neel-dey/anatomix

[ICLR 2025] Learning General-purpose Biomedical Volume Representations using Randomized Synthesis

47
/ 100
Emerging

This tool helps researchers and clinicians analyze 3D biomedical images more effectively. It takes raw 3D medical scans, like MRI or CT, and extracts detailed features that are robust to variations in imaging equipment or patient positioning. The output can then be used to precisely align different types of scans or to quickly identify specific anatomical structures, even with limited labeled data. It's designed for biomedical researchers, radiologists, and anyone working with 3D medical volume data.

Use this if you need to accurately align different 3D medical scans or perform segmentation on new biomedical datasets with minimal annotations, without extensive pretraining.

Not ideal if your primary task involves 2D images or if you require fine-grained, pixel-level control for very specific, niche image processing tasks that aren't focused on general anatomical feature extraction.

biomedical-imaging medical-image-analysis 3D-registration medical-segmentation radiology-research
No Package No Dependents
Maintenance 13 / 25
Adoption 9 / 25
Maturity 16 / 25
Community 9 / 25

How are scores calculated?

Stars

93

Forks

6

Language

Python

License

MIT

Last pushed

Mar 22, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/computer-vision/neel-dey/anatomix"

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