TAIR Multimodal Imaging Dataset for Cerebrovascular Disease
CTA and MRA, plus paired X-ray fluoroscopy video, collected during routine critical care at multiple centers. Released by Kitware and the University of North Carolina at Chapel Hill through the ARPA-H AIR program for cerebrovascular disease research.
Open the source page ↗ See it in the directory
At a glance
- What it is
- CTA and MRA, plus paired X-ray fluoroscopy video, collected during routine critical care at multiple centers. Released by Kitware and the University of North Carolina at Chapel Hill through the ARPA-H AIR program for cerebrovascular disease research.
- Area
- Neuro
- Modality
- CT, MR, XA
- Format
- NIfTI, MP4
- Size
- First release (3 Nov 2026): 500 CTA/MRA studies and 150 fluoroscopy datasets, in one zip of about 70 GB. The clinical-data page lists a planned final release on 30 Dec 2027 of 2,000 CTA/MRA studies and 1,250 fluoroscopy datasets. The 4 Nov 2026 announcement describes a full collection of 2,500 CTA/MRA datasets and 1,500 fluoroscopy recordings. Segmented, centerlined, and annotated models (VTK, NIfTI, OpenUSD, and JSON; the clinical-data page lists 100 CTA/MRA models, 50 fluoroscopy datasets, and one population shape model) are announced for December 2026 and are not in the first zip.
- Normals included
- Partial (mixed or unlabeled)
- License
- CC BY 4.0
- Commercial use
- Allowed by the license, with its attribution terms
- Access
- Open download, no account
- Status
- Verified: source page reviewed by Hounsfield Lab
- Added
- 2026-11-11
- Last checked
- 2026-11-11
- Source
- tair.kitware.com/clinical-data/
How to access
Install the AWS CLI. No AWS account is needed. Download the zip from the clinical-data page, or run: aws s3 cp s3://multimodal-imaging-for-cerebrovascular-disease/tair-project-data.zip . --no-sign-request . The same bucket can be listed by numeric patient prefix. Allow about 70 GB of disk space. Kitware and UNC-Chapel Hill announced the release on 4 Nov 2026.
Limitations
NIfTI and MP4, not DICOM, so DICOM header metadata is not in this release. CTA and MRA are NIfTI. Fluoroscopy is MP4. The publisher states that the scans were defaced, PHI was removed from metadata, and burned-in text was removed from the fluoroscopy. This is the first of four planned releases, so the counts will change. The first release has no labels or segmentations. The source pages do not say whether any study is free of cerebrovascular disease, so this is not listed as a normal-control set. The clinical-data page describes a permissive license for academic, government, and commercial use. The 4 Nov 2026 Kitware announcement names that license as CC BY 4.0 and says citation is required. Relevant to head and neck CTA work on large-vessel occlusion, aneurysm, and stenosis.
Citation required?
Yes. Lee YZ, de Siqueira Bucci AC, Ho J, Paniagua B, Vicory J, Ravikumar S, Nagler M, Webb JB, Clipp RB. TAIR: Multimodal Imaging Dataset for Cerebrovascular Disease. 2026, v1.0, AWS Registry of Open Data.
Related open-source tools
- HD-BET: Deep-learning brain extraction (skull stripping) for MRI
- FastSurfer: Deep-learning whole-brain segmentation and cortical surface reconstruction, a fast alternative to FreeSurfer recon-all
- FreeSurfer: Neuroimaging analysis and visualization suite
- BrainChop: In-browser 3D MRI rendering and segmentation
- Neuro-VL: Neuroimaging vision-language model (Neuro-VL-8B-Thinking, Qwen3-8B language model with an HLIP visual encoder) that reads complete MRI and CT exams, converses about findings, drafts reports with reasoning traces and compares serial exams; trained on nearly 1 million clinical neuroimaging studies from 450,000 patients; released with the NeuroBench benchmark
- dcm2niix: Converts DICOM to NIfTI with JSON sidecars; the standard converter in many brain MRI and BIDS pipelines
- TorchIO: Medical image loading, preprocessing and augmentation for PyTorch
More neuro datasets
- BrainMetShare: MR · Stanford AIMI research use agreement (non-commercial; commercial license sold separately)
- Burdenko-GBM-Progression: MR, CT · NIH Controlled Data Access Policy (images); CC BY 4.0 (clinical and genomic CSV)
- CFB-GBM: MR, CT · CC BY 4.0
- CPAISD: CT · CC BY 4.0
- CQ500: CT · CC BY-NC-SA 4.0
- CT-ICH: CT · PhysioNet Restricted Health Data License 1.5.0
Hounsfield Lab links to the source and does not host images. The license and access summary reflects the source page on the last-checked date; terms can change, so read the source before you download. Not legal advice.