IFAY audits healthcare ML decisions

A final-year project tackling a genuinely hard problem — ML auditability in healthcare — but at an early, unpolished stage with minimal community traction (score: 2, no comments).

• –Records decision conditions, timestamps, and model state to enable replay and traceability of classification changes
• –Focused on microscopy datasets, but the replay/audit pattern is broadly applicable to any critical ML pipeline
• –Addresses a real gap: most ML tooling optimizes for performance metrics, not decision auditability or regulatory defensibility
• –Low community signal (2 upvotes, no discussion) suggests this is early-stage personal work rather than a mature tool