rachitsaluja/BraTS-2023-Metrics

Official BraTS 2023 Segmentation Performance Metrics

30
/ 100
Emerging

This tool helps radiologists, neurologists, and medical researchers more accurately assess how well AI models segment brain tumors from MRI scans. It takes in predicted tumor segmentations from an AI model and compares them against known ground truth segmentations, outputting detailed, lesion-by-lesion performance scores (like Dice and Hausdorff distance) that highlight missed tumors or false positives. This provides a more clinically relevant evaluation than traditional image-level metrics, especially for multi-lesion cases.

No commits in the last 6 months.

Use this if you need to rigorously evaluate the performance of brain tumor segmentation models, focusing on individual tumor detection and accuracy, rather than just overall image-level correctness.

Not ideal if you are evaluating segmentation models for pathologies outside of brain tumors or if your primary goal is a quick, high-level assessment of image-wide segmentation accuracy.

brain-tumor-segmentation medical-imaging radiology-ai neurology-diagnostics biomedical-research
No License Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 6 / 25
Maturity 8 / 25
Community 16 / 25

How are scores calculated?

Stars

21

Forks

6

Language

Python

License

Last pushed

Jan 11, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/rachitsaluja/BraTS-2023-Metrics"

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