Jonghwan-dev/Awesome-Segmentation-in-Medical

Welcome, Awesome-segmentation-in-medical project. This repo is robust, reproducible, and fair benchmarking framework for breast ultrasound (BUS) image segmentation. Features 10+ models (UNet, UNet++, TransUnet, SwinUnet etc.), 5+ datasets, and strict data separation to prevent test set leakage.

30
/ 100
Emerging

This project helps medical researchers and deep learning practitioners reliably compare different breast ultrasound image segmentation models. It takes raw breast ultrasound (BUS) image datasets, trains various deep learning models like U-Net or SwinUnet, and produces fair, reproducible performance metrics. Medical image analysts and research scientists who work with diagnostic imaging for breast health would use this.

Use this if you need to rigorously benchmark deep learning models for segmenting lesions in breast ultrasound images, ensuring your comparisons are unbiased and clinically relevant.

Not ideal if you are looking for a pre-trained model for immediate clinical deployment or if your focus is on image classification rather than pixel-level segmentation.

breast-ultrasound medical-imaging image-segmentation diagnostic-imaging clinical-research
No Package No Dependents
Maintenance 10 / 25
Adoption 5 / 25
Maturity 15 / 25
Community 0 / 25

How are scores calculated?

Stars

9

Forks

Language

Python

License

MIT

Last pushed

Jan 23, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/Jonghwan-dev/Awesome-Segmentation-in-Medical"

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