Outputs of fcdproc
fcdproc generates 3 classes of outcomes:
** Derivative (preprocessed data)**
** Model & data ** includes PCA, GAUSS, FCD-detector models as well as Average Controls data fro both hemispheres
3. ** Projections ** 8 HTML reports per MRI(-) subject, indicating similarity map on top of inflated pial surface.
Layout
Assuming fcdproc is invoked with:
fmriprep <input_dir>/ <output_dir>/ participant analysis_type [OPTIONS]
Preprocessed Data
Preprocessed, or derivative data are written to <output dir>/fcdproc/<subject_label>/
<output_dir>/
fcdproc/
<subject_label>/
data/
std.60.<subject_label>_both.spec/
std.60.<subject_label>_[LR].spec/
std.60.[LR].pial.gii\
std.60.[LR].white.gii\
std.60.[LR].sphere.gii\
std.60.[LR].sphere.reg.gii\
std.60.[LR].smoothwm.gii\
std.60.[LR].inflated.gii\
std.60.[LR].inf_200.gii\
std.60.[LR].aparc.a2009s.annot.niml.dset\
SurfVol_Alnd+orig.(BRIK/HEAD)\
T1_axialize.nii.gz\
dset/
std.60.[LR].curve.1D.dset/
std.60.[LR].sulc.1D.dset/
std.60.[LR].sel_ctx.1D.dset/
std.60.[LR].thickness.fwhm5.1D.dset/
std.60.[LR].w-g.pct.fwhm5.1D.dset/
std.60.[LR].T1_features.1D.dset/
std.60.[LR].T2_features.1D.dset/
std.60.[LR].FL_features.1D.dset/
std.60.[LR].features.1D.dset/
std.60.[LR].features.globalSTD.1D.dset/
std.60.[LR].features.globalSTD.PCA.1D.dset/
std.60.[LR].features.globalSTD.PCA.GAUSS.NITER{1..10}.1D.dset/
std.60.[LR].features.globalSTD.PCA.GAUSS.fwhm10.1D.dset/
Normative Data & FCD-Detector Model
Models and normative data are written to <output dir>/fcdproc/
<output_dir>/
fcdproc/
data/
globalSTD.PCA/
features.globalSTD.PCA.GAUSS.NITER{1..10}/
lh_avg.1D.dset/
rh_avg.1D.dset/
model/
PCA.globalSTD/
GAUSS.NITER{1..10}.features.globalSTD.PCA\
fcd_detector\
Projections
Similarity map of MRI negative patients are written to
<output dir>/fcdproc/<subject_label>/data/dset/projections/
<output_dir>/
fcdproc/
<subject_label>/
data/
dset/
projections/
std.60.[LR].fcd_proj_{0,1}.1D.dset/
[LR]_[inflated/pial_inflated]_surf_[normalized/unnormalized]_projection.html/