← Home
In the research wiki
- Paige Prostate — the project’s pointer page.
- External Validation — running a vendor model on local material makes this department the external cohort for the vendor; what it shows about the department is a different question and needs its own design.
- Model Abstention — when a commercial model declines a slide the denominator becomes a choice, so coverage has to be reported beside accuracy.
- Weakly Supervised Learning — the family of methods the product comes from.
- Diagnostic Test Accuracy — the metric set for an assistive read, and which of those metrics move with prevalence.