HealthX-Lab/DiffusionSynCTSeg

[MICCAI2024] Official implementation of "CT-Based Brain Ventricle Segmentation via Diffusion Schrödinger Bridge without target domain ground truths"

34
/ 100
Emerging

This project offers an automated way to identify and outline brain ventricles from CT scans, which is crucial for emergency neurosurgeries. It takes unpaired CT and MRI images (where only MRI images have known ventricle outlines) as input and produces CT scans with accurate ventricle segmentations, along with a confidence measure for the result. Neurosurgeons, radiologists, and clinicians involved in brain injury or stroke management would use this tool.

No commits in the last 6 months.

Use this if you need a reliable method to quickly and accurately segment brain ventricles from CT scans, especially when annotated CT data is scarce.

Not ideal if you require segmentation of other brain structures besides ventricles or if you primarily work with MRI scans that already have detailed annotations.

neurosurgery radiology medical-imaging CT-segmentation emergency-medicine
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 5 / 25
Maturity 16 / 25
Community 13 / 25

How are scores calculated?

Stars

10

Forks

2

Language

Python

License

Apache-2.0

Last pushed

Nov 14, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/HealthX-Lab/DiffusionSynCTSeg"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.