Skip to content

Dataset · Forithmus

Aligned volumetric MRI scans for medical models

Forithmus/MR-RATE-coreg

10k to 100k coregistered MRI volumes brought to a single coordinate system.

What's inside

Part of the MR-RATE release, this repository contains 10k to 100k volumetric magnetic resonance imaging scans. All images within each study are spatially coregistered relative to a common T1-weighted reference grid. The dataset is designed for English multimodal tasks, including visual question answering and classification.

What you can do with it

  • Training medical Vision-Language models for MRI analysis and report generation
  • Building visual search systems and zero-shot pathology classification from tomograms
Tasks
Image → text, Text → image, Image classification, Question answering, Visual question answering, Zero shot classification
Languages
English
Size
10K–100K rows
Downloads this month
25.9K
Likes
5
Access
on request on Hugging Face

How to load it

from datasets import load_dataset

ds = load_dataset("Forithmus/MR-RATE-coreg")