Institutional Layer

Data Governance and Privacy Architecture

DT4H separates source data, reference intelligence, cohort context, Twin state, runtime computation, execution feedback, and clinical interpretation boundaries.

Governance map #

SOURCESInput data

EHR · wearables · ArenaK · clinician · SETPOINT

DT4HModeling boundary

Cohorts · reference priors · Twin state · calibration

STATEKRuntime computation

State · confidence · readiness · provenance

GOVERNANCEReview boundary

Access · privacy · clinical oversight · auditability

Data boundaries #

01

Raw source data

EHR records, wearable streams, ArenaK events, clinician notes, and SETPOINT outcomes.

02

Normalized signals

Cleaned and structured evidence used for modeling and runtime interpretation.

03

Reference intelligence

Population priors and reference-human assumptions from research and pilot cohorts.

04

Twin state

Individual longitudinal state, confidence, trajectory, and calibration history.

05

Runtime outputs

StateK confidence, readiness, transition logic, and provenance surfaces.

06

Execution feedback

SETPOINT practices, adherence, check-ins, outcomes, and recalibration evidence.

Privacy controls #

MinimizeCollect less

Only collect data needed for modeling, calibration, validation, and execution context.

SeparateLayer access

Separate raw inputs, Twin state, research views, operator views, and clinical views.

TraceAudit provenance

Track evidence source, freshness, confidence, assumptions, and recalibration history.

GovernReview use

Boundary-sensitive outputs require human review and explicit governance context.

What is not inferred automatically #

No autonomous diagnosis

DT4H does not convert model state into diagnosis.

No autonomous treatment

SETPOINT execution remains bounded and does not replace clinical treatment decisions.

No hidden certainty

Confidence must remain visible and should not be presented as clinical certainty.

LayerGovernance Framework
StatusActive Draft
SystemDT4H / StateK / SETPOINT
BoundaryInfrastructure, not diagnosis
System lineageDT4HTwinStateKSETPOINTOutcomesRecalibration
Infrastructure boundaryDT4H models cohorts, Twins, calibration, and runtime state. It does not diagnose, prescribe, or replace licensed clinical judgment.
Document statusInfrastructure draft
Last updatedMay 2026
Applies toDT4H.ai / AvatarK.ai ecosystem