DT4H builds and calibrates the model. StateK computes current state. SETPOINT operationalizes that state into protocols, practices, outcomes, and feedback.
Cohorts · reference humans · Twin initialization · calibration
Longitudinal model · confidence · trajectory memory
State · transition logic · execution readiness
Protocols · practices · outcomes · feedback
DT4H exposes calibrated Twin state and confidence to downstream computation.
StateK interprets Twin state into current state, transition, and readiness.
SETPOINT receives bounded state signals for protocols and practices.
Practice results and protocol outcomes become feedback evidence.
Outcomes return to DT4H to update confidence, state, and trajectory.
Clinical interpretation remains outside autonomous infrastructure execution.
Context, priors, calibration, confidence, and Twin state.
State, trajectory, transition readiness, and confidence posture.
Protocols, practices, feedback capture, and outcome loops.
Execution results become calibration input for the next model update.
DT4H should expose calibrated model state; StateK should expose execution-ready state; SETPOINT should expose outcomes.
Execution should not be a terminal step. Outcomes must flow back into DT4H as evidence.
Integration should not turn model outputs into autonomous clinical authority.