Pipeline / 01 · Platforms

GEDM-3DQ Decision Engine

Gradient Equilibrium Decision Modeling in 3-Dimensional Quadrant — the AI spine of every MACRO HRD program.

GEDM-3DQ CORE ENGINE ARCHITECTURE Five disease-agnostic computational engines. GEDM-3DQ DECISION INTELLIGENCE V Vector Decision Engine 3DQ state vector computation M Monte Carlo Simulator Stochastic outcome simulation T Trajectory Optimizer Patient-state evolution modeling R Recommendation Generator Clinician-facing therapy outputs C Cost-Risk Modeler Multi-objective optimization 5 CORE ENGINES · 6-LAYER BACKPROPAGATION · DISEASE-AGNOSTIC · FDA SaMD CLASS II
The GEDM-3DQ core engine architecture — five computational modules (Vector Decision Engine, Monte Carlo Simulator, Trajectory Optimizer, Recommendation Generator, Cost-Risk Modeler) operating across a six-layer cognitive stack, evaluating patient state across three axes: viral activity, immune competence, and therapy toxicity.
Premise

The problem we are solving.

Most clinical AI today gives a binary answer: treat or don't, cancer or not, high-risk or low. But medicine rarely fits a binary. A patient's viral pressure, immune competence, and therapy toxicity all move continuously, pulling against one another, settling into equilibria that shift day to day.

GEDM-3DQ models this reality directly. Rather than producing yes/no verdicts, it computes a patient state vector across three axes — viral, immune, and toxicity — and identifies the therapeutic window where all three are favorable at once. It is the reasoning layer that makes our integrated platforms possible.

Approach

How we tackle it.

GEDM-3DQ is built on five disease-agnostic computational modules — Vector Decision Engine, Monte Carlo Simulator, Trajectory Optimizer, Recommendation Generator, and Cost-Risk Modeler — operating across a six-layer cognitive architecture: perception, risk integration, memory, equilibrium, decision, and learning.

This means the same engine that guides HIV therapy coordination can — with disease-specific training data — also guide glioblastoma protocol selection or TB diagnostic interpretation. The underlying reasoning is invariant. Only the biological knowledge graph changes.

The engine has been architected toward FDA SaMD Class II software-as-medical-device classification, with stochastic simulation outputs, explainable recommendations, and continuous learning from global trial data.

3DQ STATE SPACE · PATIENT MODELING Patient state as a vector in three dimensions. X · VIRAL Y · IMMUNE Z · TOXICITY patient state at time t Clinical Biomarkers Viral + immune + safety data Knowledge Graph Mechanistic disease reasoning GEDM-3DQ Engine 3-axis state modeling Optimized Therapy Dose · timing · selection Adaptive Control Continuous feedback loop
AI-driven disease modeling applied to HIV — integrating nanomedicine, immunotherapy, and decision intelligence into a continuous feedback loop of clinical biomarkers, mechanistic reasoning, state modeling, therapy optimization, and adaptive control.
Capabilities

What makes this real.

01
Three-axis state modeling
Real-time computation of patient state across viral pressure, immune competence, and toxicity risk — the therapeutic window emerges where all three are favorable.
02
Monte Carlo uncertainty
Stochastic simulations per decision point quantify not only the recommendation but the confidence behind it.
03
Six-layer reasoning spine
Perception → Attention → Memory → Reasoning → Judgment → Purpose. The same cognitive architecture that defines MACRO HRD itself.
04
Disease-agnostic core
One engine, many diseases. The reasoning is invariant; only the biological knowledge graph changes between programs.
Where It Connects

Part of an integrated platform.

04 · Infrastructure
GEDM-ICU
A research-grade, multimodal, trajectory-based intelligence layer — deployed first at the University of Antwerp's ferroptosis-oriented critical care program.
02 · Infectious Disease
HIV Functional Cure Program
An integrated therapeutic approach combining tri-specific antibodies, CAR-like NK effectors, AH-D peptide activation, and GEDM-3DQ-guided coordination.
03 · Neuro & Oncology
Brain Mapping Platform
AI-assisted neuro-biomarker discovery through integrated proteomics and genomics at Beijing Tiantan Hospital.
01 · Platforms
Immune Cell Banking Platform
Pre-engineered, AI-preconditioned immune cell reserves — a living biological insurance for every client.
⸻ Continue the platform

“A decision is not a single moment. It is a trajectory through possibility — we model the whole path, not just the endpoint.”

See all ventures → Engage with MACRO HRD