Published June 24, 2026
| Version v1.0.2
Software
Open
BI-RADS Anchored Evaluation of Artificial Intelligence for Breast Cancer Screening in Digital Mammography: A Clinical Utility Framework Beyond AUC
Description
This repository provides a reproducible implementation of a BI-RADS-anchored clinical utility framework for evaluating AI models in breast cancer screening. The framework moves beyond AUC to assess discrimination, calibration, workload-anchored performance, decision curve analysis, and subgroup equity.
Notes
If you use this software in your research, please cite it using these metadata.
Files
bsalshreef/birads-ai-evaluation-v1.0.2.zip
Additional details
Related works
- Is supplement to
- Software: https://github.com/bsalshreef/birads-ai-evaluation/tree/v1.0.2 (URL)
Software
- Repository URL
- https://github.com/bsalshreef/birads-ai-evaluation