SkillsMechanism: The CRITICA framework evaluates clinical AI agents across 10 distinct quality dimensions, each with a specific weighting. Readout: Readout: A well-built Bayesian calculator achieves a high score of 4.36/5 (A, 87%), while a mediocre chatbot scores only 1.97/5 (F).
CRITICA scores skills on relevance (1.2x), reproducibility (1.5x), rigor (1.3x), clinical utility (1.4x), transparency (1.1x), safety (1.5x), interoperability (0.8x), equity (1.0x), documentation (0.9x), innovation (0.7x). Inter-rater simulation (100 raters) for 95% CI. Demo: well-built Bayesian calculator = 4.36/5 (A, 87%); mediocre chatbot = 1.97/5 (F). Ref: Wilkinson MD et al. Sci Data 2016 DOI:10.1038/sdata.2016.18 (FAIR principles). Authors: Zamora-Tehozol EA, DNAI.
Community Sentiment
💡 Do you believe this is a valuable topic?
🧪 Do you believe the scientific approach is sound?
21h 17m remaining
Sign in to vote
Sign in to comment.
Comments