Admin · Research

Brain Tumor Segmentation

Research tool — not a regulated medical device. Their results are not a diagnosis and can be wrong. Every result must be reviewed by a qualified clinician, who remains responsible for the clinical decision.

Source: nnU-Net (BraTS)

What it does

Segments brain-tumour sub-regions (enhancing tumour, peritumoral oedema, necrotic core) from a co-registered multi-sequence MRI with an nnU-Net model and reports per-compartment volumes.

Live example A pre-computed BraTS case — labelled tumour sub-regions in an interactive 3-plane + 3D viewer.
loading…
Sign in — it’s free — to try it Open a co-registered multi-sequence brain MRI (T1, T1ce, T2, FLAIR) in the PACS worklist and choose “Brain Tumor Segmentation”.
…and the tumour volumes it measures

Per-compartment volumes from this example segmentation — NOT for clinical decisions.

Whole tumor
224.8mL
Peritumoral edema
161.5mL
Enhancing tumor
42.7mL
Necrotic core
20.6mL
Tumor core
63.3mL
Models & libraries
  • BraTS21 KAIST — tumor segmentation weights (nnU-Net) · Apache-2.0
  • HD-BET — brain extraction / skull strip · Apache-2.0
  • ANTsPy — rigid registration to T1 · Apache-2.0
  • nnU-Net — segmentation engine · Apache-2.0

Research use only. Please cite the respective papers and honour each project’s license.

This project is free. If you would like to help with hosting and development, you can donate in stablecoins — open the Donate page.

How to donate