Skip to content

Quick Start

Run the entire CST assessment pipeline in one command.

The csttool run Command

csttool run --nifti /path/to/dwi.nii.gz --out /path/to/output \
    --subject-id sub-01 --generate-pdf --save-visualizations

csttool run --dicom /path/to/study/ --out /path/to/output \
    --subject-id sub-01 --generate-pdf --save-visualizations

Need sample data?

See recommended datasets for freely available diffusion MRI data.

Runtime & disk space

Typical runtime is 2–10 minutes depending on data size, hardware and chosen parameters. Notably, patch2self denoising and the --perform-motion-correction flag will increase runtime (in the latter case quite substantially). The pipeline generates up to 500 MB of output files (tractograms, scalar maps, reports) per pipeline run.

What Each Step Does

The pipeline runs 6 steps automatically:

Step Name Description
1 Check Validates your environment and dependencies
2 Import Loads NIfTI/DICOM data and extracts gradient information
3 Preprocess Denoises, corrects motion, and segments the brain
4 Track Generates a whole-brain tractogram using CSD-based tractography
5 Extract Isolates left and right CST using atlas-based ROI filtering
6 Metrics Computes FA/MD/RD/AD along tracts and generates reports

Example Output

A successful run looks like this:

======================================================================
PIPELINE COMPLETE
======================================================================
Subject ID:     sub-01
Total time:     3.3 minutes (201 seconds)

Step timing:
  ✓ check       : 0.0s
  ✓ import      : 0.0s
  ✓ preprocess  : 7.7s
  ✓ track       : 72.2s
  ✓ extract     : 110.9s
  ✓ metrics     : 10.1s

✓ All steps completed successfully!
======================================================================

Output Structure

csttool writes a BIDS derivatives dataset at --out. No extra flags required.

output/
├── dataset_description.json
├── participants.tsv
├── participants.json
├── sub-01/
│   ├── dwi/
│   │   ├── sub-01_space-orig_desc-preproc_dwi.nii.gz
│   │   ├── sub-01_space-orig_desc-preproc_dwi.bval
│   │   ├── sub-01_space-orig_desc-preproc_dwi.bvec
│   │   ├── sub-01_space-orig_model-DTI_param-FA_dwimap.nii.gz
│   │   ├── sub-01_space-orig_model-DTI_param-MD_dwimap.nii.gz
│   │   ├── sub-01_space-orig_model-DTI_param-RD_dwimap.nii.gz
│   │   ├── sub-01_space-orig_model-DTI_param-AD_dwimap.nii.gz
│   │   └── tractography/
│   │       ├── sub-01_space-orig_desc-wholebrain_tractogram.trk
│   │       ├── sub-01_space-orig_desc-CSTleft_tractogram.trk
│   │       ├── sub-01_space-orig_desc-CSTright_tractogram.trk
│   │       └── sub-01_space-orig_desc-CSTbilateral_tractogram.trk
│   ├── figures/                        (if --save-visualizations)
│   │   ├── sub-01_stage-preproc_qc-brainmask.png
│   │   ├── sub-01_stage-tracking_qc-tensormaps.png
│   │   ├── sub-01_stage-extraction_qc-registration.png
│   │   └── ...
│   └── reports/
│       ├── sub-01_report.html
│       ├── sub-01_report.pdf           (if --generate-pdf)
│       ├── sub-01_metrics.json
│       └── sub-01_metrics.csv
└── sub-01_pipeline_report.json

See Output formats for a full description of every file.

Key Options

Flag Description
--nifti / --dicom Input format (provide path to data)
--out Output directory
--denoise-method patch2self (default) or nlmeans
--generate-pdf Generate PDF report
--save-visualizations Save QC visualizations at each step
--subject-id Subject ID
--verbose Verbose output

Run csttool run --help for all options.

What's Next?