Derivatives
ABCD-BIDS Pipeline
Derivatives generated from the DCAN Labs ABCD-BIDS MRI (version 0.1.4) processing pipeline (Sturgeon et al., 2023) are provided, including Human Connectome Project (HCP) Minimal Preprocessing Pipelines-style data in both volume and surface spaces as well as freesurfer-5.3.0-HCP segmentation statistics and surface morphometrics (Glasser et al., 2013).
The BIDS structure of the output derivatives will look like the following (see below for descriptions of anat/ and func/ file contents). Note that the .html file is the executive summary output used to perform quality control, with the associated images included in the file stored under img/ (file contents not listed below for readability). The file contents of each folder is detailed below - see full details in the ABCD-BIDS Pipeline documentation
|__ abcc/
|__derivatives/
|__ abcd-hcp-pipeline_v0.1.4/
|__ sub-<label>/
|__ ses-<label>/
|__ anat/
|__ func/
|__ img/*
|__ SUBSES.htmlDetails: anat/
The derivative files included under the anat/ directory are displayed below (BIDS entities sub-<label>_ses-<label> replaced with SUBSES for readability). Please reference the table below for further information on each file type based on the number to the right of the file as listed in the file tree visual:
|_ anat/
|__ SUBSES_<T1w|T2w>_space-MNI_brain.nii.gz [1]
|__ SUBSES_T1w_space-MNI_desc-wmparc_dseg.nii.gz [2]
|__ SUBSES_atlas-MNI_space-fsLR32k_desc-smoothed_myelinmap.dscalar.nii [3A]
|__ SUBSES_atlas-MNI_space-fsLR32k_myelinmap.dscalar.nii [3B] _
|__ SUBSES_hemi-<L|R>_space-MNI_mesh-fsLR164k_midthickness.surf.gii |
|__ SUBSES_hemi-<L|R>_space-MNI_mesh-fsLR32k_midthickness.surf.gii |
|__ SUBSES_hemi-<L|R>_space-MNI_mesh-native_midthickness.surf.gii | [4]
|__ SUBSES_hemi-<L|R>_space-T1w_mesh-fsLR32k_midthickness.surf.gii |
|__ SUBSES_hemi-<L|R>_space-T1w_mesh-native_midthickness.surf.gii _|
|__ SUBSES_space-ACPC_dseg.nii.gz [5] _
|__ SUBSES_space-fsLR32k_curv.dscalar.nii |
|__ SUBSES_space-fsLR32k_sulc.dscalar.nii | [6]
|__ SUBSES_space-fsLR32k_thickness.dscalar.nii _|
|__ SUBSES_space-fsLR32k_sulc.pscalar.nii [7]| Key | File Information |
|---|---|
| 1 | T1w & T2w brain & head images in MNI space |
| 2 | White matter segmentation in MNI space |
| 3A/3B | Smoothed & unsmoothed myelin map |
| 4 | L/R midthickness in MNI & native space, each with 32k, 146k, & native mesh |
| 5 | Discrete segmentation (native volume space) |
| 6 | Dense curvature, sulcal depth, & cortical thickness |
| 7 | Parcellated dense subject sulcal depth |
Details: func/
The derivative files included under the func/ directory are displayed below (BIDS entities sub-<label>_ses-<label> replaced with SUBSES for readability). Please reference the table below for further information on each file type based on the number to the right of the file as listed in the file tree visual:
|__ func/
|__ SUBSES_task-<TASK>_bold_desc-filtered_timeseries.dtseries.nii
|__ SUBSES_task-<TASK>_bold_atlas-<ATLAS>_desc-filtered_timeseries.ptseries.nii
|__ SUBSES_task-<TASK>_desc-filtered_motion_mask.mat
|__ SUBSES_task-<TASK>_desc-filteredwithoutliers_motion_mask.mat
|__ SUBSES_task-<TASK>_run-<label>_bold_timeseries.dtseries.nii
|__ SUBSES_task-<TASK>_run-<label>_desc-filtered_motion.tsv
|__ SUBSES_task-<TASK>_run-<label>_desc-filteredincludingFD_motion.tsv
|__ SUBSES_task-<TASK>_run-<label>_desc-includingFD_motion.tsv
|__ SUBSES_task-<TASK>_run-<label>_motion.tsv
|__ SUBSES_task-<TASK>_run-<label>_space-MNI_bold.nii.gz
|__ SUBSES_task-rest_bold_atlas-<ATLAS>_desc-filtered_timeseries_thresh-<THRESH>_censor.txt
|__ SUBSES_task-rest_bold_atlas-<ATLAS>_desc-filtered_timeseries_thresh-<THRESH>_connectivity.pconn.niiThe dtseries.nii files represent dense time series data in CIFTI format at different processing stages. Individual run files (*_run-#_bold_timeseries.dtseries.nii) contain preprocessed BOLD signal mapped to atlas space for each functional run. Concatenated filtered files (*_bold_desc-filtered_timeseries.dtseries.nii) represent time series data combined across all runs for a given task after application of nuisance regression and bandpass filtering steps performed by DCANBOLDProc.Each dtseries.nii is accompanied by corresponding motion censoring files (motion.mat), which provide framewise displacement (FD) information and temporal masks used to identify and exclude high-motion frames during analysis (Power et al., 2014).The motion files contain a 1x101 MATLAB cell of MATLAB structs where each struct is the censoring info at a given FD threshold (0 to 0.3 millimeters in steps of 0.01 millimeters). Dense time series are available for each task:
- Values for
<TASK>include:MIDSSTnbackrest
In addition to these surface-based files, volumetric data prior to surface registration is also provided in the file SUBSES_task-<TASK>_run-<label>_space-MNI_bold.nii.gz.
The ptseries.nii files are parcellated timeseries extracted based on atlas templates listed below - all parcellations additionally include 19 individualized subcortical parcellations with ROIs taken from the standard HCP 91k-grayordinate CIFTI template.
- Values for
<ATLAS>BIDS entity and the atlas template it corresponds to are as follows:Gordon2014FreeSurferSubcortical- Gordon’s 333 ROI template, parcellated using boundary detection on rs-fMRI data from 120 young adults with 14 minutes of data collected on average (Gordon et al., 2016)HCP2016FreeSurferSubcortical- HCP’s 360 ROI template, parcellated from mutli-modal task, rest, and diffusion MRI data on 210 young adults (Glasser et al., 2016)Markov2012FreeSurferSubcortical- Markov’s cortical parcellation comprising 91 areas using retrograde tracer injections in rhesus macaquesPower2011FreeSurferSubcortical- Power’s 264 ROI template, parcellated from meta-analysis of task and resting-fMRI data across multiple datasets (Power et al., 2011)Yeo2011FreeSurferSubcortical- Yeo’s 118 ROI template, parcellated from rs-fMRI data collected on 1000 subjects (Yeo et al., 2011)
Finally, the parcellated connectivity matrices, or ‘pconns’ (*.pconn.nii), generated post-processing using the DCAN-Labs/cifti-connectivity tools. Pconns were generated by calculating the lag-zero pearson’s correlation coefficient between each ROI pair as defined by each of the parcellation schemes listed above, resulting in ROIxROI matrices of scalar values reporting the pairwise correlation between ROIs. Pconns generated from both 5 and 10 minutes of low-motion (FD<0.2 mm) data (as available) are provided. Note that the inverse hyperbolic tangent (a variance stabilization procedure) was applied to the correlations: z = arctanh(r). The maximum value displayed is 7.254329. Applying the hyperbolic tangent will recover the pearson’s correlation: r = tanh(z).
- Values for
<THRESH>include:fd0p2mm_censor-10min_conndata-networkfd0p2mm_censor-5min_conndata-networkfd0p2mm_censor-belowthresh_conndata-network
Details: SUBSES.html
Quality control (QC) procedures for structural and functional derivatives involve manual visual inspection to detect image artifacts. The ExecutiveSummary stage of the ABCD-BIDS pipeline produces an HTML visual quality control page that displays a BrainSprite viewer of the T1w and T2w segmentation, an overlay of the atlas registration on each single band reference created by FSL’s slicer, and a visualization of the movement and grayordinate time series for each fMRI run pre- and post-regression. The Executive Summary is intended for use in performing quality control on each subject.
For instruction on how to judge the quality of images contained in the Executive Summary, please see the BrainSwipes Tutorials (note: you must create a free account first). Please also see the section on BrainSwipes under Quality control procedures to learn about how this utility is being used to provide QC ratings based on Executive Summaries for all participant data.
QSIPrep
The QSIPrep pipeline is used for preprocessing the ABCD diffusion-weighted MRI (dMRI) data. Preprocessing includes head motion correction, susceptibility distortion correction, MP-PCA denoising, coregistration to T1w images, ANTS spatial normalization, and tissue segmentation. To see an explanation of the QSIPrep pipeline derivatives, please visit the section on Outputs of QSIPrep in the QSIPrep documentation.
|__ abcc/
|__derivatives/
|__ qsiprep/
|__ sub-<label>/
|__ ses-<label>/
|__ anat/
| |__ SUBSES_from-T1wACPC_to-T1wNative_mode-image_xfm.mat
| |__ SUBSES_from-T1wNative_to-T1wACPC_mode-image_xfm.mat
| |__ SUBSES_from-MNI152NLin2009cAsym_to-T1w_mode-image_xfm.h5
| |__ SUBSES_from-T1w_to-MNI152NLin2009cAsym_mode-image_xfm.h5
| |__ SUBSES_from-orig_to-T1w_mode-image_xfm.txt
| |__ SUBSES_rec-normalized_from-orig_to-T1w_mode-image_xfm.txt
| |__ SUBSES_rec-normalized_from-orig_to-T1w_mode-image_xfm.mat
| |__ SUBSES_rec-normalized_run-<label>_from-orig_to-T1w_mode-image_xfm.mat
| |__ SUBSES_run-<label>_from-orig_to-T1w_mode-image_xfm.mat
| |__ SUBSES_dseg.nii.gz
| |__ SUBSES_desc-brain_mask.nii.gz
| |__ SUBSES_desc-preproc_T1w.nii.gz
| |__ SUBSES_desc-aseg_dseg.nii.gz
|
|__ dwi/
| |__ SUBSES_confounds.tsv
| |__ SUBSES_desc-ImageQC_dwi.csv
| |__ SUBSES_desc-SliceQC_dwi.json
| |__ SUBSES_dwiqc.json
| |__ SUBSES_space-T1w_desc-brain_mask.nii.gz
| |__ SUBSES_space-T1w_desc-eddy_cnr.nii.gz
| |__ SUBSES_space-T1w_desc-preproc_dwi.b
| |__ SUBSES_space-T1w_desc-preproc_dwi.bval
| |__ SUBSES_space-T1w_desc-preproc_dwi.bvec
| |__ SUBSES_space-T1w_desc-preproc_dwi.nii.gz
| |__ SUBSES_space-T1w_desc-preproc_dwi.txt
| |__ SUBSES_space-T1w_dwiref.nii.gz
|
|__ figures/
|__ SUBSES.htmlQSIRecon
The QSIRecon pipeline is used for reconstructing and modeling preprocessed diffusion-weighted MRI (dMRI) data from QSIPrep outputs. QSIRecon applies various reconstruction algorithms to generate microstructural tissue maps, tractography, and connectivity matrices from the preprocessed diffusion data. For this release, QSIRecon derivatives were generated using five reconstruction specifications, each providing different modeling approaches and outputs:
- DIPYDKI - Diffusion kurtosis imaging (DKI) using DIPY, providing metrics such as mean kurtosis, radial kurtosis, and axial kurtosis. Diffusion tensor metrics derived as part of the DKI fit are also included here
- DSIStudioGQI - Q-space imaging using DSI Studio’s Generalized Q-sampling Imaging (GQI) for orientation distribution functions and quantitative anisotropy. Diffusion tensor metrics computed across all b-values are included within this specification
- MSMTAutoTrack - Multi-shell multi-tissue tractography using DSI Studio’s AutoTrack algorithm on MSMT FODs generated by MRtrix3
- TORTOISE_model-MAPMRI - Mean Apparent Propagator MRI (MAP-MRI) modeling using TORTOISE for non-Gaussian diffusion characterization. Diffusion tensor metrics computed for b-values below 1200 are also included within this specification.
- wmNODDI - White matter neurite orientation dispersion and density imaging (NODDI) modeled using AMICO for white matter microstructure analysis
To see a detailed explanation of the QSIRecon pipeline derivatives and reconstruction specifications, please visit the section on Outputs of QSIRecon in the QSIRecon documentation.
|__ abcc/
|__derivatives/
|__ qsirecon-<RECON_SPEC>/
|__ sub-<label>/
|__ ses-<label>/
├── dwi
│ ├── # --- MNI152NLin2009cAsym space outputs ---
│ ├── sub-<SUBJECT_ID>_ses-<SESSION_ID>_space-MNI152NLin2009cAsym_model-<MODEL>_param-<PARAM>_dwimap.json
│ ├── sub-<SUBJECT_ID>_ses-<SESSION_ID>_space-MNI152NLin2009cAsym_model-<MODEL>_param-<PARAM>_dwimap.nii.gz
│ ├── ...
│
│ ├── # --- T1w space outputs ---
│ ├── sub-<SUBJECT_ID>_ses-<SESSION_ID>_space-T1w_model-<MODEL>_param-<PARAM>_dwimap.json
│ ├── sub-<SUBJECT_ID>_ses-<SESSION_ID>_space-T1w_model-<MODEL>_param-<PARAM>_dwimap.nii.gz
│ ├── sub-<SUBJECT_ID>_ses-<SESSION_ID>_space-T1w_bundles-<TRACTOGRAPHY_METHOD>_scalarstats.tsv
│ ├── ...
│
│ └── (other diffusion models, scalar maps, or tractography, depending on RECON_SPEC)
│
├── figures
│ ├── sub-<SUBJECT_ID>_ses-<SESSION_ID>_desc-about_T1w.html
│ └── sub-<SUBJECT_ID>_ses-<SESSION_ID>_desc-summary_T1w.html
│
└── sub-<SUBJECT_ID>_ses-<SESSION_ID>.html