Acquire

Determines which measurement to obtain next, in whom, and at what point further measurement stops earning its cost. Compares more of the same against a different measurement family against measuring only some subjects, and returns the option that buys the most for the budget.

—

Validation & data provenance

COPDGene / Bowler selective acquisition785 participants. At 30%, 40%, and 50% acquisition, the history layer recovered median 71.0%, 85.6%, and 96.9% of universal measurement value.

PRO-ACT ALS8,260 participants · 70,746 longitudinal transitions. More-same AUC gain +0.0020; laboratory measurements +0.0131 AUC and +0.0087 branching reduction.

BoundarySelective-acquisition percentages are dataset- and representation-specific; they are not a universal savings multiplier.