RichardObi/frd-score
Official implementation of the Fréchet Radiomic Distance
This tool helps medical researchers and clinicians compare distributions of medical images, especially those generated by AI models. It takes two sets of 2D or 3D radiological images (like CT or MRI scans) and their optional masks, then calculates a similarity score between them. This allows users to assess how well an AI-generated image set mimics real patient data, or compare different groups of patients.
Available on PyPI.
Use this if you need to objectively quantify the similarity between two collections of medical images based on standardized radiomic features, especially when working with deep learning models for image synthesis or segmentation.
Not ideal if you are comparing general images that aren't medical or don't require the interpretability of radiomic features.
Stars
25
Forks
3
Language
Python
License
Apache-2.0
Category
Last pushed
Feb 26, 2026
Commits (30d)
0
Dependencies
5
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/computer-vision/RichardObi/frd-score"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.