Four tutorials about one fictional company
Meridian Health Analytics Ltd supplies statistical and machine-learning models to the NHS. These four tutorials use fictional products to examine bias, explainability, privacy and environmental costs, and uncertainty in clinical AI.
Each tutorial is a group activity with ten questions in three stages. Read the evidence in each stage, then answer the questions on the control-room page. Solving a question reveals a debrief and one character of the final code. Completing a stage unlocks the next one.
Work in a group and use the evidence to support each answer. Questions may require calculations or comparisons across documents. Stage evidence opens in a separate tab; your progress is saved in this browser, so you can return to the activity later.
Choose a tutorial
- 1
The Referral That Wasn't
An audit of CARDEA, a cardiovascular risk score used to triage GP referrals, and the subgroup performance gaps it produces.
CARDEA v2.4Enter → - 2
Black Box, Open Inquest
An audit of SENTINEL, a sepsis early-warning model, and what its vendor's explainability claim can and cannot support.
SENTINEL v1.7Enter → - 3
The Price of the Model
An audit of VITAL-LM, a clinical language model, covering re-identification risk, the cost of training it, and the carbon of running it.
VITAL-LMEnter → - 4
The Number That Was Sure
An audit of HORIZON v3, an oncology prognosis model that reports a point estimate with no interval, and what that hides.
HORIZON v3Enter →