Results included in this manuscript come from preprocessing
performed using *fMRIPrep* 1.4.1
(@fmriprep1; @fmriprep2; RRID:SCR_016216),
which is based on *Nipype* 1.2.0
(@nipype1; @nipype2; RRID:SCR_002502).

Anatomical data preprocessing

: A total of 4 T1-weighted (T1w) images were found within the input
BIDS dataset.
All of them were corrected for intensity non-uniformity (INU)
with `N4BiasFieldCorrection` [@n4], distributed with ANTs 2.2.0 [@ants, RRID:SCR_004757].
The T1w-reference was then skull-stripped with a *Nipype* implementation of
the `antsBrainExtraction.sh` workflow (from ANTs), using OASIS30ANTs
as target template.
Brain tissue segmentation of cerebrospinal fluid (CSF),
white-matter (WM) and gray-matter (GM) was performed on
the brain-extracted T1w using `fast` [FSL 5.0.9, RRID:SCR_002823,
@fsl_fast].
A T1w-reference map was computed after registration of
4 T1w images (after INU-correction) using
`mri_robust_template` [FreeSurfer 6.0.1, @fs_template].
Brain surfaces were reconstructed using `recon-all` [FreeSurfer 6.0.1,
RRID:SCR_001847, @fs_reconall], and the brain mask estimated
previously was refined with a custom variation of the method to reconcile
ANTs-derived and FreeSurfer-derived segmentations of the cortical
gray-matter of Mindboggle [RRID:SCR_002438, @mindboggle].
Volume-based spatial normalization to two standard spaces (MNI152NLin2009cAsym, MNI152NLin6Asym) was performed through
nonlinear registration with `antsRegistration` (ANTs 2.2.0),
using brain-extracted versions of both T1w reference and the T1w template.
The following templates were selected for spatial normalization:
*ICBM 152 Nonlinear Asymmetrical template version 2009c* [@mni152nlin2009casym, RRID:SCR_008796; TemplateFlow ID: MNI152NLin2009cAsym], *FSL's MNI ICBM 152 non-linear 6th Generation Asymmetric Average Brain Stereotaxic Registration Model* [@mni152nlin6asym, RRID:SCR_002823; TemplateFlow ID: MNI152NLin6Asym].

Functional data preprocessing

: For each of the 9 BOLD runs found per subject (across all
tasks and sessions), the following preprocessing was performed.
First, a reference volume and its skull-stripped version were generated
using a custom methodology of *fMRIPrep*.
A deformation field to correct for susceptibility distortions was estimated
based on two echo-planar imaging (EPI) references with opposing phase-encoding
directions, using `3dQwarp` @afni (AFNI 20160207).
Based on the estimated susceptibility distortion, an
unwarped BOLD reference was calculated for a more accurate
co-registration with the anatomical reference.
Head-motion parameters with respect to the BOLD reference
(transformation matrices, and six corresponding rotation and translation
parameters) are estimated before any spatiotemporal filtering using
`mcflirt` [FSL 5.0.9, @mcflirt].
BOLD runs were slice-time corrected using `3dTshift` from
AFNI 20160207 [@afni, RRID:SCR_005927].
The BOLD time-series (including slice-timing correction when applied)
were resampled onto their original, native space by applying
a single, composite transform to correct for head-motion and
susceptibility distortions.
These resampled BOLD time-series will be referred to as *preprocessed
BOLD in original space*, or just *preprocessed BOLD*.
A T2* map was estimated from the preprocessed BOLD by fitting to a monoexponential signal
decay model with log-linear regression.
For each voxel, the maximal number of echoes with reliable signal in that voxel were
used to fit the model.
The calculated T2* map was then used to optimally combine preprocessed BOLD across
echoes following the method described in [@posse_t2s].
The optimally combined time series was carried forward as the *preprocessed BOLD*, and the T2* map was also retained as the BOLD reference.
The BOLD reference was then co-registered to the T1w reference using
`bbregister` (FreeSurfer) which implements boundary-based registration [@bbr].
Co-registration was configured with nine degrees of freedom to account
for distortions remaining in the BOLD reference.
The BOLD time-series were resampled into several standard spaces,
correspondingly generating the following *spatially-normalized,
preprocessed BOLD runs*: MNI152NLin2009cAsym, MNI152NLin6Asym.
First, a reference volume and its skull-stripped version were generated
using a custom methodology of *fMRIPrep*.
Automatic removal of motion artifacts using independent component analysis
[ICA-AROMA, @aroma] was performed on the *preprocessed BOLD on MNI space*
time-series after removal of non-steady state volumes and spatial smoothing
with an isotropic, Gaussian kernel of 6mm FWHM (full-width half-maximum).
Corresponding "non-aggresively" denoised runs were produced after such
smoothing.
Additionally, the "aggressive" noise-regressors were collected and placed
in the corresponding confounds file.
Several confounding time-series were calculated based on the
*preprocessed BOLD*: framewise displacement (FD), DVARS and
three region-wise global signals.
FD and DVARS are calculated for each functional run, both using their
implementations in *Nipype* [following the definitions by @power_fd_dvars].
The three global signals are extracted within the CSF, the WM, and
the whole-brain masks.
Additionally, a set of physiological regressors were extracted to
allow for component-based noise correction [*CompCor*, @compcor].
Principal components are estimated after high-pass filtering the
*preprocessed BOLD* time-series (using a discrete cosine filter with
128s cut-off) for the two *CompCor* variants: temporal (tCompCor)
and anatomical (aCompCor).
tCompCor components are then calculated from the top 5% variable
voxels within a mask covering the subcortical regions.
This subcortical mask is obtained by heavily eroding the brain mask,
which ensures it does not include cortical GM regions.
For aCompCor, components are calculated within the intersection of
the aforementioned mask and the union of CSF and WM masks calculated
in T1w space, after their projection to the native space of each
functional run (using the inverse BOLD-to-T1w transformation). Components
are also calculated separately within the WM and CSF masks.
For each CompCor decomposition, the *k* components with the largest singular
values are retained, such that the retained components' time series are
sufficient to explain 50 percent of variance across the nuisance mask (CSF,
WM, combined, or temporal). The remaining components are dropped from
consideration.
The head-motion estimates calculated in the correction step were also
placed within the corresponding confounds file.
The confound time series derived from head motion estimates and global
signals were expanded with the inclusion of temporal derivatives and
quadratic terms for each [@confounds_satterthwaite_2013].
Frames that exceeded a threshold of 0.5 mm FD or 1.5 standardised DVARS
were annotated as motion outliers.
The BOLD time-series, were resampled to surfaces on the following
spaces: *fsaverage6*.
All resamplings can be performed with *a single interpolation
step* by composing all the pertinent transformations (i.e. head-motion
transform matrices, susceptibility distortion correction when available,
and co-registrations to anatomical and output spaces).
Gridded (volumetric) resamplings were performed using `antsApplyTransforms` (ANTs),
configured with Lanczos interpolation to minimize the smoothing
effects of other kernels [@lanczos].
Non-gridded (surface) resamplings were performed using `mri_vol2surf`
(FreeSurfer).


Functional data preprocessing

: For each of the 9 BOLD runs found per subject (across all
tasks and sessions), the following preprocessing was performed.
First, a reference volume and its skull-stripped version were generated
using a custom methodology of *fMRIPrep*.
A deformation field to correct for susceptibility distortions was estimated
based on two echo-planar imaging (EPI) references with opposing phase-encoding
directions, using `3dQwarp` @afni (AFNI 20160207).
Based on the estimated susceptibility distortion, an
unwarped BOLD reference was calculated for a more accurate
co-registration with the anatomical reference.
The BOLD reference was then co-registered to the T1w reference using
`bbregister` (FreeSurfer) which implements boundary-based registration [@bbr].
Co-registration was configured with nine degrees of freedom to account
for distortions remaining in the BOLD reference.
Head-motion parameters with respect to the BOLD reference
(transformation matrices, and six corresponding rotation and translation
parameters) are estimated before any spatiotemporal filtering using
`mcflirt` [FSL 5.0.9, @mcflirt].
BOLD runs were slice-time corrected using `3dTshift` from
AFNI 20160207 [@afni, RRID:SCR_005927].
The BOLD time-series, were resampled to surfaces on the following
spaces: *fsaverage6*.
The BOLD time-series (including slice-timing correction when applied)
were resampled onto their original, native space by applying
a single, composite transform to correct for head-motion and
susceptibility distortions.
These resampled BOLD time-series will be referred to as *preprocessed
BOLD in original space*, or just *preprocessed BOLD*.
The BOLD time-series were resampled into several standard spaces,
correspondingly generating the following *spatially-normalized,
preprocessed BOLD runs*: MNI152NLin2009cAsym, MNI152NLin6Asym.
First, a reference volume and its skull-stripped version were generated
using a custom methodology of *fMRIPrep*.
Automatic removal of motion artifacts using independent component analysis
[ICA-AROMA, @aroma] was performed on the *preprocessed BOLD on MNI space*
time-series after removal of non-steady state volumes and spatial smoothing
with an isotropic, Gaussian kernel of 6mm FWHM (full-width half-maximum).
Corresponding "non-aggresively" denoised runs were produced after such
smoothing.
Additionally, the "aggressive" noise-regressors were collected and placed
in the corresponding confounds file.
Several confounding time-series were calculated based on the
*preprocessed BOLD*: framewise displacement (FD), DVARS and
three region-wise global signals.
FD and DVARS are calculated for each functional run, both using their
implementations in *Nipype* [following the definitions by @power_fd_dvars].
The three global signals are extracted within the CSF, the WM, and
the whole-brain masks.
Additionally, a set of physiological regressors were extracted to
allow for component-based noise correction [*CompCor*, @compcor].
Principal components are estimated after high-pass filtering the
*preprocessed BOLD* time-series (using a discrete cosine filter with
128s cut-off) for the two *CompCor* variants: temporal (tCompCor)
and anatomical (aCompCor).
tCompCor components are then calculated from the top 5% variable
voxels within a mask covering the subcortical regions.
This subcortical mask is obtained by heavily eroding the brain mask,
which ensures it does not include cortical GM regions.
For aCompCor, components are calculated within the intersection of
the aforementioned mask and the union of CSF and WM masks calculated
in T1w space, after their projection to the native space of each
functional run (using the inverse BOLD-to-T1w transformation). Components
are also calculated separately within the WM and CSF masks.
For each CompCor decomposition, the *k* components with the largest singular
values are retained, such that the retained components' time series are
sufficient to explain 50 percent of variance across the nuisance mask (CSF,
WM, combined, or temporal). The remaining components are dropped from
consideration.
The head-motion estimates calculated in the correction step were also
placed within the corresponding confounds file.
The confound time series derived from head motion estimates and global
signals were expanded with the inclusion of temporal derivatives and
quadratic terms for each [@confounds_satterthwaite_2013].
Frames that exceeded a threshold of 0.5 mm FD or 1.5 standardised DVARS
were annotated as motion outliers.
All resamplings can be performed with *a single interpolation
step* by composing all the pertinent transformations (i.e. head-motion
transform matrices, susceptibility distortion correction when available,
and co-registrations to anatomical and output spaces).
Gridded (volumetric) resamplings were performed using `antsApplyTransforms` (ANTs),
configured with Lanczos interpolation to minimize the smoothing
effects of other kernels [@lanczos].
Non-gridded (surface) resamplings were performed using `mri_vol2surf`
(FreeSurfer).


Many internal operations of *fMRIPrep* use
*Nilearn* 0.5.2 [@nilearn, RRID:SCR_001362],
mostly within the functional processing workflow.
For more details of the pipeline, see [the section corresponding
to workflows in *fMRIPrep*'s documentation](https://fmriprep.readthedocs.io/en/latest/workflows.html "FMRIPrep's documentation").


### References

Results included in this manuscript come from preprocessing performed
using \emph{fMRIPrep} 1.4.1 (\citet{fmriprep1}; \citet{fmriprep2};
RRID:SCR\_016216), which is based on \emph{Nipype} 1.2.0
(\citet{nipype1}; \citet{nipype2}; RRID:SCR\_002502).

\begin{description}
\item[Anatomical data preprocessing]
A total of 4 T1-weighted (T1w) images were found within the input BIDS
dataset. All of them were corrected for intensity non-uniformity (INU)
with \texttt{N4BiasFieldCorrection} \citep{n4}, distributed with ANTs
2.2.0 \citep[RRID:SCR\_004757]{ants}. The T1w-reference was then
skull-stripped with a \emph{Nipype} implementation of the
\texttt{antsBrainExtraction.sh} workflow (from ANTs), using OASIS30ANTs
as target template. Brain tissue segmentation of cerebrospinal fluid
(CSF), white-matter (WM) and gray-matter (GM) was performed on the
brain-extracted T1w using \texttt{fast} \citep[FSL 5.0.9,
RRID:SCR\_002823,][]{fsl_fast}. A T1w-reference map was computed after
registration of 4 T1w images (after INU-correction) using
\texttt{mri\_robust\_template} \citep[FreeSurfer 6.0.1,][]{fs_template}.
Brain surfaces were reconstructed using \texttt{recon-all}
\citep[FreeSurfer 6.0.1, RRID:SCR\_001847,][]{fs_reconall}, and the
brain mask estimated previously was refined with a custom variation of
the method to reconcile ANTs-derived and FreeSurfer-derived
segmentations of the cortical gray-matter of Mindboggle
\citep[RRID:SCR\_002438,][]{mindboggle}. Volume-based spatial
normalization to two standard spaces (MNI152NLin2009cAsym,
MNI152NLin6Asym) was performed through nonlinear registration with
\texttt{antsRegistration} (ANTs 2.2.0), using brain-extracted versions
of both T1w reference and the T1w template. The following templates were
selected for spatial normalization: \emph{ICBM 152 Nonlinear
Asymmetrical template version 2009c} {[}\citet{mni152nlin2009casym},
RRID:SCR\_008796; TemplateFlow ID: MNI152NLin2009cAsym{]}, \emph{FSL's
MNI ICBM 152 non-linear 6th Generation Asymmetric Average Brain
Stereotaxic Registration Model} {[}\citet{mni152nlin6asym},
RRID:SCR\_002823; TemplateFlow ID: MNI152NLin6Asym{]}.
\item[Functional data preprocessing]
For each of the 9 BOLD runs found per subject (across all tasks and
sessions), the following preprocessing was performed. First, a reference
volume and its skull-stripped version were generated using a custom
methodology of \emph{fMRIPrep}. A deformation field to correct for
susceptibility distortions was estimated based on two echo-planar
imaging (EPI) references with opposing phase-encoding directions, using
\texttt{3dQwarp} \citet{afni} (AFNI 20160207). Based on the estimated
susceptibility distortion, an unwarped BOLD reference was calculated for
a more accurate co-registration with the anatomical reference.
Head-motion parameters with respect to the BOLD reference
(transformation matrices, and six corresponding rotation and translation
parameters) are estimated before any spatiotemporal filtering using
\texttt{mcflirt} \citep[FSL 5.0.9,][]{mcflirt}. BOLD runs were
slice-time corrected using \texttt{3dTshift} from AFNI 20160207
\citep[RRID:SCR\_005927]{afni}. The BOLD time-series (including
slice-timing correction when applied) were resampled onto their
original, native space by applying a single, composite transform to
correct for head-motion and susceptibility distortions. These resampled
BOLD time-series will be referred to as \emph{preprocessed BOLD in
original space}, or just \emph{preprocessed BOLD}. A T2* map was
estimated from the preprocessed BOLD by fitting to a monoexponential
signal decay model with log-linear regression. For each voxel, the
maximal number of echoes with reliable signal in that voxel were used to
fit the model. The calculated T2* map was then used to optimally combine
preprocessed BOLD across echoes following the method described in
\citep{posse_t2s}. The optimally combined time series was carried
forward as the \emph{preprocessed BOLD}, and the T2* map was also
retained as the BOLD reference. The BOLD reference was then
co-registered to the T1w reference using \texttt{bbregister}
(FreeSurfer) which implements boundary-based registration \citep{bbr}.
Co-registration was configured with nine degrees of freedom to account
for distortions remaining in the BOLD reference. The BOLD time-series
were resampled into several standard spaces, correspondingly generating
the following \emph{spatially-normalized, preprocessed BOLD runs}:
MNI152NLin2009cAsym, MNI152NLin6Asym. First, a reference volume and its
skull-stripped version were generated using a custom methodology of
\emph{fMRIPrep}. Automatic removal of motion artifacts using independent
component analysis \citep[ICA-AROMA,][]{aroma} was performed on the
\emph{preprocessed BOLD on MNI space} time-series after removal of
non-steady state volumes and spatial smoothing with an isotropic,
Gaussian kernel of 6mm FWHM (full-width half-maximum). Corresponding
``non-aggresively'' denoised runs were produced after such smoothing.
Additionally, the ``aggressive'' noise-regressors were collected and
placed in the corresponding confounds file. Several confounding
time-series were calculated based on the \emph{preprocessed BOLD}:
framewise displacement (FD), DVARS and three region-wise global signals.
FD and DVARS are calculated for each functional run, both using their
implementations in \emph{Nipype} \citep[following the definitions
by][]{power_fd_dvars}. The three global signals are extracted within the
CSF, the WM, and the whole-brain masks. Additionally, a set of
physiological regressors were extracted to allow for component-based
noise correction \citep[\emph{CompCor},][]{compcor}. Principal
components are estimated after high-pass filtering the
\emph{preprocessed BOLD} time-series (using a discrete cosine filter
with 128s cut-off) for the two \emph{CompCor} variants: temporal
(tCompCor) and anatomical (aCompCor). tCompCor components are then
calculated from the top 5\% variable voxels within a mask covering the
subcortical regions. This subcortical mask is obtained by heavily
eroding the brain mask, which ensures it does not include cortical GM
regions. For aCompCor, components are calculated within the intersection
of the aforementioned mask and the union of CSF and WM masks calculated
in T1w space, after their projection to the native space of each
functional run (using the inverse BOLD-to-T1w transformation).
Components are also calculated separately within the WM and CSF masks.
For each CompCor decomposition, the \emph{k} components with the largest
singular values are retained, such that the retained components' time
series are sufficient to explain 50 percent of variance across the
nuisance mask (CSF, WM, combined, or temporal). The remaining components
are dropped from consideration. The head-motion estimates calculated in
the correction step were also placed within the corresponding confounds
file. The confound time series derived from head motion estimates and
global signals were expanded with the inclusion of temporal derivatives
and quadratic terms for each \citep{confounds_satterthwaite_2013}.
Frames that exceeded a threshold of 0.5 mm FD or 1.5 standardised DVARS
were annotated as motion outliers. The BOLD time-series, were resampled
to surfaces on the following spaces: \emph{fsaverage6}. All resamplings
can be performed with \emph{a single interpolation step} by composing
all the pertinent transformations (i.e.~head-motion transform matrices,
susceptibility distortion correction when available, and
co-registrations to anatomical and output spaces). Gridded (volumetric)
resamplings were performed using \texttt{antsApplyTransforms} (ANTs),
configured with Lanczos interpolation to minimize the smoothing effects
of other kernels \citep{lanczos}. Non-gridded (surface) resamplings were
performed using \texttt{mri\_vol2surf} (FreeSurfer).
\item[Functional data preprocessing]
For each of the 9 BOLD runs found per subject (across all tasks and
sessions), the following preprocessing was performed. First, a reference
volume and its skull-stripped version were generated using a custom
methodology of \emph{fMRIPrep}. A deformation field to correct for
susceptibility distortions was estimated based on two echo-planar
imaging (EPI) references with opposing phase-encoding directions, using
\texttt{3dQwarp} \citet{afni} (AFNI 20160207). Based on the estimated
susceptibility distortion, an unwarped BOLD reference was calculated for
a more accurate co-registration with the anatomical reference. The BOLD
reference was then co-registered to the T1w reference using
\texttt{bbregister} (FreeSurfer) which implements boundary-based
registration \citep{bbr}. Co-registration was configured with nine
degrees of freedom to account for distortions remaining in the BOLD
reference. Head-motion parameters with respect to the BOLD reference
(transformation matrices, and six corresponding rotation and translation
parameters) are estimated before any spatiotemporal filtering using
\texttt{mcflirt} \citep[FSL 5.0.9,][]{mcflirt}. BOLD runs were
slice-time corrected using \texttt{3dTshift} from AFNI 20160207
\citep[RRID:SCR\_005927]{afni}. The BOLD time-series, were resampled to
surfaces on the following spaces: \emph{fsaverage6}. The BOLD
time-series (including slice-timing correction when applied) were
resampled onto their original, native space by applying a single,
composite transform to correct for head-motion and susceptibility
distortions. These resampled BOLD time-series will be referred to as
\emph{preprocessed BOLD in original space}, or just \emph{preprocessed
BOLD}. The BOLD time-series were resampled into several standard spaces,
correspondingly generating the following \emph{spatially-normalized,
preprocessed BOLD runs}: MNI152NLin2009cAsym, MNI152NLin6Asym. First, a
reference volume and its skull-stripped version were generated using a
custom methodology of \emph{fMRIPrep}. Automatic removal of motion
artifacts using independent component analysis
\citep[ICA-AROMA,][]{aroma} was performed on the \emph{preprocessed BOLD
on MNI space} time-series after removal of non-steady state volumes and
spatial smoothing with an isotropic, Gaussian kernel of 6mm FWHM
(full-width half-maximum). Corresponding ``non-aggresively'' denoised
runs were produced after such smoothing. Additionally, the
``aggressive'' noise-regressors were collected and placed in the
corresponding confounds file. Several confounding time-series were
calculated based on the \emph{preprocessed BOLD}: framewise displacement
(FD), DVARS and three region-wise global signals. FD and DVARS are
calculated for each functional run, both using their implementations in
\emph{Nipype} \citep[following the definitions by][]{power_fd_dvars}.
The three global signals are extracted within the CSF, the WM, and the
whole-brain masks. Additionally, a set of physiological regressors were
extracted to allow for component-based noise correction
\citep[\emph{CompCor},][]{compcor}. Principal components are estimated
after high-pass filtering the \emph{preprocessed BOLD} time-series
(using a discrete cosine filter with 128s cut-off) for the two
\emph{CompCor} variants: temporal (tCompCor) and anatomical (aCompCor).
tCompCor components are then calculated from the top 5\% variable voxels
within a mask covering the subcortical regions. This subcortical mask is
obtained by heavily eroding the brain mask, which ensures it does not
include cortical GM regions. For aCompCor, components are calculated
within the intersection of the aforementioned mask and the union of CSF
and WM masks calculated in T1w space, after their projection to the
native space of each functional run (using the inverse BOLD-to-T1w
transformation). Components are also calculated separately within the WM
and CSF masks. For each CompCor decomposition, the \emph{k} components
with the largest singular values are retained, such that the retained
components' time series are sufficient to explain 50 percent of variance
across the nuisance mask (CSF, WM, combined, or temporal). The remaining
components are dropped from consideration. The head-motion estimates
calculated in the correction step were also placed within the
corresponding confounds file. The confound time series derived from head
motion estimates and global signals were expanded with the inclusion of
temporal derivatives and quadratic terms for each
\citep{confounds_satterthwaite_2013}. Frames that exceeded a threshold
of 0.5 mm FD or 1.5 standardised DVARS were annotated as motion
outliers. All resamplings can be performed with \emph{a single
interpolation step} by composing all the pertinent transformations
(i.e.~head-motion transform matrices, susceptibility distortion
correction when available, and co-registrations to anatomical and output
spaces). Gridded (volumetric) resamplings were performed using
\texttt{antsApplyTransforms} (ANTs), configured with Lanczos
interpolation to minimize the smoothing effects of other kernels
\citep{lanczos}. Non-gridded (surface) resamplings were performed using
\texttt{mri\_vol2surf} (FreeSurfer).
\end{description}

Many internal operations of \emph{fMRIPrep} use \emph{Nilearn} 0.5.2
\citep[RRID:SCR\_001362]{nilearn}, mostly within the functional
processing workflow. For more details of the pipeline, see
\href{https://fmriprep.readthedocs.io/en/latest/workflows.html}{the
section corresponding to workflows in \emph{fMRIPrep}'s documentation}.

\hypertarget{references}{%
\subsubsection{References}\label{references}}

\bibliography{/usr/local/miniconda/lib/python3.7/site-packages/fmriprep/data/boilerplate.bib}

Bibliography

@article{fmriprep1,
    author = {Esteban, Oscar and Markiewicz, Christopher and Blair, Ross W and Moodie, Craig and Isik, Ayse Ilkay and Erramuzpe Aliaga, Asier and Kent, James and Goncalves, Mathias and DuPre, Elizabeth and Snyder, Madeleine and Oya, Hiroyuki and Ghosh, Satrajit and Wright, Jessey and Durnez, Joke and Poldrack, Russell and Gorgolewski, Krzysztof Jacek},
    title = {{fMRIPrep}: a robust preprocessing pipeline for functional {MRI}},
    year = {2018},
    doi = {10.1038/s41592-018-0235-4},
    journal = {Nature Methods}
}

@article{fmriprep2,
    author = {Esteban, Oscar and Blair, Ross and Markiewicz, Christopher J. and Berleant, Shoshana L. and Moodie, Craig and Ma, Feilong and Isik, Ayse Ilkay and Erramuzpe, Asier and Kent, James D. andGoncalves, Mathias and DuPre, Elizabeth and Sitek, Kevin R. and Gomez, Daniel E. P. and Lurie, Daniel J. and Ye, Zhifang and Poldrack, Russell A. and Gorgolewski, Krzysztof J.},
    title = {fMRIPrep},
    year = 2018,
    doi = {10.5281/zenodo.852659},
    publisher = {Zenodo},
    journal = {Software}
}

@article{nipype1,
    author = {Gorgolewski, K. and Burns, C. D. and Madison, C. and Clark, D. and Halchenko, Y. O. and Waskom, M. L. and Ghosh, S.},
    doi = {10.3389/fninf.2011.00013},
    journal = {Frontiers in Neuroinformatics},
    pages = 13,
    shorttitle = {Nipype},
    title = {Nipype: a flexible, lightweight and extensible neuroimaging data processing framework in Python},
    volume = 5,
    year = 2011
}

@article{nipype2,
    author = {Gorgolewski, Krzysztof J. and Esteban, Oscar and Markiewicz, Christopher J. and Ziegler, Erik and Ellis, David Gage and Notter, Michael Philipp and Jarecka, Dorota and Johnson, Hans and Burns, Christopher and Manhães-Savio, Alexandre and Hamalainen, Carlo and Yvernault, Benjamin and Salo, Taylor and Jordan, Kesshi and Goncalves, Mathias and Waskom, Michael and Clark, Daniel and Wong, Jason and Loney, Fred and Modat, Marc and Dewey, Blake E and Madison, Cindee and Visconti di Oleggio Castello, Matteo and Clark, Michael G. and Dayan, Michael and Clark, Dav and Keshavan, Anisha and Pinsard, Basile and Gramfort, Alexandre and Berleant, Shoshana and Nielson, Dylan M. and Bougacha, Salma and Varoquaux, Gael and Cipollini, Ben and Markello, Ross and Rokem, Ariel and Moloney, Brendan and Halchenko, Yaroslav O. and Wassermann , Demian and Hanke, Michael and Horea, Christian and Kaczmarzyk, Jakub and Gilles de Hollander and DuPre, Elizabeth and Gillman, Ashley and Mordom, David and Buchanan, Colin and Tungaraza, Rosalia and Pauli, Wolfgang M. and Iqbal, Shariq and Sikka, Sharad and Mancini, Matteo and Schwartz, Yannick and Malone, Ian B. and Dubois, Mathieu and Frohlich, Caroline and Welch, David and Forbes, Jessica and Kent, James and Watanabe, Aimi and Cumba, Chad and Huntenburg, Julia M. and Kastman, Erik and Nichols, B. Nolan and Eshaghi, Arman and Ginsburg, Daniel and Schaefer, Alexander and Acland, Benjamin and Giavasis, Steven and Kleesiek, Jens and Erickson, Drew and Küttner, René and Haselgrove, Christian and Correa, Carlos and Ghayoor, Ali and Liem, Franz and Millman, Jarrod and Haehn, Daniel and Lai, Jeff and Zhou, Dale and Blair, Ross and Glatard, Tristan and Renfro, Mandy and Liu, Siqi and Kahn, Ari E. and Pérez-García, Fernando and Triplett, William and Lampe, Leonie and Stadler, Jörg and Kong, Xiang-Zhen and Hallquist, Michael and Chetverikov, Andrey and Salvatore, John and Park, Anne and Poldrack, Russell and Craddock, R. Cameron and Inati, Souheil and Hinds, Oliver and Cooper, Gavin and Perkins, L. Nathan and Marina, Ana and Mattfeld, Aaron and Noel, Maxime and Lukas Snoek and Matsubara, K and Cheung, Brian and Rothmei, Simon and Urchs, Sebastian and Durnez, Joke and Mertz, Fred and Geisler, Daniel and Floren, Andrew and Gerhard, Stephan and Sharp, Paul and Molina-Romero, Miguel and Weinstein, Alejandro and Broderick, William and Saase, Victor and Andberg, Sami Kristian and Harms, Robbert and Schlamp, Kai and Arias, Jaime and Papadopoulos Orfanos, Dimitri and Tarbert, Claire and Tambini, Arielle and De La Vega, Alejandro and Nickson, Thomas and Brett, Matthew and Falkiewicz, Marcel and Podranski, Kornelius and Linkersdörfer, Janosch and Flandin, Guillaume and Ort, Eduard and Shachnev, Dmitry and McNamee, Daniel and Davison, Andrew and Varada, Jan and Schwabacher, Isaac and Pellman, John and Perez-Guevara, Martin and Khanuja, Ranjeet and Pannetier, Nicolas and McDermottroe, Conor and Ghosh, Satrajit},
    title = {Nipype},
    year = 2018,
    doi = {10.5281/zenodo.596855},
    publisher = {Zenodo},
    journal = {Software}
}

@article{n4,
    author = {Tustison, N. J. and Avants, B. B. and Cook, P. A. and Zheng, Y. and Egan, A. and Yushkevich, P. A. and Gee, J. C.},
    doi = {10.1109/TMI.2010.2046908},
    issn = {0278-0062},
    journal = {IEEE Transactions on Medical Imaging},
    number = 6,
    pages = {1310-1320},
    shorttitle = {N4ITK},
    title = {N4ITK: Improved N3 Bias Correction},
    volume = 29,
    year = 2010
}

@article{fs_reconall,
    author = {Dale, Anders M. and Fischl, Bruce and Sereno, Martin I.},
    doi = {10.1006/nimg.1998.0395},
    issn = {1053-8119},
    journal = {NeuroImage},
    number = 2,
    pages = {179-194},
    shorttitle = {Cortical Surface-Based Analysis},
    title = {Cortical Surface-Based Analysis: I. Segmentation and Surface Reconstruction},
    url = {http://www.sciencedirect.com/science/article/pii/S1053811998903950},
    volume = 9,
    year = 1999
}



@article{mindboggle,
    author = {Klein, Arno and Ghosh, Satrajit S. and Bao, Forrest S. and Giard, Joachim and Häme, Yrjö and Stavsky, Eliezer and Lee, Noah and Rossa, Brian and Reuter, Martin and Neto, Elias Chaibub and Keshavan, Anisha},
    doi = {10.1371/journal.pcbi.1005350},
    issn = {1553-7358},
    journal = {PLOS Computational Biology},
    number = 2,
    pages = {e1005350},
    title = {Mindboggling morphometry of human brains},
    url = {http://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1005350},
    volume = 13,
    year = 2017
}

@article{mni152lin,
    title = {A {Probabilistic} {Atlas} of the {Human} {Brain}: {Theory} and {Rationale} for {Its} {Development}: {The} {International} {Consortium} for {Brain} {Mapping} ({ICBM})},
    author = {Mazziotta, John C. and Toga, Arthur W. and Evans, Alan and Fox, Peter and Lancaster, Jack},
    volume = {2},
    issn = {1053-8119},
    shorttitle = {A {Probabilistic} {Atlas} of the {Human} {Brain}},
    doi = {10.1006/nimg.1995.1012},
    number = {2, Part A},
    journal = {NeuroImage},
    year = {1995},
    pages = {89--101}
}

@article{mni152nlin2009casym,
    title = {Unbiased nonlinear average age-appropriate brain templates from birth to adulthood},
    author = {Fonov, VS and Evans, AC and McKinstry, RC and Almli, CR and Collins, DL},
    doi = {10.1016/S1053-8119(09)70884-5},
    journal = {NeuroImage},
    pages = {S102},
    volume = {47, Supplement 1},
    year = 2009
}

@article{mni152nlin6asym,
    author = {Evans, AC and Janke, AL and Collins, DL and Baillet, S},
    title = {Brain templates and atlases},
    doi = {10.1016/j.neuroimage.2012.01.024},
    journal = {NeuroImage},
    volume = {62},
    number = {2},
    pages = {911--922},
    year = 2012
}

@article{ants,
    author = {Avants, B.B. and Epstein, C.L. and Grossman, M. and Gee, J.C.},
    doi = {10.1016/j.media.2007.06.004},
    issn = {1361-8415},
    journal = {Medical Image Analysis},
    number = 1,
    pages = {26-41},
    shorttitle = {Symmetric diffeomorphic image registration with cross-correlation},
    title = {Symmetric diffeomorphic image registration with cross-correlation: Evaluating automated labeling of elderly and neurodegenerative brain},
    url = {http://www.sciencedirect.com/science/article/pii/S1361841507000606},
    volume = 12,
    year = 2008
}

@article{fsl_fast,
    author = {Zhang, Y. and Brady, M. and Smith, S.},
    doi = {10.1109/42.906424},
    issn = {0278-0062},
    journal = {IEEE Transactions on Medical Imaging},
    number = 1,
    pages = {45-57},
    title = {Segmentation of brain {MR} images through a hidden Markov random field model and the expectation-maximization algorithm},
    volume = 20,
    year = 2001
}


@article{fieldmapless1,
    author = {Wang, Sijia and Peterson, Daniel J. and Gatenby, J. C. and Li, Wenbin and Grabowski, Thomas J. and Madhyastha, Tara M.},
    doi = {10.3389/fninf.2017.00017},
    issn = {1662-5196},
    journal = {Frontiers in Neuroinformatics},
    language = {English},
    title = {Evaluation of Field Map and Nonlinear Registration Methods for Correction of Susceptibility Artifacts in Diffusion {MRI}},
    url = {http://journal.frontiersin.org/article/10.3389/fninf.2017.00017/full},
    volume = 11,
    year = 2017
}

@phdthesis{fieldmapless2,
    address = {Berlin},
    author = {Huntenburg, Julia M.},
    language = {eng},
    school = {Freie Universität},
    title = {Evaluating nonlinear coregistration of {BOLD} {EPI} and T1w images},
    type = {Master's Thesis},
    url = {http://hdl.handle.net/11858/00-001M-0000-002B-1CB5-A},
    year = 2014
}

@article{fieldmapless3,
    author = {Treiber, Jeffrey Mark and White, Nathan S. and Steed, Tyler Christian and Bartsch, Hauke and Holland, Dominic and Farid, Nikdokht and McDonald, Carrie R. and Carter, Bob S. and Dale, Anders Martin and Chen, Clark C.},
    doi = {10.1371/journal.pone.0152472},
    issn = {1932-6203},
    journal = {PLOS ONE},
    number = 3,
    pages = {e0152472},
    title = {Characterization and Correction of Geometric Distortions in 814 Diffusion Weighted Images},
    url = {http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0152472},
    volume = 11,
    year = 2016
}

@article{flirt,
    title = {A global optimisation method for robust affine registration of brain images},
    volume = {5},
    issn = {1361-8415},
    url = {http://www.sciencedirect.com/science/article/pii/S1361841501000366},
    doi = {10.1016/S1361-8415(01)00036-6},
    number = {2},
    urldate = {2018-07-27},
    journal = {Medical Image Analysis},
    author = {Jenkinson, Mark and Smith, Stephen},
    year = {2001},
    keywords = {Affine transformation, flirt, fsl, Global optimisation, Multi-resolution search, Multimodal registration, Robustness},
    pages = {143--156}
}

@article{mcflirt,
    author = {Jenkinson, Mark and Bannister, Peter and Brady, Michael and Smith, Stephen},
    doi = {10.1006/nimg.2002.1132},
    issn = {1053-8119},
    journal = {NeuroImage},
    number = 2,
    pages = {825-841},
    title = {Improved Optimization for the Robust and Accurate Linear Registration and Motion Correction of Brain Images},
    url = {http://www.sciencedirect.com/science/article/pii/S1053811902911328},
    volume = 17,
    year = 2002
}

@article{bbr,
    author = {Greve, Douglas N and Fischl, Bruce},
    doi = {10.1016/j.neuroimage.2009.06.060},
    issn = {1095-9572},
    journal = {NeuroImage},
    number = 1,
    pages = {63-72},
    title = {Accurate and robust brain image alignment using boundary-based registration},
    volume = 48,
    year = 2009
}

@article{aroma,
    author = {Pruim, Raimon H. R. and Mennes, Maarten and van Rooij, Daan and Llera, Alberto and Buitelaar, Jan K. and Beckmann, Christian F.},
    doi = {10.1016/j.neuroimage.2015.02.064},
    issn = {1053-8119},
    journal = {NeuroImage},
    number = {Supplement C},
    pages = {267-277},
    shorttitle = {ICA-AROMA},
    title = {ICA-{AROMA}: A robust {ICA}-based strategy for removing motion artifacts from fMRI data},
    url = {http://www.sciencedirect.com/science/article/pii/S1053811915001822},
    volume = 112,
    year = 2015
}

@article{power_fd_dvars,
    author = {Power, Jonathan D. and Mitra, Anish and Laumann, Timothy O. and Snyder, Abraham Z. and Schlaggar, Bradley L. and Petersen, Steven E.},
    doi = {10.1016/j.neuroimage.2013.08.048},
    issn = {1053-8119},
    journal = {NeuroImage},
    number = {Supplement C},
    pages = {320-341},
    title = {Methods to detect, characterize, and remove motion artifact in resting state fMRI},
    url = {http://www.sciencedirect.com/science/article/pii/S1053811913009117},
    volume = 84,
    year = 2014
}

@article{confounds_satterthwaite_2013,
    author = {Satterthwaite, Theodore D. and Elliott, Mark A. and Gerraty, Raphael T. and Ruparel, Kosha and Loughead, James and Calkins, Monica E. and Eickhoff, Simon B. and Hakonarson, Hakon and Gur, Ruben C. and Gur, Raquel E. and Wolf, Daniel H.},
    doi = {10.1016/j.neuroimage.2012.08.052},
    issn = {10538119},
    journal = {NeuroImage},
    number = 1,
    pages = {240--256},
    title = {{An improved framework for confound regression and filtering for control of motion artifact in the preprocessing of resting-state functional connectivity data}},
    url = {http://linkinghub.elsevier.com/retrieve/pii/S1053811912008609},
    volume = 64,
    year = 2013
}


@article{nilearn,
    author = {Abraham, Alexandre and Pedregosa, Fabian and Eickenberg, Michael and Gervais, Philippe and Mueller, Andreas and Kossaifi, Jean and Gramfort, Alexandre and Thirion, Bertrand and Varoquaux, Gael},
    doi = {10.3389/fninf.2014.00014},
    issn = {1662-5196},
    journal = {Frontiers in Neuroinformatics},
    language = {English},
    title = {Machine learning for neuroimaging with scikit-learn},
    url = {https://www.frontiersin.org/articles/10.3389/fninf.2014.00014/full},
    volume = 8,
    year = 2014
}

@article{lanczos,
    author = {Lanczos, C.},
    doi = {10.1137/0701007},
    issn = {0887-459X},
    journal = {Journal of the Society for Industrial and Applied Mathematics Series B Numerical Analysis},
    number = 1,
    pages = {76-85},
    title = {Evaluation of Noisy Data},
    url = {http://epubs.siam.org/doi/10.1137/0701007},
    volume = 1,
    year = 1964
}

@article{compcor,
    author = {Behzadi, Yashar and Restom, Khaled and Liau, Joy and Liu, Thomas T.},
    doi = {10.1016/j.neuroimage.2007.04.042},
    issn = {1053-8119},
    journal = {NeuroImage},
    number = 1,
    pages = {90-101},
    title = {A component based noise correction method ({CompCor}) for {BOLD} and perfusion based fMRI},
    url = {http://www.sciencedirect.com/science/article/pii/S1053811907003837},
    volume = 37,
    year = 2007
}

@article{hcppipelines,
    author = {Glasser, Matthew F. and Sotiropoulos, Stamatios N. and Wilson, J. Anthony and Coalson, Timothy S. and Fischl, Bruce and Andersson, Jesper L. and Xu, Junqian and Jbabdi, Saad and Webster, Matthew and Polimeni, Jonathan R. and Van Essen, David C. and Jenkinson, Mark},
    doi = {10.1016/j.neuroimage.2013.04.127},
    issn = {1053-8119},
    journal = {NeuroImage},
    pages = {105-124},
    series = {Mapping the Connectome},
    title = {The minimal preprocessing pipelines for the Human Connectome Project},
    url = {http://www.sciencedirect.com/science/article/pii/S1053811913005053},
    volume = 80,
    year = 2013
}

@article{fs_template,
    author = {Reuter, Martin and Rosas, Herminia Diana and Fischl, Bruce},
    doi = {10.1016/j.neuroimage.2010.07.020},
    journal = {NeuroImage},
    number = 4,
    pages = {1181-1196},
    title = {Highly accurate inverse consistent registration: A robust approach},
    volume = 53,
    year = 2010
}

@article{afni,
    author = {Cox, Robert W. and Hyde, James S.},
    doi = {10.1002/(SICI)1099-1492(199706/08)10:4/5<171::AID-NBM453>3.0.CO;2-L},
    journal = {NMR in Biomedicine},
    number = {4-5},
    pages = {171-178},
    title = {Software tools for analysis and visualization of fMRI data},
    volume = 10,
    year = 1997
}

@article{posse_t2s,
    author = {Posse, Stefan and Wiese, Stefan and Gembris, Daniel and Mathiak, Klaus and Kessler, Christoph and Grosse-Ruyken, Maria-Liisa and Elghahwagi, Barbara and Richards, Todd and Dager, Stephen R. and Kiselev, Valerij G.},
    doi = {10.1002/(SICI)1522-2594(199907)42:1<87::AID-MRM13>3.0.CO;2-O},
    journal = {Magnetic Resonance in Medicine},
    number = 1,
    pages = {87-97},
    title = {Enhancement of {BOLD}-contrast sensitivity by single-shot multi-echo functional {MR} imaging},
    volume = 42,
    year = 1999
}