UCSD-BMETS-Longitudinal
Brain metastases, longitudinal pre- and post-treatment MRI with neuroradiologist-approved segmentations
Open the source page ↗ See it in the directory
At a glance
- What it is
- Brain metastases, longitudinal pre- and post-treatment MRI with neuroradiologist-approved segmentations
- Area
- Neuro
- Modality
- MR
- Format
- NIfTI
- Size
- 300 subjects; 1,066 studies; about 163 GB
- Normals included
- No
- 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
- 2026-10-21
- Last checked
- 2026-11-11
- Source
- www.cancerimagingarchive.net/collection/ucsd-bmets-longitudinal/
How to access
Open the collection page and use the download control in the Data Access table (TCIA Data Retriever or Aspera, whichever that row names). A public Creative Commons row does not need an application. Read the Citations & Data Usage Policy tab before use. The MRI and segmentation package is NIfTI, not DICOM, and that row requires the IBM Aspera Connect plugin. The clinical TSV, data dictionaries, and MRI metadata TSV are separate small downloads on the same table.
Limitations
Not DICOM. DICOM was converted with dcm2niix and skull-stripped with HD-BET, then registered to the contrast T1 from the first time point. Segmentations are NIfTI masks in that same space (BraTS compartments: enhancing tissue, non-enhancing tumor core, FLAIR hyperintensity, and resection cavity), approved by a neuroradiologist after nnU-Net pre-segmentation. Not healthy controls. Two to four time points per patient, before and after treatment; a pre-diagnosis MRI was included for some patients with fewer than four scans. Every exam has pre-contrast T1, post-contrast T1, T2, and FLAIR; other sequences are missing on some exams. The download table lists 1,066 studies and 11,405 images. The collection text says 1,065 MRI time points. Clinical data are in a TSV. Updated 2026-10-05.
Citation required?
Yes. Cite the collection DOI on the TCIA page (10.7937/5ejb-9g94) and follow the TCIA Data Usage Policy.
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.