minimum value for any class's mixing fractionz -mixfloor %fz Number of iterations to perform.z -main_N %dNrSr)r!r"r#r r;r r=r?r%Z blur_methr&Z bias_fwhmrclassesZbmrfZ bias_classesprefixZmixfracZmixfloorr'Zmain_Nr,r,r,r-r3sRr3c@s&eZdZdZdZeZeZdddZ dS)Segav3dSeg segments brain volumes into tissue classes. The program allows for adding a variety of global and voxelwise priors. However for the moment, only mixing fractions and MRF are documented. For complete details, see the `3dSeg Documentation. `_ Examples -------- >>> from nipype.interfaces.afni import preprocess >>> seg = preprocess.Seg() >>> seg.inputs.in_file = 'structural.nii' >>> seg.inputs.mask = 'AUTO' >>> seg.cmdline '3dSeg -mask AUTO -anat structural.nii' >>> res = seg.run() # doctest: +SKIP Z3dSegNcCs^ddl}|j}t|jjr6tjjtj|jjd}ntjjtjdd}|j|d|_ |S)NrzClasses+*.BRIKZSegsy) globrr rWr5rrrXrrl)r\rrr7r]rr,r,r-rs zSeg.aggregate_outputs)NN) r!r"r#r$r_r3r`rrTrr,r,r,r-r6s r6c@s0eZdZedddddddZeddd d d Zd S) SkullStripInputSpeczinput file to 3dSkullStripz -input %srTF)rrrkr/rr0z %s_skullstripzoutput image file namez -prefix %sr;)rcrrrdN)r!r"r#r r;rlr,r,r,r-r8 sr8cs0eZdZdZdZdZeZeZ fddZ Z S) SkullStriparA program to extract the brain from surrounding tissue from MRI T1-weighted images. TODO Add optional arguments. For complete details, see the `3dSkullStrip Documentation. `_ Examples -------- >>> from nipype.interfaces import afni >>> skullstrip = afni.SkullStrip() >>> skullstrip.inputs.in_file = 'functional.nii' >>> skullstrip.inputs.args = '-o_ply' >>> skullstrip.cmdline '3dSkullStrip -input functional.nii -o_ply -prefix functional_skullstrip' >>> res = skullstrip.run() # doctest: +SKIP r6Tc s:tt|jf|ts6tj}|dkr6|dkr6d|_dS)Nrrr+F)r:rr)r:rr+)rr9rrrrr)r\rWr$)rr,r-r3 s zSkullStrip.__init__) r!r"r#r$r_rr8r`rrTrrr,r,)rr-r9 s r9c@seZdZedddddddZedddddd Zed d d d ddZejdddddgddZ ejdddddgddZ ejdddddgddZ ejdddddgddZ dS)TCorr1DInputSpecz3d+time dataset inputz %srTF)rrrkr/rr0z1D time series file inputr)rrrkr/rzoutput filename prefixz%s_correlation.nii.gzz -prefix %sxset)rrcrrdrz9Correlation is the normal Pearson correlation coefficientz -pearsonr rktaub)rrrfrkz:Correlation is the Spearman (rank) correlation coefficientz -spearmanpearsonz3Correlation is the quadrant correlation coefficientz -quadrantz:Correlation is the Kendall's tau_b correlation coefficientz -ktaubNrrS) r!r"r#r r<Zy_1drlr r)r>r rr=r,r,r,r-r;> sNr;c@seZdZedddZdS)TCorr1DOutputSpecz#output file containing correlationsT)rrN)r!r"r#r rlr,r,r,r-r?o sr?c@seZdZdZdZeZeZdS)TCorr1Da=Computes the correlation coefficient between each voxel time series in the input 3D+time dataset. For complete details, see the `3dTcorr1D Documentation. `_ >>> from nipype.interfaces import afni >>> tcorr1D = afni.TCorr1D() >>> tcorr1D.inputs.xset= 'u_rc1s1_Template.nii' >>> tcorr1D.inputs.y_1d = 'seed.1D' >>> tcorr1D.cmdline '3dTcorr1D -prefix u_rc1s1_Template_correlation.nii.gz u_rc1s1_Template.nii seed.1D' >>> res = tcorr1D.run() # doctest: +SKIP Z 3dTcorr1DN) r!r"r#r$r_r;r`r?rTr,r,r,r-r@s sr@c@speZdZedddddZeddddZeddd Zejd d Z ej d d Z ej ej ej fd d Zeddd Zej dd Zej dddZeddddZeddddZeddddZeddddZd7Zejej Zed d!ded"Zed#d$ded"Zed%d&ded"Zed'dd(Zed)dd(Zd8ZeZ ed-d.ded"Z!ed/d0ded"Z"ed1d2ded"Z#ej Z$edd3d4d5Z%d6S)9TCorrMapInputSpecTz -input %sF)rrr/r0z-seed %s seeds_width)rrrfz-mask %s)rrz -automask)rz -polort %dz -bpass %f %fz-ort %sz -Gblur %fz -Mseed %fseeds)rrfz-Mean %sZ_meanr;)rrCrdz -Zmean %sZ_zmeanz -Qmean %sZ_qmeanz -Pmean %sZ_pmeanabsolute_thresholdvar_absolute_threshold var_absolute_threshold_normalizez -Thresh %f %sZ_thresh)rrCrdrfz-VarThresh %f %f %f %sZ _varthreshz-VarThreshN %f %f %f %sZ _varthreshnz -CorrMap %s)rrdz -CorrMask %s average_expraverage_expr_nonzerosum_exprz -Aexpr %s %sZ_aexprz -Cexpr %s %sZ_cexprz -Sexpr %s %sZ_sexprz -Hist %d %sZ_hist)rdrrCN)rDrErF)rGrHrI)&r!r"r#r r;rCr%r r)r+r'r(Tupler&bandpassZregress_out_timeseriesZ blur_fwhmrB mean_filezmeanqmeanpmean _thresh_optsr thresholdsrDrErFcorrelation_mapscorrelation_maps_masked _expr_optsrexprrGrHrIhistogram_bin_numbers histogramr,r,r,r-rA sX       rAc@sZeZdZeZeZeZeZeZeZ eZ eZ eZ eZ eZeZeZdS)TCorrMapOutputSpecN)r!r"r#r rLrMrNrOrDrErFrRrSrGrHrIrWr,r,r,r-rX srXcs2eZdZdZdZeZeZdgZ fddZ Z S)TCorrMapaFor each voxel time series, computes the correlation between it and all other voxels, and combines this set of values into the output dataset(s) in some way. For complete details, see the `3dTcorrMap Documentation. `_ Examples -------- >>> from nipype.interfaces import afni >>> tcm = afni.TCorrMap() >>> tcm.inputs.in_file = 'functional.nii' >>> tcm.inputs.mask = 'mask.nii' >>> tcm.mean_file = 'functional_meancorr.nii' >>> tcm.cmdline # doctest: +SKIP '3dTcorrMap -input functional.nii -mask mask.nii -Mean functional_meancorr.nii' >>> res = tcm.run() # doctest: +SKIP Z 3dTcorrMaprCcsp||jjkr |j|jj|gS||jjkr>|j|jj|fS|dkrX|j|jj|fStt|j |||SdS)NrW) rWrPrrQrTrUrVrrYr&)r\rr'r)rr,r-r& s  zTCorrMap._format_arg) r!r"r#r$r_rAr`rXrTZ_additional_metadatar&rr,r,)rr-rY s rYc@seZdZedddddZedddddZedddd Zed dd d Zej d d dZ ej dddZ ej dddZ ej dddZ ej dddZej dddZej dddZej dddZej dddZej dd dZej d!d"dZej d#d$dZed%d&d'd(d)d*Zd+S),NetCorrInputSpecz$input time series file (4D data set)Tz -inset %s)rrrr/z7input set of ROIs, each labelled with distinct integersz -in_rois %szncan include a whole brain mask within which to calculate correlation. Otherwise, data should be masked alreadyz-mask %s)rrrainput a 1D file WTS of weights that will be applied multiplicatively to each ROI's average time series. WTS can be a column- or row-file of values, but it must have the same length as the input time series volume. If the initial average time series was A[n] for n=0,..,(N-1) time points, then applying a set of weights W[n] of the same length from WTS would produce a new time series: B[n] = A[n] * W[n]z -weight_ts %sa switch to also output a matrix of Fisher Z-transform values for the corr coefs (r): Z = atanh(r) , (with Z=4 being output along matrix diagonals where r=1, as the r-to-Z conversion is ceilinged at Z = atanh(r=0.999329) = 4, which is still *quite* a high Pearson-r valuez-fish_z)rrz%output the partial correlation matrixz -part_corrzswitch to output the mean time series of the ROIs that have been used to generate the correlation matrices. Output filenames mirror those of the correlation matrix files, with a '.netts' postfixz-ts_outa5additional switch when using '-ts_out'. Using this option will insert the integer ROI label at the start of each line of the *.netts file created. Thus, for a time series of length N, each line will have N+1 numbers, where the first is the integer ROI label and the subsequent N are scientific notation valuesz -ts_labelasswitch to create a directory for each network that contains the average time series for each ROI in individual files (each file has one line). The directories are labelled PREFIX_000_INDIV/, PREFIX_001_INDIV/, etc. (one per network). Within each directory, the files are labelled ROI_001.netts, ROI_002.netts, etc., with the numbers given by the actual ROI integer labelsz -ts_indivaswitch to create a set of whole brain correlation maps. Performs whole brain correlation for each ROI's average time series; this will automatically create a directory for each network that contains the set of whole brain correlation maps (Pearson 'r's). The directories are labelled as above for '-ts_indiv' Within each directory, the files are labelled WB_CORR_ROI_001+orig, WB_CORR_ROI_002+orig, etc., with the numbers given by the actual ROI integer labelsz -ts_wb_corrasame as above in '-ts_wb_corr', except that the maps have been Fisher transformed to Z-scores the relation: Z=atanh(r). To avoid infinities in the transform, Pearson values are effectively capped at |r| = 0.999329 (where |Z| = 4.0). Files are labelled WB_Z_ROI_001+orig, etcz-ts_wb_Za by default, '-ts_wb_{corr,Z}' output files are named using the int number of a given ROI, such as: WB_Z_ROI_001+orig. With this option, one can replace the int (such as '001') with the string label (such as 'L-thalamus') *if* one has a labeltable attached to the filez-ts_wb_strlabelz|output any correlation map files as NIFTI files (default is BRIK/HEAD). Only useful if using '-ts_wb_corr' and/or '-ts_wb_Z'z-niftizinternally, this program checks for where there are nonnull time series, because we don't like those, in general. With this flag, the user can output the determined mask of non-null time series.z-output_mask_nonnullaby default, this program will grind to a halt and refuse to calculate if any ROI contains >10 percent of voxels with null times series (i.e., each point is 0), as of April, 2017. This is because it seems most likely that hidden badness is responsible. However, if the user still wants to carry on the calculation anyways, then this option will allow one to push on through. However, if any ROI *only* has null time series, then the program will not calculate and the user will really, really, really need to address their maskingz-push_thru_many_zerosz`switch to ignore any label table labels in the '-in_rois' file, if there are any labels attachedz -ignore_LTzoutput file name partz %s_netcorrz -prefix %srr;)rrcrrkrdN)r!r"r#r r;Zin_roisr%Z weight_tsr r)Zfish_zZ part_corrZts_outZts_labelZts_indiv ts_wb_corrZts_wb_ZZts_wb_strlabelZniftiZoutput_mask_nonnullZpush_thru_many_zerosZ ignore_LTrlr,r,r,r-rZ sv    rZc@s&eZdZeddZejeddZdS)NetCorrOutputSpeczPoutput correlation matrix between ROIs written to a text file with .netcc suffix)rz2output correlation maps in Pearson and/or Z-scoresN)r!r"r#r out_corr_matrixr r out_corr_mapsr,r,r,r-r\ sr\c@s$eZdZdZdZeZeZddZ dS)NetCorracCalculate correlation matrix of a set of ROIs (using mean time series of each). Several networks may be analyzed simultaneously, one per brick. For complete details, see the `3dNetCorr Documentation `_. Examples -------- >>> from nipype.interfaces import afni >>> ncorr = afni.NetCorr() >>> ncorr.inputs.in_file = 'functional.nii' >>> ncorr.inputs.mask = 'mask.nii' >>> ncorr.inputs.in_rois = 'maps.nii' >>> ncorr.inputs.ts_wb_corr = True >>> ncorr.inputs.ts_wb_Z = True >>> ncorr.inputs.fish_z = True >>> ncorr.inputs.out_file = 'sub0.tp1.ncorr' >>> ncorr.cmdline '3dNetCorr -prefix sub0.tp1.ncorr -fish_z -inset functional.nii -in_rois maps.nii -mask mask.nii -ts_wb_Z -ts_wb_corr' >>> res = ncorr.run() # doctest: +SKIP Z 3dNetCorrcCsddl}|jj}t|jjs4|j|jjdd}n|jj}tj j tj j |}|jtj j |dd|d<t|jj st|jjrtj j ||d}|jtj j |d|d<|S) NrZ_netcorr)rCz*.netccr]Z _000_INDIVz*.nii.gzr^)r7rTrUr rWrlrVr;rrdirnamerrXr[Z ts_Z_corr)r\r7r]r5ZodirZcorrdirr,r,r-r^ s  zNetCorr._list_outputsN) r!r"r#r$r_rZr`r\rTr^r,r,r,r-r_ s r_c@s`eZdZedddddddZedddddddZed d d d d ZejdddZ ej dddZ dS)TCorrelateInputSpecz input xsetz%srTF)rrrkr/rr0z input ysetrz%s_tcorrzoutput image file namez -prefix %sr<)rcrrrdz9Correlation is the normal Pearson correlation coefficientz-pearson)rrz#Remove polynomical trend of order mz -polort %dNrrS) r!r"r#r r<Zysetrlr r)r>r'r(r,r,r,r-ra s.rac@seZdZdZdZeZeZdS) TCorrelateaEComputes the correlation coefficient between corresponding voxel time series in two input 3D+time datasets 'xset' and 'yset' For complete details, see the `3dTcorrelate Documentation. `_ Examples -------- >>> from nipype.interfaces import afni >>> tcorrelate = afni.TCorrelate() >>> tcorrelate.inputs.xset= 'u_rc1s1_Template.nii' >>> tcorrelate.inputs.yset = 'u_rc1s2_Template.nii' >>> tcorrelate.inputs.out_file = 'functional_tcorrelate.nii.gz' >>> tcorrelate.inputs.polort = -1 >>> tcorrelate.inputs.pearson = True >>> tcorrelate.cmdline '3dTcorrelate -prefix functional_tcorrelate.nii.gz -pearson -polort -1 u_rc1s1_Template.nii u_rc1s2_Template.nii' >>> res = tcarrelate.run() # doctest: +SKIP Z 3dTcorrelateN) r!r"r#r$r_rar`rrTr,r,r,r-rb srbc@seZdZedddddddZeddd d d Zejd d dZejdddZ ejdddZ ejdddZ ej dddZ ejdddZdS)TNormInputSpeczinput file to 3dTNormz%srTF)rrrkr/rr0z%s_tnormzoutput image file namez -prefix %sr;)rcrrrdz+L2 normalize (sum of squares = 1) [DEFAULT]z-norm2)rrzHnormalize so sum of squares = number of time points \* e.g., so RMS = 1.z-normRz)L1 normalize (sum of absolute values = 1)z-norm1z1Scale so max absolute value = 1 (L_infinity norm)z-normxzDetrend with polynomials of order p before normalizing [DEFAULT = don't do this]. Use '-polort 0' to remove the mean, for examplez -polort %szZDetrend with L1 regression (L2 is the default) This option is here just for the hell of itz-L1fitNrS)r!r"r#r r;rlr r)Znorm2ZnormRZnorm1Znormxr'r(ZL1fitr,r,r,r-rc s6   rcc@seZdZdZdZeZeZdS)TNormajShifts voxel time series from input so that separate slices are aligned to the same temporal origin. For complete details, see the `3dTnorm Documentation. `_ Examples -------- >>> from nipype.interfaces import afni >>> tnorm = afni.TNorm() >>> tnorm.inputs.in_file = 'functional.nii' >>> tnorm.inputs.norm2 = True >>> tnorm.inputs.out_file = 'rm.errts.unit errts+tlrc' >>> tnorm.cmdline '3dTnorm -norm2 -prefix rm.errts.unit errts+tlrc functional.nii' >>> res = tshift.run() # doctest: +SKIP Z3dTnormN) r!r"r#r$r_rcr`rrTr,r,r,r-rd& srdc@s,eZdZedddddddZeddd6d d d Zed d ddZejej dddZ ej ddddddZ eddddZ ejdddZeddddZejdddZeeddd d!d"d#Zejejejd$d%dZejejejd&d'dZejd(d)dZedd*d+d,Zejd-d.gd/d0Zejd1d2dZejd3d4dZd5S)7TProjectInputSpeczinput file to 3dTprojectz -input %srTF)rrrkr/rr0z %s_tprojectzoutput image file namez -prefix %sr;)rcrrkrrdzFilename of censor .1D time series. This is a file of 1s and 0s, indicating which time points are to be included (1) and which are to be excluded (0).z -censor %s)rrrazList of strings that specify time indexes to be removed from the analysis. Each string is of one of the following forms: * ``37`` => remove global time index #37 * ``2:37`` => remove time index #37 in run #2 * ``37..47`` => remove global time indexes #37-47 * ``37-47`` => same as above * ``2:37..47`` => remove time indexes #37-47 in run #2 * ``*:0-2`` => remove time indexes #0-2 in all runs * Time indexes within each run start at 0. * Run indexes start at 1 (just be to confusing). * N.B.: 2:37,47 means index #37 in run #2 and global time index 47; it does NOT mean index #37 in run #2 AND index #47 in run #2. z -CENSORTR %s)rrZKILLZZEROZNTRPa Specifies how censored time points are treated in the output dataset: * mode = ZERO -- put zero values in their place; output datset is same length as input * mode = KILL -- remove those time points; output dataset is shorter than input * mode = NTRP -- censored values are replaced by interpolated neighboring (in time) non-censored values, BEFORE any projections, and then the analysis proceeds without actual removal of any time points -- this feature is to keep the Spanish Inquisition happy. * The default mode is KILL !!! z -cenmode %saThe catenation file, as in 3dDeconvolve, containing the TR indexes of the start points for each contiguous run within the input dataset (the first entry should be 0). * Also as in 3dDeconvolve, if the input dataset is automatically catenated from a collection of datasets, then the run start indexes are determined directly, and '-concat' is not needed (and will be ignored). * Each run must have at least 9 time points AFTER censoring, or the program will not work! * The only use made of this input is in setting up the bandpass/stopband regressors. * '-ort' and '-dsort' regressors run through all time points, as read in. If you want separate projections in each run, then you must either break these ort files into appropriate components, OR you must run 3dTproject for each run separately, using the appropriate pieces from the ort files via the ``{...}`` selector for the 1D files and the ``[...]`` selector for the datasets. z -concat %s)rrrzAlso as in 3dDeconvolve, if you want the program to treat an auto-catenated dataset as one long run, use this option. However, '-noblock' will not affect catenation if you use the '-concat' option.z-noblockzCRemove each column in file. Each column will have its mean removed.z-ort %saRemove polynomials up to and including degree pp. * Default value is 2. * It makes no sense to use a value of pp greater than 2, if you are bandpassing out the lower frequencies! * For catenated datasets, each run gets a separate set set of pp+1 Legendre polynomial regressors. * Use of -polort -1 is not advised (if data mean != 0), even if -ort contains constant terms, as all means are removed. z -polort %d)rr0z -dsort %s...zRemove the 3D+time time series in dataset fset. * That is, 'fset' contains a different nuisance time series for each voxel (e.g., from AnatICOR). * Multiple -dsort options are allowed. )rrz0Remove all frequencies EXCEPT those in the rangez-bandpass %g %gz#Remove all frequencies in the rangez-stopband %g %gzdUse time step dd for the frequency calculations, rather than the value stored in the dataset header.z-TR %gzOnly operate on voxels nonzero in the mset dataset. * Voxels outside the mask will be filled with zeros. * If no masking option is given, then all voxels will be processed. z-mask %s)rrrzGenerate a mask automaticallyr%z -automask)rrfrzBlur (inside the mask only) with a filter that has width (FWHM) of fff millimeters. Spatial blurring (if done) is after the time series filtering.z-blur %gzZ Normalize each output time series to have sum of squares = 1. This is the LAST operation.z-normNrS)r!r"r#r r;rlZcensorr rrZcensortrr?Zcenmodeconcatr)ZnoblockZortr'r(rZdsortrJr&rKZstopbandZTRr%r+rZnormr,r,r,r-re? s  rec@seZdZdZdZeZeZdS)TProjecta This program projects (detrends) out various 'nuisance' time series from each voxel in the input dataset. Note that all the projections are done via linear regression, including the frequency-based options such as ``-passband``. In this way, you can bandpass time-censored data, and at the same time, remove other time series of no interest (e.g., physiological estimates, motion parameters). Shifts voxel time series from input so that seperate slices are aligned to the same temporal origin. Examples -------- >>> from nipype.interfaces import afni >>> tproject = afni.TProject() >>> tproject.inputs.in_file = 'functional.nii' >>> tproject.inputs.bandpass = (0.00667, 99999) >>> tproject.inputs.polort = 3 >>> tproject.inputs.automask = True >>> tproject.inputs.out_file = 'projected.nii.gz' >>> tproject.cmdline '3dTproject -input functional.nii -automask -bandpass 0.00667 99999 -polort 3 -prefix projected.nii.gz' >>> res = tproject.run() # doctest: +SKIP See Also -------- For complete details, see the `3dTproject Documentation. `__ Z 3dTprojectN) r!r"r#r$r_rer`rrTr,r,r,r-rg srgc @seZdZeddd9ddddZeddd d d Zed d dZej dddgdZ ej dddgdZ ej dddZ ejd:dddZejejdd d!d"d#d$d%d&d'd( ed)d*d+gdZejedd,ejej d-d.d/gdZejd0d1dd2d3Zejd4d5dZejd6d7dZd8S);TShiftInputSpeczinput file to 3dTshiftz%srTF)rrrkr/rr0z %s_tshiftzoutput image file namez -prefix %sr;)rcrrrdzTmanually set the TR. You can attach suffix "s" for seconds or "ms" for milliseconds.z-TR %s)rrz%align each slice to given time offsetz -tzero %stslice)rrrfz.align each slice to time offset of given slicez -slice %stzeroz(ignore the first set of points specifiedz -ignore %srrqrrrsrwzLdifferent interpolation methods (see 3dTshift for details) default = Fourierz-%szalt+zZaltpluszalt+z2zalt-zZaltminuszalt-z2zseq+zZseqpluszseq-zZseqminusz:use specified slice time pattern rather than one in headerz -tpattern %s slice_timing)rz=time offsets from the volume acquisition onset for each slicez -tpattern @%stpatternkzk-aBDirection in which slice_timing is specified (default: k). If negative,slice_timing is defined in reverse order, that is, the first entry corresponds to the slice with the largest index, and the final entry corresponds to slice index zero. Only in effect when slice_timing is passed as list, not when it is passed as file.)r5rz1Before shifting, remove the mean and linear trendz-rltzMBefore shifting, remove the mean and linear trend and later put back the meanz-rlt+NrS)rrqrrrsrw)r!r"r#r r;rlrrr r&rjr'riignorer?interpr=rlrrkslice_encoding_directionr)ZrltZrltplusr,r,r,r-rh sv      rhc@seZdZeddZdS)TShiftOutputSpecz9AFNI formatted timing file, if ``slice_timing`` is a list)rN)r!r"r#r timing_filer,r,r,r-rqr srqcs@eZdZdZdZeZeZfddZ ddZ fddZ Z S) TShiftaW Shifts voxel time series from input so that seperate slices are aligned to the same temporal origin. For complete details, see the `3dTshift Documentation. `_ Examples -------- Slice timing details may be specified explicitly via the ``slice_timing`` input: >>> from nipype.interfaces import afni >>> TR = 2.5 >>> tshift = afni.TShift() >>> tshift.inputs.in_file = 'functional.nii' >>> tshift.inputs.tzero = 0.0 >>> tshift.inputs.tr = '%.1fs' % TR >>> tshift.inputs.slice_timing = list(np.arange(40) / TR) >>> tshift.cmdline '3dTshift -prefix functional_tshift -tpattern @slice_timing.1D -TR 2.5s -tzero 0.0 functional.nii' When the ``slice_timing`` input is used, the ``timing_file`` output is populated, in this case with the generated file. >>> tshift._list_outputs()['timing_file'] # doctest: +ELLIPSIS '.../slice_timing.1D' >>> np.loadtxt(tshift._list_outputs()['timing_file']).tolist()[:5] [0.0, 0.4, 0.8, 1.2, 1.6] If ``slice_encoding_direction`` is set to ``'k-'``, the slice timing is reversed: >>> tshift.inputs.slice_encoding_direction = 'k-' >>> tshift.cmdline '3dTshift -prefix functional_tshift -tpattern @slice_timing.1D -TR 2.5s -tzero 0.0 functional.nii' >>> np.loadtxt(tshift._list_outputs()['timing_file']).tolist()[:5] [15.6, 15.2, 14.8, 14.4, 14.0] This method creates a ``slice_timing.1D`` file to be passed to ``3dTshift``. A pre-existing slice-timing file may be used in the same way: >>> tshift = afni.TShift() >>> tshift.inputs.in_file = 'functional.nii' >>> tshift.inputs.tzero = 0.0 >>> tshift.inputs.tr = '%.1fs' % TR >>> tshift.inputs.slice_timing = 'slice_timing.1D' >>> tshift.cmdline '3dTshift -prefix functional_tshift -tpattern @slice_timing.1D -TR 2.5s -tzero 0.0 functional.nii' When a pre-existing file is provided, ``timing_file`` is simply passed through. >>> tshift._list_outputs()['timing_file'] # doctest: +ELLIPSIS '.../slice_timing.1D' Alternatively, pre-specified slice timing patterns may be specified with the ``tpattern`` input. For example, to specify an alternating, ascending slice timing pattern: >>> tshift = afni.TShift() >>> tshift.inputs.in_file = 'functional.nii' >>> tshift.inputs.tzero = 0.0 >>> tshift.inputs.tr = '%.1fs' % TR >>> tshift.inputs.tpattern = 'alt+z' >>> tshift.cmdline '3dTshift -prefix functional_tshift -tpattern alt+z -TR 2.5s -tzero 0.0 functional.nii' For backwards compatibility, ``tpattern`` may also take filenames prefixed with ``@``. However, in this case, filenames are not validated, so this usage will be deprecated in future versions of Nipype. >>> tshift = afni.TShift() >>> tshift.inputs.in_file = 'functional.nii' >>> tshift.inputs.tzero = 0.0 >>> tshift.inputs.tr = '%.1fs' % TR >>> tshift.inputs.tpattern = '@slice_timing.1D' >>> tshift.cmdline '3dTshift -prefix functional_tshift -tpattern @slice_timing.1D -TR 2.5s -tzero 0.0 functional.nii' In these cases, ``timing_file`` is undefined. >>> tshift._list_outputs()['timing_file'] # doctest: +ELLIPSIS In any configuration, the interface may be run as usual: >>> res = tshift.run() # doctest: +SKIP Z3dTshiftcsL|dkr|jdrtjdn|dkr8t|tr8|j}tt|j|||S)Nrl@zVPassing a file prefixed by "@" will be deprecated; please use the `slice_timing` inputrk) startswithifloggerwarning isinstancer_write_slice_timingrrsr&)r\rr'r)rr,r-r& s zTShift._format_argc CsVt|jj}|jjjdr"|jd}t|d}|jdjt t |WdQRX|S)N-zslice_timing.1Dr ) rrWrkrpendswithreverserrrXmapstr)r\rkfnameZfobjr,r,r-ry s   zTShift._write_slice_timingcsRtt|j}t|jjrNt|jjtr:tj j d|d<ntj j |jj|d<|S)Nzslice_timing.1Drr) rrsr^r rWrkrxrrrr)r\r])rr,r-r^ s  zTShift._list_outputs) r!r"r#r$r_rhr`rqrTr&ryr^rr,r,)rr-rsv sX  rsc@seZdZeddd!ddddZeddd d d Zejd d dZej dddZ ej dddZ ej dddZ ej dddZej dddZej dddZedddZej dddZd S)"TSmoothInputSpeczinput file to 3dTSmoothz%srTF)rrrkr/rr0z %s_smoothzoutput file from 3dTSmoothz -prefix %sr;)rcrrrdz(Sets the data type of the output datasetz -datum %s)rrzY3 point linear filter: :math:`0.15\,a + 0.70\,b + 0.15\,c` [This is the default smoother]z-linz$3 point median filter: median(a,b,c)z-medza3 point order statistics filter::math:`0.15\,min(a,b,c) + 0.70\,median(a,b,c) + 0.15\,max(a,b,c)`z-osfzt3 point linear filter: :math:`0.5\,(1-m)\,a + m\,b + 0.5\,(1-m)\,c`. Here, 'm' is a number strictly between 0 and 1.z-3lin %dz -hamming %dz?Use N point Hamming windows. (N must be odd and bigger than 1.))rrz -blackman %dz@Use N point Blackman windows. (N must be odd and bigger than 1.)z -custom %sz>odd # of coefficients must be in a single column in ASCII filez -adaptive %dzOuse adaptive mean filtering of width N (where N must be odd and bigger than 3).NrS)r!r"r#r r;rlr rrr)ZlinZmedZosfr'Zlin3ZhammingZblackmanZcustomZadaptiver,r,r,r-r sH rc@seZdZdZdZeZeZdS)TSmoothaSmooths each voxel time series in a 3D+time dataset and produces as output a new 3D+time dataset (e.g., lowpass filter in time). For complete details, see the `3dTsmooth Documentation. `_ Examples -------- >>> from nipype.interfaces import afni >>> from nipype.testing import example_data >>> smooth = afni.TSmooth() >>> smooth.inputs.in_file = 'functional.nii' >>> smooth.inputs.adaptive = 5 >>> smooth.cmdline '3dTsmooth -adaptive 5 -prefix functional_smooth functional.nii' >>> res = smooth.run() # doctest: +SKIP Z 3dTsmoothN) r!r"r#r$r_rr`rrTr,r,r,r-r' src@seZdZeddd3ddddZejejeddejedddd d Z ed d d ddZ eddd4ddZ ejddd5dZ edddddd6dZ edddddd Zejd!d"d Zejd#d$d Zejd%d&d Zed'd(d)ddd*Zejd7d0d1d Zd2S)8VolregInputSpeczinput file to 3dvolregz%srTF)rrrkr/rr0)rzYweights for each voxel specified by a file with an optional volume number (defaults to 0)z-weight '%s[%d]')rrz %s_volregzoutput image file namez -prefix %sr;)rcrrrdzbase file for registrationz-base %sr*)rrrkrz7Zeropad around the edges by 'n' voxels during rotationsz-zpad %dr))rrrkz%s_md.1Dzmax displacement output filez -maxdisp1D %sr)rcrrrdrrkz%s.1Dz"1D movement parameters output filez -1Dfile %s)rcrrrdrz(more detailed description of the processz-verbosez$time shift to mean slice time offsetz -tshift 0z© base file origin coords to outputz-twodupz %s.aff12.1DzSave the matrix transformationz-1Dmatrix_save %s)rcrrrrdrrrrwrsrqz0spatial interpolation methods [default = heptic]z-%sNrSir-r,)rrrrwrsrq)r!r"r#r r;r r=rJr'in_weight_volumerlZbasefilezpad md1d_filerr)rZ timeshiftZ copyoriginoned_matrix_saver?ror,r,r,r-r@ sh   rc@s<eZdZedddZedddZedddZedddZdS)VolregOutputSpeczregistered fileT)rrzmax displacement info filezmovement parameters info filez(matrix transformation from base to inputN)r!r"r#r rlrrrr,r,r,r-r s    rcs,eZdZdZdZeZeZfddZ Z S)VolregaRegister input volumes to a base volume using AFNI 3dvolreg command For complete details, see the `3dvolreg Documentation. `_ Examples -------- >>> from nipype.interfaces import afni >>> volreg = afni.Volreg() >>> volreg.inputs.in_file = 'functional.nii' >>> volreg.inputs.args = '-Fourier -twopass' >>> volreg.inputs.zpad = 4 >>> volreg.inputs.outputtype = 'NIFTI' >>> volreg.cmdline # doctest: +ELLIPSIS '3dvolreg -Fourier -twopass -1Dfile functional.1D -1Dmatrix_save functional.aff12.1D -prefix functional_volreg.nii -zpad 4 -maxdisp1D functional_md.1D functional.nii' >>> res = volreg.run() # doctest: +SKIP >>> from nipype.interfaces import afni >>> volreg = afni.Volreg() >>> volreg.inputs.in_file = 'functional.nii' >>> volreg.inputs.interp = 'cubic' >>> volreg.inputs.verbose = True >>> volreg.inputs.zpad = 1 >>> volreg.inputs.basefile = 'functional.nii' >>> volreg.inputs.out_file = 'rm.epi.volreg.r1' >>> volreg.inputs.oned_file = 'dfile.r1.1D' >>> volreg.inputs.oned_matrix_save = 'mat.r1.tshift+orig.1D' >>> volreg.cmdline '3dvolreg -cubic -1Dfile dfile.r1.1D -1Dmatrix_save mat.r1.tshift+orig.1D -prefix rm.epi.volreg.r1 -verbose -base functional.nii -zpad 1 -maxdisp1D functional_md.1D functional.nii' >>> res = volreg.run() # doctest: +SKIP Z3dvolregcs0|dkrt|t r|df}tt|j|||S)Nrr)rxtuplerrr&)r\rr'r)rr,r-r& szVolreg._format_arg) r!r"r#r$r_rr`rrTr&rr,r,)rr-r s "rc@seZdZeddd*ddddZeddd d dd Zejd d dZejdddZ eddddZ eddddZ ejdddZ ej d+dddZeddddZejd d!dZejd"d#dZejd$d%dZejd&d'gd(Zd)S), WarpInputSpeczinput file to 3dWarpz%srTF)rrrkr/rr0z%s_warpzoutput image file namez -prefix %sr;)rcrrrdrz*transform dataset from Talairach to MNI152z-tta2mni)rrz*transform dataset from MNI152 to Talaraichz-mni2ttaz%apply transformation from 3dWarpDrivez -matparent %s)rrrzlRead in the oblique transformation matrix from an oblique dataset and make cardinal dataset oblique to matchz-oblique_parent %sz*transform dataset from oblique to cardinalz -deobliquerqrrNNrsz0spatial interpolation methods [default = linear]z-%szcopy grid of specified datasetz -gridset %szspecify grid of this size (mm)z -newgrid %fz5pad input dataset with N planes of zero on all sides.z-zpad %dz)Print out some information along the way.z-verbzsave warp as .mat filer)rrNrS)rqrrrrs)r!r"r#r r;rlr r)Ztta2mniZmni2ttaZ matparentZoblique_parentZ deobliquer?roZgridsetr&rr'rr save_warpr,r,r,r-r sN      rc@s"eZdZedddZeddZdS)WarpOutputSpecz Warped file.T)rrzwarp transform .mat file)rN)r!r"r#r rl warp_filer,r,r,r-r s rcs:eZdZdZdZeZeZd fdd Z fddZ Z S) Warpa]Use 3dWarp for spatially transforming a dataset. Examples -------- >>> from nipype.interfaces import afni >>> warp = afni.Warp() >>> warp.inputs.in_file = 'structural.nii' >>> warp.inputs.deoblique = True >>> warp.inputs.out_file = 'trans.nii.gz' >>> warp.cmdline '3dWarp -deoblique -prefix trans.nii.gz structural.nii' >>> res = warp.run() # doctest: +SKIP >>> warp_2 = afni.Warp() >>> warp_2.inputs.in_file = 'structural.nii' >>> warp_2.inputs.newgrid = 1.0 >>> warp_2.inputs.out_file = 'trans.nii.gz' >>> warp_2.cmdline '3dWarp -newgrid 1.000000 -prefix trans.nii.gz structural.nii' >>> res = warp_2.run() # doctest: +SKIP See Also -------- For complete details, see the `3dWarp Documentation. `__. Z3dWarprcsJtt|j||}|jjrFddl}|jd}|j||jgt dd|S)Nrrz%s)fmt) rrrrWrnumpyr^Zsavetxtrr)r\rr npr)rr,r-rs  zWarp._run_interfacecs0tt|j}|jjr,t|dddd|d<|S)Nrlz_transform.matF)rCuse_extr)rrr^rWrr)r\r])rr,r-r^!s zWarp._list_outputsr)r) r!r"r#r$r_rr`rrTrr^rr,r,)rr-r s  rc @seZdZeddddddZeddddddZedd d gd d Zejd ddZ ejdddZ ej dddgdZ ejdddZ ejdddgdZejdddZejdddZejd d!dZejd"d#d$Zejd%d&dZed'd(dd)Zejejd*d+d,d,d-gd.Zejejejfd/d0ejedd1ejfd2d3d4gdZed5d6d$Zejejd7d8d9d:d;Zejejdd?ddd@ZejdAdBdZejdCdDdZ ejdEdFdZ!ejedddGdHdIdJgdZ"ejdKdLdJgdZ#ejdMdNdZ$ejdKdOdJgddPZ%edQdRdddJdgdSZ&ejdTdUdVdJdgdZ'ejdWdXdYdZd[d\dd]gdZ(ejd^d_d`dagdZ)ejdbdcdZ*ejdddedZ+ejdfdgdJd]dhgdZ,ejdidjdZ-ejdkdld]dmgdZ.ejdndodVgdZ/ejdpdqdrd`gdZ0ejdsdtdrdagdZ1ejdudvdwgdZ2ejdxdydzgdZ3ejd{d|dZ4ejd}d~dddddgddPZ5ejdddddddgdZ6ejdddddddgdZ7ejdddddddgdZ8ejdddddddgdZ9dS)QwarpInputSpeczASource image (opposite phase encoding direction than base image).z -source %sTF)rrr/rr0zABase image (opposite phase encoding direction than source image).z-base %sz -prefix %szppp_%sr;aSets the prefix/suffix for the output datasets. * The source dataset is warped to match the base and gets prefix 'ppp'. (Except if '-plusminus' is used * The final interpolation to this output dataset is done using the 'wsinc5' method. See the output of 3dAllineate -HELP (in the "Modifying '-final wsinc5'" section) for the lengthy technical details. * The 3D warp used is saved in a dataset with prefix 'ppp_WARP' -- this dataset can be used with 3dNwarpApply and 3dNwarpCat, for example. * To be clear, this is the warp from source dataset coordinates to base dataset coordinates, where the values at each base grid point are the xyz displacments needed to move that grid point's xyz values to the corresponding xyz values in the source dataset: base( (x,y,z) + WARP(x,y,z) ) matches source(x,y,z) Another way to think of this warp is that it 'pulls' values back from source space to base space. * 3dNwarpApply would use 'ppp_WARP' to transform datasets aligned with the source dataset to be aligned with the base dataset. **If you do NOT want this warp saved, use the option '-nowarp'**. (However, this warp is usually the most valuable possible output!) * If you want to calculate and save the inverse 3D warp, use the option '-iwarp'. This inverse warp will then be saved in a dataset with prefix 'ppp_WARPINV'. * This inverse warp could be used to transform data from base space to source space, if you need to do such an operation. * You can easily compute the inverse later, say by a command like 3dNwarpCat -prefix Z_WARPINV 'INV(Z_WARP+tlrc)' or the inverse can be computed as needed in 3dNwarpApply, like 3dNwarpApply -nwarp 'INV(Z_WARP+tlrc)' -source Dataset.nii ... )rrcrdraThis option simply resamples the source dataset to match the base dataset grid. You can use this if the two datasets overlap well (as seen in the AFNI GUI), but are not on the same 3D grid. * If they don't overlap well, allineate them first * The reampling here is done with the 'wsinc5' method, which has very little blurring artifact. * If the base and source datasets ARE on the same 3D grid, then the -resample option will be ignored. * You CAN use -resample with these 3dQwarp options: -plusminus -inilev -iniwarp -duplo z -resample)rrzThis option will make 3dQwarp run 3dAllineate first, to align the source dataset to the base with an affine transformation. It will then use that alignment as a starting point for the nonlinear warping.z -allineatezBadd extra options to the 3dAllineate command to be run by 3dQwarp.z-allineate_opts %s allineate)rrrzDo not save the _WARP file.z-nowarpz&Do compute and save the _WARPINV file.z-iwarp plusminus)rrrfzUse strict Pearson correlation for matching.Not usually recommended, since the 'clipped Pearson' methodused by default will reduce the impact of outlier values.z-pearaBReplace negative values in either input volume with 0. * If there ARE negative input values, and you do NOT use -noneg, then strict Pearson correlation will be used, since the 'clipped' method only is implemented for non-negative volumes. * '-noneg' is not the default, since there might be situations where you want to align datasets with positive and negative values mixed. * But, in many cases, the negative values in a dataset are just the result of interpolation artifacts (or other peculiarities), and so they should be ignored. That is what '-noneg' is for. z-nonegaAReplace negative values in either input volume with 0. * If there ARE negative input values, and you do NOT use -noneg, then strict Pearson correlation will be used, since the 'clipped' method only is implemented for non-negative volumes. * '-noneg' is not the default, since there might be situations where you want to align datasets with positive and negative values mixed. * But, in many cases, the negative values in a dataset are just the result of interpolation artifacts (or other peculiarities), and so they should be ignored. That is what '-noneg' is for. z -nopenaltyz -penfac %fa3Use this value to weight the penalty. The default value is 1. Larger values mean the penalty counts more, reducing grid distortions, insha'Allah; '-nopenalty' is the same as '-penfac 0'. In 23 Sep 2013 Zhark increased the default value of the penalty by a factor of 5, and also made it get progressively larger with each level of refinement. Thus, warping results will vary from earlier instances of 3dQwarp. * The progressive increase in the penalty at higher levels means that the 'cost function' can actually look like the alignment is getting worse when the levels change. * IF you wish to turn off this progression, for whatever reason (e.g., to keep compatibility with older results), use the option '-penold'.To be completely compatible with the older 3dQwarp, you'll also have to use '-penfac 0.2'. )rrzIf you want a binary weight (the old default), use this option.That is, each voxel in the base volume automask will beweighted the same in the computation of the cost functional.z -noweightzInstead of computing the weight from the base dataset,directly input the weight volume from dataset 'www'.Useful if you know what over parts of the base image youwant to emphasize or de-emphasize the matching functional.z -weight %s)rrra"``-wball x y z r f`` Enhance automatic weight from '-useweight' by a factor of 1+f\*Gaussian(FWHM=r) centered in the base image at DICOM coordinates (x,y,z) and with radius 'r'. The goal of this option is to try and make the alignment better in a specific part of the brain. Example: -wball 0 14 6 30 40 to emphasize the thalamic area (in MNI/Talairach space). * The 'r' parameter must be positive! * The 'f' parameter must be between 1 and 100 (inclusive). * '-wball' does nothing if you input your own weight with the '-weight' option. * '-wball' does change the binary weight created by the '-noweight' option. * You can only use '-wball' once in a run of 3dQwarp. **The effect of '-wball' is not dramatic.** The example above makes the average brain image across a collection of subjects a little sharper in the thalamic area, which might have some small value. If you care enough about alignment to use '-wball', then you should examine the results from 3dQwarp for each subject, to see if the alignments are good enough for your purposes.z -wball %sr)wmask)rrminlenmaxlenrfz -bpass %f %f)r)raSimilar to '-wball', but here, you provide a dataset 'ws' that indicates where to increase the weight. * The 'ws' dataset must be on the same 3D grid as the base dataset. * 'ws' is treated as a mask -- it only matters where it is nonzero -- otherwise, the values inside are not used. * After 'ws' comes the factor 'f' by which to increase the automatically computed weight. Where 'ws' is nonzero, the weighting will be multiplied by (1+f). * As with '-wball', the factor 'f' should be between 1 and 100. z -wpass %s %fwballz -wtprefix %sz,Write the weight volume to disk as a datasetaGaussian blur the input images by 'bb' (FWHM) voxels before doing the alignment (the output dataset will not be blurred). The default is 2.345 (for no good reason). * Optionally, you can provide 2 values for 'bb', and then the first one is applied to the base volume, the second to the source volume. e.g., '-blur 0 3' to skip blurring the base image (if the base is a blurry template, for example). * A negative blur radius means to use 3D median filtering, rather than Gaussian blurring. This type of filtering will better preserve edges, which can be important in alignment. * If the base is a template volume that is already blurry, you probably don't want to blur it again, but blurring the source volume a little is probably a good idea, to help the program avoid trying to match tiny features. * Note that -duplo will blur the volumes some extra amount for the initial small-scale warping, to make that phase of the program converge more rapidly. z-blur %srr)rrrraUse progressive blurring; that is, for larger patch sizes, the amount of blurring is larger. The general idea is to avoid trying to match finer details when the patch size and incremental warps are coarse. When '-blur' is used as well, it sets a minimum amount of blurring that will be used. [06 Aug 2014 -- '-pblur' may become the default someday]. * You can optionally give the fraction of the patch size that is used for the progressive blur by providing a value between 0 and 0.25 after '-pblur'. If you provide TWO values, the the first fraction is used for progressively blurring the base image and the second for the source image. The default parameters when just '-pblur' is given is the same as giving the options as '-pblur 0.09 0.09'. * '-pblur' is useful when trying to match 2 volumes with high amounts of detail; e.g, warping one subject's brain image to match another's, or trying to warp to match a detailed template. * Note that using negative values with '-blur' means that the progressive blurring will be done with median filters, rather than Gaussian linear blurring. Note: The combination of the -allineate and -pblur options will make the results of using 3dQwarp to align to a template somewhat less sensitive to initial head position and scaling.z -pblur %sa>Here, 'ee' is a dataset to specify a mask of voxelsto EXCLUDE from the analysis -- all voxels in 'ee'that are NONZERO will not be used in the alignment.The base image always automasked -- the emask isextra, to indicate voxels you definitely DON'T wantincluded in the matching process, even if they areinside the brain.z -emask %s)rrrr0z%Warp will not displace in x directionz-noXdisz%Warp will not displace in y directionz-noYdisz%Warp will not displace in z directionz-noZdis)rr0aA dataset with an initial nonlinear warp to use. * If this option is not used, the initial warp is the identity. * You can specify a catenation of warps (in quotes) here, as in program 3dNwarpApply. * As a special case, if you just input an affine matrix in a .1D file, that will work also -- it is treated as giving the initial warp via the string "IDENT(base_dataset) matrix_file.aff12.1D". * You CANNOT use this option with -duplo !! * -iniwarp is usually used with -inilev to re-start 3dQwarp from a previous stopping point. z -iniwarp %sduploaThe initial refinement 'level' at which to start. * Usually used with -iniwarp; CANNOT be used with -duplo. * The combination of -inilev and -iniwarp lets you take the results of a previous 3dQwarp run and refine them further: Note that the source dataset in the second run is the SAME as in the first run. If you don't see why this is necessary, then you probably need to seek help from an AFNI guru. z -inilev %da4The value of mm should be an odd integer. * The default value of mm is 25. * For more accurate results than mm=25, try 19 or 13. * The smallest allowed patch size is 5. * You may want stop at a larger patch size (say 7 or 9) and use the -Qfinal option to run that final level with quintic warps, which might run faster and provide the same degree of warp detail. * Trying to make two different brain volumes match in fine detail is usually a waste of time, especially in humans. There is too much variability in anatomy to match gyrus to gyrus accurately. For this reason, the default minimum patch size is 25 voxels. Using a smaller '-minpatch' might try to force the warp to match features that do not match, and the result can be useless image distortions -- another reason to LOOK AT THE RESULTS. z -minpatch %dz -maxlev %d)rrrfrkaThis option provides an alternate way to specify the patch grid sizes used in the warp optimization process. 'gl' is a 1D file with a list of patches to use -- in most cases, you will want to use it in the following form: ``-gridlist '1D: 0 151 101 75 51'`` * Here, a 0 patch size means the global domain. Patch sizes otherwise should be odd integers >= 5. * If you use the '0' patch size again after the first position, you will actually get an iteration at the size of the default patch level 1, where the patch sizes are 75% of the volume dimension. There is no way to force the program to literally repeat the sui generis step of lev=0. z -gridlist %s)rrrr0rfz This option lets you save the output warps from each level" of the refinement process. Mostly used for experimenting." Will only save all the outputs if the program terminates" normally -- if it crashes, or freezes, then all these" warps are lost.z-allsave nopadWARPaStart off with 1/2 scale versions of the volumes," for getting a speedy coarse first alignment." * Then scales back up to register the full volumes." The goal is greater speed, and it seems to help this" positively piggish program to be more expeditious." * However, accuracy is somewhat lower with '-duplo'," for reasons that currenly elude Zhark; for this reason," the Emperor does not usually use '-duplo'. z-duplogridlistmaxlevinileviniwarpallsavea Iterate more times, which can help when the volumes are hard to align at all, or when you hope to get a more precise alignment. * Slows the program down (possibly a lot), of course. * When you combine '-workhard' with '-duplo', only the full size volumes get the extra iterations. * For finer control over which refinement levels work hard, you can use this option in the form (for example) ``-workhard:4:7`` which implies the extra iterations will be done at levels 4, 5, 6, and 7, but not otherwise. * You can also use '-superhard' to iterate even more, but this extra option will REALLY slow things down. * Under most circumstances, you should not need to use either ``-workhard`` or ``-superhard``. * The fastest way to register to a template image is via the ``-duplo`` option, and without the ``-workhard`` or ``-superhard`` options. * If you use this option in the form '-Workhard' (first letter in upper case), then the second iteration at each level is done with quintic polynomial warps. z -workhardZboxoptballoptaAt the finest patch size (the final level), use Hermite quintic polynomials for the warp instead of cubic polynomials. * In a 3D 'patch', there are 2x2x2x3=24 cubic polynomial basis function parameters over which to optimize (2 polynomials dependent on each of the x,y,z directions, and 3 different directions of displacement). * There are 3x3x3x3=81 quintic polynomial parameters per patch. * With -Qfinal, the final level will have more detail in the allowed warps, at the cost of yet more CPU time. * However, no patch below 7x7x7 in size will be done with quintic polynomials. * This option is also not usually needed, and is experimental. z-QfinalzUse Hermite quintic polynomials at all levels. * Very slow (about 4 times longer). Also experimental. * Will produce a (discrete representation of a) C2 warp. z-QonlyaNormally, the warp displacements dis(x) are defined to match base(x) to source(x+dis(x)). With this option, the match is between base(x-dis(x)) and source(x+dis(x)) -- the two images 'meet in the middle'. * One goal is to mimic the warping done to MRI EPI data by field inhomogeneities, when registering between a 'blip up' and a 'blip down' down volume, which will have opposite distortions. * Define Wp(x) = x+dis(x) and Wm(x) = x-dis(x). Then since base(Wm(x)) matches source(Wp(x)), by substituting INV(Wm(x)) wherever we see x, we have base(x) matches source(Wp(INV(Wm(x)))); that is, the warp V(x) that one would get from the 'usual' way of running 3dQwarp is V(x) = Wp(INV(Wm(x))). * Conversely, we can calculate Wp(x) in terms of V(x) as follows: If V(x) = x + dv(x), define Vh(x) = x + dv(x)/2; then Wp(x) = V(INV(Vh(x))) * With the above formulas, it is possible to compute Wp(x) from V(x) and vice-versa, using program 3dNwarpCalc. The requisite commands are left as an exercise for the aspiring AFNI Jedi Master. * You can use the semi-secret '-pmBASE' option to get the V(x) warp and the source dataset warped to base space, in addition to the Wp(x) '_PLUS' and Wm(x) '_MINUS' warps. * Alas: -plusminus does not work with -duplo or -allineate :-( * However, you can use -iniwarp with -plusminus :-) * The outputs have _PLUS (from the source dataset) and _MINUS (from the base dataset) in their filenames, in addition to the prefix. The -iwarp option, if present, will be ignored. z -plusminusiwarpa\Do NOT use zero-padding on the 3D base and source images. [Default == zero-pad, if needed] * The underlying model for deformations goes to zero at the edge of the volume being warped. However, if there is significant data near an edge of the volume, then it won't get displaced much, and so the results might not be good. * Zero padding is designed as a way to work around this potential problem. You should NOT need the '-nopad' option for any reason that Zhark can think of, but it is here to be symmetrical with 3dAllineate. * Note that the output (warped from source) dataset will be on the base dataset grid whether or not zero-padding is allowed. However, unless you use the following option, allowing zero-padding (i.e., the default operation) will make the output WARP dataset(s) be on a larger grid (also see '-expad' below). z-nopadzIf for some reason you require the warp volume tomatch the base volume, then use this option to have the outputWARP dataset(s) truncated.z -nopadWARPexpadaThis option instructs the program to pad the warp by an extra'EE' voxels (and then 3dQwarp starts optimizing it).This option is seldom needed, but can be useful if youmight later catenate the nonlinear warp -- via 3dNwarpCat --with an affine transformation that contains a large shift.Under that circumstance, the nonlinear warp might be shiftedpartially outside its original grid, so expanding that gridcan avoid this problem.Note that this option perforce turns off '-nopadWARP'.z -expad %daNormally, the incremental warp parameters are optimized insidea rectangular 'box' (24 dimensional for cubic patches, 81 forquintic patches), whose limits define the amount of distortionallowed at each step. Using '-ballopt' switches these limitsto be applied to a 'ball' (interior of a hypersphere), whichcan allow for larger incremental displacements. Use thisoption if you think things need to be able to move farther.z-balloptworkhardaUse the 'box' optimization limits instead of the 'ball'[this is the default at present].Note that if '-workhard' is used, then ball and box optimizationare alternated in the different iterations at each level, sothese two options have no effect in that case.z-boxoptz(more detailed description of the processz-verbrz5Cut out most of the fun fun fun progress messages :-(z-quietverbzOverwrite outputsz -overwriteaLocal Pearson minimization (i.e., EPI-T1 registration)This option has not be extensively testedIf you use '-lpc', then '-maxlev 0' is automatically set.If you want to go to more refined levels, you can set '-maxlev'This should be set up to have lpc as the second to last argumentand maxlev as the second to last argument, as needed by AFNIUsing maxlev > 1 is not recommended for EPI-T1 alignment.z-lpcrornrplpapearzELocal Pearson maximization. This option has not be extensively testedz-lpalpczHellinger distance: a matching function for the adventurousThis option has NOT be extensively tested for usefullnessand should be considered experimental at this infundibulum.z-helzMutual Information: a matching function for the adventurousThis option has NOT be extensively tested for usefullnessand should be considered experimental at this infundibulum.z-mizNormalized Mutual Information: a matching function for the adventurousThis option has NOT been extensively tested for usefullnessand should be considered experimental at this infundibulum.z-nmiNrSr):r!r"r#r r;Z base_filerlr r)Zresamplerrallineate_optsnowarprrZnonegZ nopenaltyr&ZpenfacZnoweightrtrr'rrJrrmrZpblurZemaskZnoXdisZnoYdisZnoZdisrrZminpatchrrrrrZQfinalZQonlyrZnopadrrrZbaxoptrrrrrrprnror,r,r,r-r+sd'                        rc@s>eZdZeddZeddZeddZeddZeddZdS)QwarpOutputSpeczMWarped source file. If plusminus is used, this is the undistortedsource file.)rzUndistorted base file.zDisplacement in mm for the source image.If plusminus is used this is the field suceptibility correctionwarp (in 'mm') for source image.zDisplacement in mm for the base image.If plus minus is used, this is the field suceptibility correctionwarp (in 'mm') for base image. This is only output if plusminusor iwarp options are passedzAuto-computed weight volume.N) r!r"r#r warped_source warped_base source_warp base_warpweightsr,r,r,r-rs rcs<eZdZdZdZeZeZfddZ ddZ ddZ Z S) Qwarpa` Allineate your images prior to passing them to this workflow. Examples -------- >>> from nipype.interfaces import afni >>> qwarp = afni.Qwarp() >>> qwarp.inputs.in_file = 'sub-01_dir-LR_epi.nii.gz' >>> qwarp.inputs.nopadWARP = True >>> qwarp.inputs.base_file = 'sub-01_dir-RL_epi.nii.gz' >>> qwarp.inputs.plusminus = True >>> qwarp.cmdline '3dQwarp -base sub-01_dir-RL_epi.nii.gz -source sub-01_dir-LR_epi.nii.gz -nopadWARP -prefix ppp_sub-01_dir-LR_epi -plusminus' >>> res = qwarp.run() # doctest: +SKIP >>> from nipype.interfaces import afni >>> qwarp = afni.Qwarp() >>> qwarp.inputs.in_file = 'structural.nii' >>> qwarp.inputs.base_file = 'mni.nii' >>> qwarp.inputs.resample = True >>> qwarp.cmdline '3dQwarp -base mni.nii -source structural.nii -prefix ppp_structural -resample' >>> res = qwarp.run() # doctest: +SKIP >>> from nipype.interfaces import afni >>> qwarp = afni.Qwarp() >>> qwarp.inputs.in_file = 'structural.nii' >>> qwarp.inputs.base_file = 'epi.nii' >>> qwarp.inputs.out_file = 'anatSSQ.nii.gz' >>> qwarp.inputs.resample = True >>> qwarp.inputs.lpc = True >>> qwarp.inputs.verb = True >>> qwarp.inputs.iwarp = True >>> qwarp.inputs.blur = [0,3] >>> qwarp.cmdline '3dQwarp -base epi.nii -blur 0.0 3.0 -source structural.nii -iwarp -prefix anatSSQ.nii.gz -resample -verb -lpc' >>> res = qwarp.run() # doctest: +SKIP >>> from nipype.interfaces import afni >>> qwarp = afni.Qwarp() >>> qwarp.inputs.in_file = 'structural.nii' >>> qwarp.inputs.base_file = 'mni.nii' >>> qwarp.inputs.duplo = True >>> qwarp.inputs.blur = [0,3] >>> qwarp.cmdline '3dQwarp -base mni.nii -blur 0.0 3.0 -duplo -source structural.nii -prefix ppp_structural' >>> res = qwarp.run() # doctest: +SKIP >>> from nipype.interfaces import afni >>> qwarp = afni.Qwarp() >>> qwarp.inputs.in_file = 'structural.nii' >>> qwarp.inputs.base_file = 'mni.nii' >>> qwarp.inputs.duplo = True >>> qwarp.inputs.minpatch = 25 >>> qwarp.inputs.blur = [0,3] >>> qwarp.inputs.out_file = 'Q25' >>> qwarp.cmdline '3dQwarp -base mni.nii -blur 0.0 3.0 -duplo -source structural.nii -minpatch 25 -prefix Q25' >>> res = qwarp.run() # doctest: +SKIP >>> qwarp2 = afni.Qwarp() >>> qwarp2.inputs.in_file = 'structural.nii' >>> qwarp2.inputs.base_file = 'mni.nii' >>> qwarp2.inputs.blur = [0,2] >>> qwarp2.inputs.out_file = 'Q11' >>> qwarp2.inputs.inilev = 7 >>> qwarp2.inputs.iniwarp = ['Q25_warp+tlrc.HEAD'] >>> qwarp2.cmdline '3dQwarp -base mni.nii -blur 0.0 2.0 -source structural.nii -inilev 7 -iniwarp Q25_warp+tlrc.HEAD -prefix Q11' >>> res2 = qwarp2.run() # doctest: +SKIP >>> res2 = qwarp2.run() # doctest: +SKIP >>> qwarp3 = afni.Qwarp() >>> qwarp3.inputs.in_file = 'structural.nii' >>> qwarp3.inputs.base_file = 'mni.nii' >>> qwarp3.inputs.allineate = True >>> qwarp3.inputs.allineate_opts = '-cose lpa -verb' >>> qwarp3.cmdline "3dQwarp -allineate -allineate_opts '-cose lpa -verb' -base mni.nii -source structural.nii -prefix ppp_structural" >>> res3 = qwarp3.run() # doctest: +SKIP See Also -------- For complete details, see the `3dQwarp Documentation. `__ Z3dQwarpcs.|dkr|jd|dStt|j|||S)Nr')rrrr&)r\rr'r)rr,r-r&szQwarp._format_argcCs|jj}t|jjsT|j|jjdd}|jj}|dkrDd}d}qtj |}d}nJ|jj}t |j j d|j j dg}|dkrd}d}n||d}d}t jjt jj|}t||d |d ||d <|jjst|d |d |d ||d<|jjrt|d|d |d ||d<t|jjr6t jj|jj|d<|jjrt|d|d |d ||d <t|d|d |d ||d<t|d|d |d ||d<t|d|d |d ||d<|S)N_QW)rCrQz.HEADz+tlrcr z.nii.gzz.niirF)rCrnewpathrZ_WARPrZ_WARPINVrrZ_PLUSZ_MINUSrZ _PLUS_WARPZ _MINUS_WARPrS)rTrUr rWrlrVr;rZrr[r4rrfindrrr`rrrrrmr)r\r]r5rZrRrCZext_indZout_dirr,r,r-r^s\            zQwarp._list_outputscCs|dkr|j|jjddSdS)Nrlr)rC)rVrWr;)r\rr,r,r- _gen_filenamecszQwarp._gen_filename) r!r"r#r$r_rr`rrTr&r^rrr,r,)rr-rs^ Mrc @sNeZdZedddddddZedd d dd d Zejddd dddddgdZdS)QwarpPlusMinusInputSpecz@Source image (opposite phase encoding direction than base image)z -source %sTz1.1.2r;F)rrrrurvr0z Qwarp.nii.gzz -prefix %srz Output file)rrkr5rrzNormally, the warp displacements dis(x) are defined to matchbase(x) to source(x+dis(x)). With this option, the matchis between base(x-dis(x)) and source(x+dis(x)) -- the twoimages 'meet in the middle'. For more info, view Qwarp` interfacez -plusminusrrr)r5rkrrrfN) r!r"r#r Z source_filerlr r)rr,r,r,r-rhs(rc@seZdZdZeZdS)QwarpPlusMinusaA version of 3dQwarp for performing field susceptibility correction using two images with opposing phase encoding directions. Examples -------- >>> from nipype.interfaces import afni >>> qwarp = afni.QwarpPlusMinus() >>> qwarp.inputs.in_file = 'sub-01_dir-LR_epi.nii.gz' >>> qwarp.inputs.nopadWARP = True >>> qwarp.inputs.base_file = 'sub-01_dir-RL_epi.nii.gz' >>> qwarp.cmdline '3dQwarp -prefix Qwarp.nii.gz -plusminus -base sub-01_dir-RL_epi.nii.gz -source sub-01_dir-LR_epi.nii.gz -nopadWARP' >>> res = warp.run() # doctest: +SKIP See Also -------- For complete details, see the `3dQwarp Documentation. `__ N)r!r"r#r$rr`r,r,r,r-rsr)|r$ros.pathrrZutils.filemaniprrrrrrr r r r r rrrrrrrrrrrrr r getLoggerrvrr.rFrOrarrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrr rrrr"r#r(r2r3r6r8r9r;r?r@rArXrYrZr\r_rarbrcrdrergrhrqrsrrrrrrrrrrrrrr,r,r,r-s 0 (  =n# C"!R O+3 #5"/ !D68y-6.<($1<% 1*>$R|5D .65r<