Model → Interpret → Act → Learn

A system that models your health and adapts in real time.

SETPOINT builds a personalized Twin, interprets your current state, and continuously adjusts what you should do next—based on your response over time.

Current State
Strained
Improving
Twin active
Today
Recovery session
8 min · reduce strain
Projected
Stabilizing
if trajectory holds
Timeline
D1
Strained
D2
Stabilizing
D3
Stable

Health systems don’t understand you over time.

Most tools track metrics, surface snapshots, or recommend static routines. They don’t model how your system changes. Without memory, interpretation, and adaptation, personalization breaks down.

Reactive

Most systems wait until symptoms become events.

Not personalized

Static recommendations ignore individual trajectory.

No longitudinal model

Data accumulates, but the system does not learn.

From input to adaptation—continuously.

SETPOINT is a closed-loop system. It builds a model, interprets state, guides action, and updates through feedback.

1
Onboarding

Signals, symptoms, goals, and context

2
Twin

A personalized model initialized from your profile

3
State

Current interpretation of where you are

4
Plan

Adaptive next actions

5
Session

Structured execution

6
Feedback

Response updates the model

7
Update

Plan recalibrates over time

Twin Layer

Your Twin is the model the system updates over time.

SETPOINT initializes a personalized Twin during onboarding, then refines it through feedback, trajectory memory, and cohort calibration.

Initialized from
Signals + context

Goals, symptoms, load, history, and current context create the starting model.

Calibrated by
DT4H

Cohorts and reference humans help establish baseline ranges and expected trajectories.

Interpreted by
StateK

The Twin feeds state interpretation, plan selection, and protocol decisions.

Updated through
Feedback

Sessions and check-ins update the model, confidence, and projected next state.

Twin lifecycle
Onboard
Initialize
Interpret
Act
Recalibrate

A structured system, not a single feature.

SETPOINT connects data, models, decisions, protocols, sessions, and feedback into one longitudinal system.

Inputs
Signals + symptoms + context + goals
DT4H
Cohorts + reference humans + calibration
Twin
Individualized model
StateK
State interpretation + decision layer
Protocols
Structured interventions
Practices
Executable sessions
Feedback Loop
Check-ins + outcomes + recalibration
Protocol Library

Structured interventions, selected by state.

SETPOINT does not serve generic content. StateK selects protocols from a structured library, then adapts sessions based on response.

Strained → Stabilizing
Recovery v1

Downregulation protocol for elevated load and reduced recovery.

Selected by StateK using state, trajectory, cohort, and feedback.
Stabilizing → Stable
Stabilization Core

Consistency protocol for maintaining progress and reducing regression.

Selected by StateK using state, trajectory, cohort, and feedback.
Plateau → Response
Adaptive Variation

Variation logic used when feedback shows no meaningful change.

Selected by StateK using state, trajectory, cohort, and feedback.

One system, multiple surfaces.

SETPOINT operates through user, clinician, and system intelligence interfaces.

/app
User System

State visibility, sessions, feedback, trajectory.

/care
Clinical Layer

Patient state, plan oversight, decision traceability.

/ops
System Intelligence

Cohorts, calibration, protocols, audit, performance.

DT4H

Cohort-aware by design.

DT4H calibrates the Twin against cohorts, reference humans, baseline patterns, and expected trajectories. Your model is not isolated—it is continuously calibrated.

SPI

A signal—not the system.

SPI is a derived longitudinal signal inside SETPOINT. It reflects state and trajectory, but it is not the product.

Live System Example

How the system adapts over time.

A simple example of how SETPOINT interprets, acts, and updates based on feedback.

D1
State: Strained
High load + low recovery
D2
Plan: Recovery
Session reduces strain
D3
State: Stabilizing
Positive response detected
Clinical boundary

Designed for rigor.

Not diagnostic

SETPOINT supports precision health workflows; it does not replace clinical judgment.

Auditable

State, plan, protocol, and updates are traceable.

Transparent

Reasoning is visible where decisions affect action.

Start your system.

See how your model evolves from input, state, action, and feedback.