DPABI: a toolbox for Data Processing & Analysis for Brain Imaging
DPABI is a GNU/GPL* toolbox for Data Processing & Analysis of Brain Imaging, evolved from DPARSF (Data Processing Assistant for Resting-State fMRI). Please refer to The R-fMRI Course to know more about how to use this toolbox. Add with subfolders for DPABI in MATLAB's path setting and enter "dpabi" in the command window to enjoy this powerful toolbox.
New features of DPABI_V2.1_160415:
1. DPARSF V4.1_160415 released.
1.1. Fixed a bug in DPARSF Basic Edition. The bug is that the white matter signal is always removed in nuisance regression (only exist in the Basic Edition). Thanks to the report of Liviu Badea.
1.2. DPARSF Advanced Edition: Add an option of “Add mean back” for nuisance regression. The mean will be added back to the residual after nuisance regression. This is useful for circumstances of ICA or task-based analysis.
1.3. DPARSF Advanced Edition: Re-run with global signal regression (DPARSFA_RerunWithGSR). Fixed a bug when “Remove first X time points” was defined, the number of time points will be adjusted accordingly now. Thanks to the report of Hua-Sheng Liu.
1.4. DPARSF Advanced Edition: Add a slice timing batch mode for MultiBand data. Users could specify a text timing file for a given participant in SliceOrderInfo.tsv. Please see http://wiki.rfmri.org/SliceTiming for more details.
2. DPABI Results Organizer (under “The R-fMRI Maps Project”): also save the text version of ROI signals.
3. Dual Regression added (under Utilities). Define a map, then regress the map on the 4D data for a participant, thus get a time series. Variance-normalize the time series, and then regress on the 4D data, thus get the dual regression map.
4. DPABI Image Calculator (under “Utilities”). Syntax changed, now includes: g1.*To4D((i1>2.3),100) Make a mask (threshold at 2.3 on i1) and then apply to each image in group 1 (group 1 has 100 images).
5. Donsenbach 160 ROIs were merged into a single mask file (Dosenbach_Science_160ROIs_Radius5_Mask.nii).
6. Fixed a bug in y_GroupAnalysis_Image: CovVolume read error.
7. Add "MultiSelect" Mode when user selected "Add Image" for several GUIs.
8. DPABI Image Calculator (under Utilities): Fixed a bug when removing image or directory. Now, when you remove the item, the identifier will be re-ordered.
9. DPABI Viewer: Added a colorbar mode for DPABI_VIEW, when you add "+" or "-" flag at the end in "Add Overlay's Colorbar" entry, DPABI_VIEW will use full colormap to display the overlay.
New features of DPABI_V2.0_151201 (together with DPARSF_V4.0_151201):
1. Compatible with MATLAB 2014b and later versions.
2. Process the data both with and without global signal regression (GSR). Check “Nuisance regressors setting” -> “Both with & without GSR”. Alternatively, you can call DPARSFA_RerunWithGSR.m. E.g., DPARSFA_RerunWithGSR(DPARSFACfg.mat); where DPARSFACfg.mat stores the previous parameters without GSR.
3. The processing steps are affixed to Results directories. The R-fMRI calculation parameters are also written to the header of the result files.
4. V4 processing parameter template is added. No smoothing before R-fMRI measure calculation (except for VMHC). This is used for comparing across studies and accumulate processed data.
5. DPABI Statistical Analysis. Add multiple comparison test after ANOVA, e.g., 'tukey-kramer' or 'hsd', 'lsd', 'dunn-sidak', 'bonferroni’ or ‘scheffe' procedures.
6. DPABI_VIEW: compatible with BrainNet Viewer 1.5.
7. Fixed a "File too small" bug when .hdr/.img files are used.
8. Fixed a bug in y_Standardize.m: error when multiple files are defined.
9. Fixed a bug in DPABI Image Calculator: error in standard deviation calculation along the 4th dimension.
10. Results Organizer module: with this module, the users could organize the intermediate files for future processing with DPABI. In addition, the results could be organized for future use, and to be accumulated for the future R-fMRI maps project.
New features of DPABI_V1.3_150710:
1. SPM12 Compatible.
2. DPARSF_V3.2_150710 released.
3. DPARSF for Rat data released.
The Rat module is based on a Rat T2 template generated by Dr. Adam J. Schwarz et al. Please cite this paper when appropriate: Schwarz, A.J., Danckaert, A., Reese, T., Gozzi, A., Paxinos, G., Watson, C., Merlo-Pich, E.V., Bifone, A., 2006. A stereotaxic MRI template set for the rat brain with tissue class distribution maps and co-registered anatomical atlas: application to pharmacological MRI. Neuroimage 32, 538-550. (A T1 template was included as well. It's generated by normalizing 50 rats (two scans at PND45 or PND60) to that T2 template and then averaging (by Dr. Chao-Gan Yan)).
4. Fixed a bug in generating Voxel Specific Head Motion: missing gmdmp.
5. Fixed a bug in Group Analysis: when CovVolume is not defined.
6. Fixed a bug when calling BrainNet Viewer in DPABI_VIEW.
7. Fixed a bug in Standardization: ‘/‘ is not defined in Windows.
8. Fixed a bug in Image Calculator: output will be split as the input files when calculating group images.
New features of DPABI_V1.2_141101:
1. DPARSF V3.1 Basic Edition: Fixed a bug of missing DPARSF_run.
2. DPARSF V3.1: Fixed a bug that can not find ROI templates.
3. DICOM Sorter: Fixed a bug - “Add All” button doesn’t work.
4. DPABI Viewer: Fixed a bug when execute GRF or AlphaSim correction.
5. DPABI ROI Signal Extractor: New module added to DPABI->Utilities.
7. Test-Retest Reliability: Intraclass Correlation Coefficient (ICC). New module added to DPABI->Utilities. Three ICC algorithms are supported: ANOVA Model, ReML Model and Linear Mixed Models. The algorithms are based on Dr. Xi-Nian Zuo and his colleague’s work. Please cite Dr. Zuo’s work as detailed in each function.
New features of DPABI_V1.1_140827:
1. New modules in DPABI Utilities:
1.1. Multiple T1 Images Averager. If you have multiple T1 runs for each subject, this module will coregister them and make a mean T1 image to put to "T1Img" for following analyses.
2. Bugs Fixed.
2.1. DPABI Viewer: Cluster report doesn't work correctly after GRF correction. Thanks for Vincent's report!
3. DPABI now can check the latest version and pop up a notice.
DPABI includes the following components.
1. DPARSF 3.0 Advanced Edition.
New features in DPARSF 3.0 Advanced Edition.
1.1. Quality control. Integrated GUI for QCing the functional and structural images, users can give ratings and comments during the step of interactive reorientation.
1.2. Automask generation. For checking EPI coverage and generating group mask, the automasks (as in AFNI) will be generated based on EPI images.
1.3. Brain extraction (Skullstrip). This step can improve the coregistration between functional and structural images. Most registration issues of previous DPARSF versions can be solved by including this step. For Linux and Mac users: Need to install FSL. For Windows users: Thanks to Chris Rorden's compiled version of bet in MRIcroN, the modified version can work on NIfTI images directly.
1.4. Nuisance Regression. 1) Masks can be generated based on segmentation or SPM apriori masks; 2) Methods can be mean or CompCor [Note: for CompCor, detrend (demean) and variance normalization will be applied before PCA, according to (Behzadi et al., 2007)]; 3) Global Signal can be extracted based on Automasks.
2. DPARSF 3.0 Basic Edition.
2.1. DPARSF Basic Edition now is using the engine of DPARSF Advanced Edition.
2.2. Nuisance Regression (in MNI space) is placed before filtering, according to (Hallquist et al., 2013).
3. DPARSF for Monkey data.
3.1. The monkey module is based on Rhesus Macaque Atlases for functional and structural imaging studies generated by Wisconsin ADRC Imaging Core. Please cite their papers when appropriate: (McLaren et al., 2010; McLaren et al., 2009).
3.2. Of note, the origin of monkey atlas is different from human MNI atlas. Please make sure the correct origins are set at the steps of "reorienting Fun*" and "reorienting T1*".
4. Preprocessing for task fMRI.
Task fMRI data can be preprocessed via DPABI-DPARSF.
VBM analyses can be performed via DPABI-DPARSF.
6. Quality Control.
6.1. QC Raw T1 images.
6.2. QC Raw functional images.
6.3. QC normalization effects. 1) QC on the pictures for checking spatial normalization. 2) Dynamically checking normalized T1, gray matter and functional images.
6.4. Thresholding QC scores and removing un-qualified subjects.
6.5. Generating Group masks based on normalized Automasks of each subject.
6.6. Thresholding EPI coverage.
6.7. Head motion report.
6.8. Thresholding head motion.
7. Standardization. Perform the following standardization according to (Yan et al., 2013).
7.1. Mean Regression
7.2. Mean Regression & SD Division
7.3. Mean Regression & Log SD Regression
7.4. Z - Standardization
7.5. Mean Division
7.6. Mean Subtraction
7.7. Median-IQR Standardization
7.9. Quantile Standardization
7.10. Gaussian Fit
8. Statistical Analysis.
Smoothness estimation based on the 4D residual is built in regression function – smoothness is written to the NIfTI headers automatically. For AlphaSim and GRF multiple comparison correction, only using smooth kernel applied in preprocessing is NOT sufficient, please use the estimated smoothness instead.
The DPABI_VIEW is based on spm_orthviews, but powered with convenient functions. Please try it out!
Utilities including DICOM Sorter, Image Calculator and Image Reslicer.
Behzadi, Y., Restom, K., Liau, J., Liu, T.T., 2007. A component based noise correction method (CompCor) for BOLD and perfusion based fMRI. Neuroimage 37, 90-101.
Hallquist, M.N., Hwang, K., Luna, B., 2013. The nuisance of nuisance regression: spectral misspecification in a common approach to resting-state fMRI preprocessing reintroduces noise and obscures functional connectivity. Neuroimage 82, 208-225.
McLaren, D.G., Kosmatka, K.J., Kastman, E.K., Bendlin, B.B., Johnson, S.C., 2010. Rhesus macaque brain morphometry: a methodological comparison of voxel-wise approaches. Methods 50, 157-165.
McLaren, D.G., Kosmatka, K.J., Oakes, T.R., Kroenke, C.D., Kohama, S.G., Matochik, J.A., Ingram, D.K., Johnson, S.C., 2009. A population-average MRI-based atlas collection of the rhesus macaque. Neuroimage 45, 52-59.
Yan, C.G., Craddock, R.C., Zuo, X.N., Zang, Y.F., Milham, M.P., 2013. Standardizing the intrinsic brain: towards robust measurement of inter-individual variation in 1000 functional connectomes. Neuroimage 80, 246-262.
*Some codes from REST V1.0 and V1.1 that are not complying with GNU GPL have been re-written here to comply GNU GPL. Some GNU GPL functions of REST V1.2 ~ V1.8 have been modified to be integrated here under GNU GPL.
Old Versions: V1.0 V1.1 V1.2 V1.3 V2.0