Brain MRI Collections

Curated external human brain MRI datasets for research, benchmarking, and education.
This repository aggregates links and light metadata to facilitate access from neuroimaging tools (e.g., DSI Studio).

Important: Always follow the original license / DUA of each dataset and cite the original authors.

Datasets in This Repository

NODDI Lifespan dMRI (ucsf-noddi)
DTI Predicts Mandarin Learning (mit-casl)
Southwest University SLIM (swu-slim)
Stark Cross-Sectional Aging (stark-aging)
Penthera 3T (penthera)
NTU-90 (ntu90)
NKI Rockland Sample (nki-rockland)
Multi-Modal MRI Reproducibility Resource (mmrr)
IXI – Institute of Psychiatry (ixi-iop)
IXI – Hammersmith Hospital (ixi-hh)
IXI – Guy’s Hospital (ixi-guys)
Healthy Brain Network (hbn)
Cambridge Centre for Ageing and Neuroscience (cam-can)

1) NODDI Lifespan dMRI (`ucsf-noddi`)

Diffusion MRI for lifespan white-matter maturation with NODDI (ages 7–63).

Participants: 67 healthy controls; includes test–retest subsets
Modalities:
- DTI: single-shell (b≈1000 s/mm²)
- HARDI: two shells for NODDI modeling
- T1-weighted MRI
License: Governed by the dataset’s UCSF Datashare DUA (see Dryad record)
Citation:
Chang, Y., & Mukherjee, P. (2015). NODDI-PLOS-ONE-Chang et al 2015 [Data set]. Dryad. https://doi.org/10.7272/Q6D798BD
Primary article: PLOS ONE, “Neurite orientation dispersion and density imaging of brain development across the lifespan” (2015).
Source: Dryad – https://doi.org/10.7272/Q6D798BD
Typical uses: ND/OD lifespan curves, age prediction, NODDI vs. DTI comparisons, reliability.

GitHub one-line download (`ucsf-noddi`)

Linux / macOS

curl -s https://api.github.com/repos/data-others/brain/releases/tags/ucsf-noddi | jq -r '.assets[].browser_download_url' | xargs -n1 -P4 curl -LO

Windows PowerShell

(Invoke-RestMethod "https://api.github.com/repos/data-others/brain/releases/tags/ucsf-noddi").assets | ForEach-Object { Invoke-WebRequest $_.browser_download_url -OutFile (Split-Path $_.browser_download_url -Leaf) }

2) DTI Predicts Mandarin Learning (`mit-casl`)

White-matter microstructure and Mandarin second-language learning outcomes.

Participants: De-identified adult learners of Mandarin as a second language
Modalities: DTI (NIfTI DWI + .bval/.bvec), behavioral / demographic CSV
License: CC BY 4.0
Citation: Qi, Z. (2017). DTI Predicts Mandarin Learning [Data set]. Zenodo. https://doi.org/10.5281/zenodo.260007 Qi et al. (2014). J Neurolinguistics, 33, 14–28.
Source: Zenodo – https://doi.org/10.5281/zenodo.260007
Typical uses: Language aptitude biomarkers, FA/tract-based analyses, structure–learning relationships.

GitHub one-line download (`mit-casl`)

Linux / macOS

curl -s https://api.github.com/repos/data-others/brain/releases/tags/mit-casl | jq -r '.assets[].browser_download_url' | xargs -n1 -P4 curl -LO

Windows PowerShell

(Invoke-RestMethod "https://api.github.com/repos/data-others/brain/releases/tags/mit-casl").assets | ForEach-Object { Invoke-WebRequest $_.browser_download_url -OutFile (Split-Path $_.browser_download_url -Leaf) }

3) Southwest University SLIM (`swu-slim`)

Longitudinal multimodal MRI test–retest sample of young adults.

Participants: Healthy young adults, 3.5-year test–retest design
Modalities: Eddy-corrected DWI (SRC/FIB), T1w, demographics; behavioral assessments
License: CC BY-NC
Reference: Liu et al., Sci Data, 2017
Source: CoRR / SLIM project; OneDrive links in release
Typical uses: Reliability studies, longitudinal modeling, diffusion reconstruction benchmarks.

GitHub one-line download (`swu-slim`)

Linux / macOS

curl -s https://api.github.com/repos/data-others/brain/releases/tags/swu-slim | jq -r '.assets[].browser_download_url' | xargs -n1 -P4 curl -LO

Windows PowerShell

(Invoke-RestMethod "https://api.github.com/repos/data-others/brain/releases/tags/swu-slim").assets | ForEach-Object { Invoke-WebRequest $_.browser_download_url -OutFile (Split-Path $_.browser_download_url -Leaf) }

4) Stark Cross-Sectional Aging (`stark-aging`)

Cross-sectional aging dataset with detailed imaging and behavior.

Participants: ~120 adults, ages 18–89
Modalities: Whole-brain DTI, high-res MTL DTI, structural MRI, FreeSurfer segmentations
License: CC BY-NC-SA
Project page: https://www.nitrc.org/projects/stark_aging/
Typical uses: Aging effects on hippocampus, structure–behavior relations, lifespan analyses.

GitHub one-line download (`stark-aging`)

Linux / macOS

curl -s https://api.github.com/repos/data-others/brain/releases/tags/stark-aging | jq -r '.assets[].browser_download_url' | xargs -n1 -P4 curl -LO

Windows PowerShell

(Invoke-RestMethod "https://api.github.com/repos/data-others/brain/releases/tags/stark-aging").assets | ForEach-Object { Invoke-WebRequest $_.browser_download_url -OutFile (Split-Path $_.browser_download_url -Leaf) }

5) Penthera 3T (`penthera`)

Scan–rescan multi-shell DWI on Philips 3T.

Participants: 13 young healthy adults, 6 scans per subject (two sessions × three scans)
Modalities:
- DWI (b=300/1000/2000; 8/32/60 directions; 7×b0; reverse-PE b0)
- T1w MPRAGE
License / Source: Zenodo – https://doi.org/10.5281/zenodo.2602049
Typical uses: Multi-shell modeling, distortion correction, scan–rescan reliability.

GitHub one-line download (`penthera`)

Linux / macOS

curl -s https://api.github.com/repos/data-others/brain/releases/tags/penthera | jq -r '.assets[].browser_download_url' | xargs -n1 -P4 curl -LO

Windows PowerShell

(Invoke-RestMethod "https://api.github.com/repos/data-others/brain/releases/tags/penthera").assets | ForEach-Object { Invoke-WebRequest $_.browser_download_url -OutFile (Split-Path $_.browser_download_url -Leaf) }

6) NTU-90 (`ntu90`)

Classic DSI dataset (NTU-90 atlas source).

Participants: 90 subjects (3 mm DSI), plus 1 subject at 2 mm
Modalities: DSI, 203 directions, b-max 6000 s/mm², NIfTI
License: Non-commercial, educational, research; redistribution requires including original license paragraph
Key refs:
- Yeh & Tseng, NeuroImage 58(1), 91–99 (2011)
- Yeh et al., IEEE TMI 29(9), 1626–1635 (2010)
Typical uses: Tractography evaluation, high-angular sampling methods, atlas work.

GitHub one-line download (`ntu90`)

Linux / macOS

curl -s https://api.github.com/repos/data-others/brain/releases/tags/ntu90 | jq -r '.assets[].browser_download_url' | xargs -n1 -P4 curl -LO

Windows PowerShell

(Invoke-RestMethod "https://api.github.com/repos/data-others/brain/releases/tags/ntu90").assets | ForEach-Object { Invoke-WebRequest $_.browser_download_url -OutFile (Split-Path $_.browser_download_url -Leaf) }

7) NKI Rockland Sample (`nki-rockland`)

Lifespan community neuroimaging with deep phenotyping.

Participants: >1,500, ages 6–85
Data: MRI, extensive phenotypes, psychiatric/cognitive measures, genetics, etc.
License: CC BY-NC; high-dimensional phenotypes require DUA
Reference: Tobe et al., Sci Data 9(1), 300 (2022)
Phenotypes: Access via COINS (requires approved DUA)
Typical uses: Development, psychiatric associations, longitudinal/connectome studies.

GitHub one-line download (`nki-rockland`)

Linux / macOS

curl -s https://api.github.com/repos/data-others/brain/releases/tags/nki-rockland | jq -r '.assets[].browser_download_url' | xargs -n1 -P4 curl -LO

Windows PowerShell

(Invoke-RestMethod "https://api.github.com/repos/data-others/brain/releases/tags/nki-rockland").assets | ForEach-Object { Invoke-WebRequest $_.browser_download_url -OutFile (Split-Path $_.browser_download_url -Leaf) }

8) Multi-Modal MRI Reproducibility Resource (`mmrr`)

Scan–rescan multi-parametric MRI resource (3T, 1-hour protocol).

Participants: 21 healthy volunteers (no neurological history), scan–rescan
Modalities: MPRAGE, FLAIR, DTI, rs-fMRI, B0/B1, ASL, VASO, quantitative T1/T2, MT (NIfTI)
Citation: Landman et al., NeuroImage, 2010. doi:10.1016/j.neuroimage.2010.11.047
License: BIRN Data License (see NITRC project page)
Project page: https://www.nitrc.org/projects/multimodal/
Typical uses: Cross-modality reproducibility, power analysis, QC pipelines.

GitHub one-line download (`mmrr`)

Linux / macOS

curl -s https://api.github.com/repos/data-others/brain/releases/tags/mmrr | jq -r '.assets[].browser_download_url' | xargs -n1 -P4 curl -LO

Windows PowerShell

(Invoke-RestMethod "https://api.github.com/repos/data-others/brain/releases/tags/mmrr").assets | ForEach-Object { Invoke-WebRequest $_.browser_download_url -OutFile (Split-Path $_.browser_download_url -Leaf) }

9) IXI – Institute of Psychiatry (`ixi-iop`)

Site-specific subset of the IXI Dataset.

Participants: Healthy subjects scanned at Institute of Psychiatry (GE 1.5T)
Modalities: T1w, T2w, PD, MRA, DWI (15 directions)
License: CC BY-SA 3.0
Project page: https://brain-development.org/ixi-dataset/
Typical uses: Site-specific baselines, morphometry + diffusion, multi-site comparisons.

GitHub one-line download (`ixi-iop`)

Linux / macOS

curl -s https://api.github.com/repos/data-others/brain/releases/tags/ixi-iop | jq -r '.assets[].browser_download_url' | xargs -n1 -P4 curl -LO

Windows PowerShell

(Invoke-RestMethod "https://api.github.com/repos/data-others/brain/releases/tags/ixi-iop").assets | ForEach-Object { Invoke-WebRequest $_.browser_download_url -OutFile (Split-Path $_.browser_download_url -Leaf) }

10) IXI – Hammersmith Hospital (`ixi-hh`)

Site-specific subset of the IXI Dataset.

Participants: Healthy subjects scanned at Hammersmith Hospital (Philips 3T)
Modalities: T1w, T2w, PD, MRA, DWI (15 directions)
License: CC BY-SA 3.0
Project page: https://brain-development.org/ixi-dataset/
Typical uses: High-field site reference, multi-site harmonization, structural + diffusion baselines.

GitHub one-line download (`ixi-hh`)

Linux / macOS

curl -s https://api.github.com/repos/data-others/brain/releases/tags/ixi-hh | jq -r '.assets[].browser_download_url' | xargs -n1 -P4 curl -LO

Windows PowerShell

(Invoke-RestMethod "https://api.github.com/repos/data-others/brain/releases/tags/ixi-hh").assets | ForEach-Object { Invoke-WebRequest $_.browser_download_url -OutFile (Split-Path $_.browser_download_url -Leaf) }

11) IXI – Guy’s Hospital (`ixi-guys`)

Site-specific subset of the IXI Dataset.

Participants: Healthy subjects scanned at Guy’s Hospital (Philips 1.5T)
Modalities: T1w, T2w, PD, MRA, DWI (15 directions)
License: CC BY-SA 3.0
Project page: https://brain-development.org/ixi-dataset/
Typical uses: 1.5T baselines, site-comparison studies, structural + DWI benchmarks.

GitHub one-line download (`ixi-guys`)

Linux / macOS

curl -s https://api.github.com/repos/data-others/brain/releases/tags/ixi-guys | jq -r '.assets[].browser_download_url' | xargs -n1 -P4 curl -LO

Windows PowerShell

(Invoke-RestMethod "https://api.github.com/repos/data-others/brain/releases/tags/ixi-guys").assets | ForEach-Object { Invoke-WebRequest $_.browser_download_url -OutFile (Split-Path $_.browser_download_url -Leaf) }

12) Healthy Brain Network (`hbn`)

Pediatric / adolescent transdiagnostic biobank (NYC).

Participants: Target 10,000 children/adolescents (ages 5–21); multiple phases (baseline, follow-up, retest)
Data: Multimodal MRI, EEG, psychiatric/behavioral/cognitive measures, lifestyle, genetics, actigraphy, voice/video
License: CC BY-NC; follow HBN citation and DUA guidance
Reference: Alexander et al., Sci Data 4(1), 1–26 (2017)
Diffusion methods (this repo SRC/FIB): Multishell (b=1000, 2000; 64+64 dirs), TOPUP+eddy via TinyFSL + DSI Studio, RDI + GQI reconstruction (see release).
Typical uses: Pediatric mental health, transdiagnostic phenotypes, method development for large pediatric data.

GitHub one-line download (`hbn`)

Linux / macOS

curl -s https://api.github.com/repos/data-others/brain/releases/tags/hbn | jq -r '.assets[].browser_download_url' | xargs -n1 -P4 curl -LO

Windows PowerShell

(Invoke-RestMethod "https://api.github.com/repos/data-others/brain/releases/tags/hbn").assets | ForEach-Object { Invoke-WebRequest $_.browser_download_url -OutFile (Split-Path $_.browser_download_url -Leaf) }

13) Cambridge Centre for Ageing and Neuroscience (`cam-can`)

Large-scale ageing resource with rich cognitive and imaging data.

Participants: Community adults across the adult lifespan
Data (this repo):
- FIB files (GQI reconstruction, ready for DSI Studio tracking)
- SRC files available only to users with Cam-CAN approval (via private link on request)
Diffusion methods: Multishell (b=1000 & 2000; 30+30 dirs), QC-checked b-table, GQI reconstruction (DSI Studio)
License: Follows original CamCAN agreement: non-commercial, open-access publication, acknowledgement of CamCAN
Typical uses: Ageing and connectomics, tract-specific analyses, method development on well-characterized adults.

GitHub one-line download (`cam-can`)

Linux / macOS

curl -s https://api.github.com/repos/data-others/brain/releases/tags/cam-can | jq -r '.assets[].browser_download_url' | xargs -n1 -P4 curl -LO

Windows PowerShell

(Invoke-RestMethod "https://api.github.com/repos/data-others/brain/releases/tags/cam-can").assets | ForEach-Object { Invoke-WebRequest $_.browser_download_url -OutFile (Split-Path $_.browser_download_url -Leaf) }

Quick Start

Use the Releases tab to inspect per-dataset assets (archives, CSVs, FIB/SRC, metadata).
Prefer the one-line commands above to mirror all assets for a given tag.
For DSI Studio:
- Load .nii.gz + .bval + .bvec or SRC/FIB as provided.
- Run reconstruction (e.g., GQI/DTI), set QA/FA thresholds, then track/export.

For reproducible pipelines, keep software versions and reconstruction parameters under version control with your outputs.

Licensing & Attribution

Each release retains its original license / DUA (e.g., CC BY, CC BY-NC, CC BY-NC-SA, BIRN, CC BY-SA, institutional DUA).
Always cite the original dataset and primary articles listed in each section.
If this aggregation is useful, you may add: “Data access was facilitated via the data-others/brain repository.”

Notes & Caveats

Some releases represent partial uploads or derived subsets (e.g., SRC/FIB only).
Phenotypic and other sensitive data may require a Data Use Agreement (DUA) or site-specific approval.
Anonymization / defacing policies differ across datasets; verify before face-related analyses.

Fiber Data Hub

Repositories

Group Templates

Tractography Atlases

Discussion Forum (Reddit)

Discussion Forum (Google)

DSI Studio

Brain MRI Collections

Datasets in This Repository

1) NODDI Lifespan dMRI (`ucsf-noddi`)

GitHub one-line download (`ucsf-noddi`)

2) DTI Predicts Mandarin Learning (`mit-casl`)

GitHub one-line download (`mit-casl`)

3) Southwest University SLIM (`swu-slim`)

GitHub one-line download (`swu-slim`)

4) Stark Cross-Sectional Aging (`stark-aging`)

GitHub one-line download (`stark-aging`)

5) Penthera 3T (`penthera`)

GitHub one-line download (`penthera`)

6) NTU-90 (`ntu90`)

GitHub one-line download (`ntu90`)

7) NKI Rockland Sample (`nki-rockland`)

GitHub one-line download (`nki-rockland`)

GitHub one-line download (`mmrr`)

9) IXI – Institute of Psychiatry (`ixi-iop`)

GitHub one-line download (`ixi-iop`)

10) IXI – Hammersmith Hospital (`ixi-hh`)

GitHub one-line download (`ixi-hh`)

11) IXI – Guy’s Hospital (`ixi-guys`)

GitHub one-line download (`ixi-guys`)

12) Healthy Brain Network (`hbn`)

GitHub one-line download (`hbn`)

13) Cambridge Centre for Ageing and Neuroscience (`cam-can`)

GitHub one-line download (`cam-can`)

Quick Start

Licensing & Attribution

Notes & Caveats

Brain MRI Collections

Datasets in This Repository

1) NODDI Lifespan dMRI (ucsf-noddi)

GitHub one-line download (ucsf-noddi)

2) DTI Predicts Mandarin Learning (mit-casl)

GitHub one-line download (mit-casl)

3) Southwest University SLIM (swu-slim)

GitHub one-line download (swu-slim)

4) Stark Cross-Sectional Aging (stark-aging)

GitHub one-line download (stark-aging)

5) Penthera 3T (penthera)

GitHub one-line download (penthera)

6) NTU-90 (ntu90)

GitHub one-line download (ntu90)

7) NKI Rockland Sample (nki-rockland)

GitHub one-line download (nki-rockland)

8) Multi-Modal MRI Reproducibility Resource (mmrr)

GitHub one-line download (mmrr)

9) IXI – Institute of Psychiatry (ixi-iop)

GitHub one-line download (ixi-iop)

10) IXI – Hammersmith Hospital (ixi-hh)

GitHub one-line download (ixi-hh)

11) IXI – Guy’s Hospital (ixi-guys)

GitHub one-line download (ixi-guys)

12) Healthy Brain Network (hbn)

GitHub one-line download (hbn)

13) Cambridge Centre for Ageing and Neuroscience (cam-can)

GitHub one-line download (cam-can)

Quick Start

Licensing & Attribution

Notes & Caveats

1) NODDI Lifespan dMRI (`ucsf-noddi`)

GitHub one-line download (`ucsf-noddi`)

2) DTI Predicts Mandarin Learning (`mit-casl`)

GitHub one-line download (`mit-casl`)

3) Southwest University SLIM (`swu-slim`)

GitHub one-line download (`swu-slim`)

4) Stark Cross-Sectional Aging (`stark-aging`)

GitHub one-line download (`stark-aging`)

5) Penthera 3T (`penthera`)

GitHub one-line download (`penthera`)

6) NTU-90 (`ntu90`)

GitHub one-line download (`ntu90`)

7) NKI Rockland Sample (`nki-rockland`)

GitHub one-line download (`nki-rockland`)

8) Multi-Modal MRI Reproducibility Resource (`mmrr`)

GitHub one-line download (`mmrr`)

9) IXI – Institute of Psychiatry (`ixi-iop`)

GitHub one-line download (`ixi-iop`)

10) IXI – Hammersmith Hospital (`ixi-hh`)

GitHub one-line download (`ixi-hh`)

11) IXI – Guy’s Hospital (`ixi-guys`)

GitHub one-line download (`ixi-guys`)

12) Healthy Brain Network (`hbn`)

GitHub one-line download (`hbn`)

13) Cambridge Centre for Ageing and Neuroscience (`cam-can`)

GitHub one-line download (`cam-can`)