MedAI Eval
Trustworthy AI evaluation utilities: calibration, subgroup analysis, robustness checks, reliability curves, and model-card generation.
MedAI Eval is the shared evaluation layer used across MedCoria systems. It supports classification metrics, calibration analysis, reliability curves, subgroup evaluation, robustness checks, and model-card generation.
The goal is to make biomedical AI systems more transparent, measurable, and honest about limitations.
Planned demos include a calibration viewer, subgroup performance explorer, and an interactive model-card builder.
View on GitHubClinical deployment note
These system pages describe technical capabilities and evaluation workflows. Any diagnostic, treatment, or patient-specific use would require appropriate validation, governance, and regulatory review.