Follow:
Subscribe:

Official Website: https://dsi-studio.labsolver.org
User Forum: https://groups.google.com/g/dsi-studio

DSI Studio is a standalone software package for diffusion MRI (dMRI) reconstruction, deterministic tractography, and connectome analysis. It is designed to be lightweight and practical for everyday use, while still supporting advanced analysis features for large studies and reproducible pipelines.

DSI Studio supports:

• End-to-end workflows from raw diffusion data to tractography and connectomes
• Interactive analysis in the GUI and automated batch processing via CLI
• Multiple reconstruction models (DTI, GQI, QSDR, and related derivatives)
• Export of tracts, voxelwise maps, region-to-region connectomes, and tract-to-region connectomes

If you use DSI Studio in your work, please cite the relevant methods (see Citations).

• Import diffusion datasets from DICOM or NIfTI
• Reconstruct diffusion models and generate fiber data (.fz)
• Support for common workflows including:
  • DTI (tensor-derived metrics)
  • GQI (model-free ODF reconstruction)
  • QSDR (template space reconstruction for group analysis)

• Deterministic fiber tracking with configurable thresholds and stopping criteria
• ROI-based tractography with interactive editing
• Automated bundle mapping using atlas-based definitions (template-based workflows)
• Tools for tract visualization, clustering, and shape-related analyses

• Generate region-to-region connectivity matrices
• Generate tract-to-region connectomes (atlas-driven representations)
• Export results in common formats for downstream analysis

• Built-in QC utilities (including diffusion signal consistency measures)
• Parameter tracking: output files store the key settings used to generate them
• Structured file formats (.sz, .fz, .tt.gz) designed for reproducibility and portability

• GUI for interactive workflows and visual inspection
• CLI for reproducible pipelines, batch processing, and HPC scripting

DSI Studio is distributed as a standalone executable.

1. Download from: https://dsi-studio.labsolver.org/download.html
2. Choose your platform:
   • dsi_studio_64.exe (Windows)
   • dsi_studio_mac.dmg (macOS)
   • dsi_studio_ubuntu.zip (Linux)
3. Launch the executable.

Notes:

• Windows/Linux: no installation required
• macOS: you may need to grant execution permission (see download page)
• GPU build: requires an NVIDIA GPU and CUDA toolkit installed

• Windows: 64-bit (Windows 10 or newer)
• macOS: Intel or Apple Silicon (macOS 13+)
• Linux: Ubuntu 18.04 or newer (tested on 20.04, 22.04)

• None for CPU builds (standalone executable)
• CUDA toolkit required for GPU builds

• CPU: 4+ cores
• RAM: 8+ GB (more recommended for large datasets)
• GPU: NVIDIA GPU recommended for GPU builds and large-scale tracking

1. Launch dsi_studio
2. Go to the Fiber Data tab
3. Select a .fz file and click Open to enter the tracking window

4. Click Fiber Tracking to start tractography

5. Visualize and export results from the top menu

Typical runtime: ~1–3 minutes on a standard desktop (depends on settings and dataset)

You can also browse and download preprocessed fiber datasets from:
https://brain.labsolver.org

A typical GUI workflow is:

1. Import diffusion data: File → Open → DICOM/NIfTI
2. Convert to .sz: use Step T1 conversion
3. Reconstruct to .fz: choose a reconstruction method (e.g., DTI, GQI)
4. Run tractography: use ROIs, atlas-based bundles, or your own definitions
5. Export: tracts, maps, connectomes, and summary reports

Documentation entry point: https://dsi-studio.labsolver.org

DSI Studio supports CLI scripting for reproducible batch processing.

• CLI documentation (T1): https://dsi-studio.labsolver.org/doc/cli_t1.html
• CLI documentation (general): https://dsi-studio.labsolver.org/doc/

Example use cases:

• Batch reconstruction across a cohort
• Automated tractography with standardized parameters
• Connectome export for statistical pipelines

Common file types you may encounter:

• .sz : converted source data container (input stage)
• .fz : reconstructed fiber data used for tracking and analysis
• .tt.gz : tractography output
• .nii.gz : exported voxelwise maps (e.g., anisotropy measures)
• .connectivity.mat / matrices : connectome outputs (depending on export options)

DSI Studio supports reproducible analysis by:

• Saving parameters alongside output files
• Enabling the same operations through both GUI and CLI
• Standardizing intermediate files (.sz, .fz) to reduce variability across runs

• Tutorial videos: https://practicum.labsolver.org
• User forum (bugs, suggestions, troubleshooting): https://groups.google.com/g/dsi-studio
• Manual: https://dsi-studio.labsolver.org/manual
• Issue tracker: https://github.com/frankyeh/DSI-Studio/issues

When reporting an issue, please include:

• OS and version
• DSI Studio version (release tag)
• Steps to reproduce
• Screenshots/logs if available

This repository hosts the source code for DSI Studio. Most users will use the standalone binaries from the official site. If you would like to contribute:

• Use GitHub issues for bug reports and feature requests
• Keep changes focused and well-documented
• Include a short rationale and (when possible) minimal test cases

Please refer to the repository license and the official website for current licensing and usage terms.

Please cite the methods you used (select only those applied to your study). Citation names are formatted consistently as Yeh, FC.

DSI Studio platform and Fiber Data Hub (2025)

Yeh, FC. DSI Studio: an integrated tractography platform and fiber data hub for accelerating brain research. Nature Methods. 2025 Aug;22(8):1617-1619. doi:10.1038/s41592-025-02762-8.

Population-based atlas and tract-to-region connectome (2022)

Yeh, FC. Population-based tract-to-region connectome of the human brain and its hierarchical topology. Nature Communications. 2022 Aug 22;13(1):1-3.

Shape analysis (2020)

Yeh, FC. Shape Analysis of the Human Association Pathways. NeuroImage. 2020.

Augmented fiber tracking (2020)

Yeh, FC. Shape Analysis of the Human Association Pathways. NeuroImage. 2020.

Differential tractography (2019)

Yeh, FC, et al. Differential tractography as a track-based biomarker for neuronal injury. NeuroImage. 2019;202:116131.

Topology-informed pruning (TIP, 2019)

Yeh, FC, et al. Automatic Removal of False Connections in Diffusion MRI Tractography Using Topology-Informed Pruning (TIP). Neurotherapeutics. 2019.

Connectometry (2016)

Yeh, FC, Badre D, Verstynen T. Connectometry: A statistical approach harnessing the analytical potential of the local connectome. NeuroImage. 2016;125:162-171.

Restricted diffusion imaging (RDI, 2016)

Yeh, FC, Liu L, Hitchens TK, Wu YL. Mapping Immune Cell Infiltration Using Restricted Diffusion MRI. Magnetic Resonance in Medicine. 2016.

Local connectome fingerprint (LCF, 2016)

Yeh, FC, et al. Quantifying differences and similarities in whole-brain white matter architecture using local connectome fingerprints. PLoS Computational Biology. 2016;12(11):e1005203.

Individual connectometry (2013)

Yeh, FC, et al. Diffusion MRI connectometry automatically reveals affected fiber pathways in individuals with chronic stroke. NeuroImage: Clinical. 2013;2:912-921.

Generalized deterministic tracking (2013)

Yeh, FC, et al. Deterministic diffusion fiber tracking improved by quantitative anisotropy. PLoS ONE. 2013;8(11):e80713. doi:10.1371/journal.pone.0080713.

QSDR / NTU-90 atlas (2011)

Yeh, FC, Tseng WI. NTU-90: a high angular resolution brain atlas constructed by q-space diffeomorphic reconstruction. NeuroImage. 2011;58(1):91-99.

GQI (2010)

Yeh, FC, Wedeen VJ, Tseng WI. Generalized q-sampling imaging. IEEE Transactions on Medical Imaging. 2010;29(9):1626-1635.