Methodology

Methodological Foundations

A comprehensive methodology framework for understanding, managing, and studying complex health conditions — organized as a closed-loop learning system from data to evidence and back.

Overview

Why a New Methodology

Complex chronic health conditions cannot be addressed by any single analytical method. They require a systematic approach that integrates data from multiple dimensions, structures knowledge formally, models trajectories over time, supports adaptive decision-making, and generates evidence from real-world practice.

The CIM Methodology Framework provides this systematic approach. It is not a single method, but an integrated methodological system — five functional layers supported by a computational and scientific foundation, operating as a continuous closed-loop learning cycle.

The Framework

CIM Methodology Framework

Five functional layers, each building on the outputs of the previous, with evidence feeding back continuously to refine the entire system.

Layer AProblem & Data
Layer BStructure & Knowledge
Layer CModeling & Computation
Layer DDecision & Intervention
Layer EEvidence & Learning
↻ Closed-loop Learning
Foundation — Computational & Scientific Foundations

Framework Layers

Layer-by-Layer

Layer A

Problem & Data

CIM focuses on health conditions that are complex, chronic, multi-dimensional, and embedded in real-world environments. For these conditions, the data approach must be correspondingly comprehensive.

CIM develops methods for collecting and integrating multi-dimensional data — clinical records, behavioral observations, environmental assessments, biomarkers, subjective reports — organized as longitudinal datasets that capture change over time rather than single-point snapshots.

multi-dimensional datalongitudinal dataregistryclinical record form

Layer B

Structure & Knowledge

Raw data, no matter how comprehensive, is not yet knowledge. Layer B transforms multi-dimensional longitudinal data into structured, formally defined knowledge representations that computational models can reason about.

Ontology provides the formal definitions of concepts and relationships. Knowledge graphs represent relationships between concepts. Data models define how individual health trajectories are represented computationally — as dynamic, multi-dimensional state representations that evolve over time.

ontologyknowledge graphdata modelstate definition

Layer C

Modeling & Computation

This is the broadest layer of the framework. Trajectory modeling captures how health states evolve over time. Pathway modeling identifies causal and sequential pathways. State transition analysis models how individuals move between health states. Simulation enables virtual experiments. World models provide comprehensive representations of individual health dynamics.

CIM does not prescribe a single modeling approach. The methodological contribution lies in providing a framework for selecting, combining, and integrating multiple modeling approaches for complex health problems.

trajectorypathwaystate transitionsimulationworld modelcausal inferenceagent-based modeling

Layer D

Decision & Intervention

In complex chronic conditions, decision-making is fundamentally different from acute medicine. Decisions must be made continuously, adaptively, over long time horizons. Decision support frameworks integrate trajectory predictions and state assessments to provide guidance that evolves as new data becomes available.

Intervention sequencing addresses the challenge unique to complex conditions: the order, timing, and combination of multiple interventions matter. Adaptive management treats health management as a continuous optimization process.

decision supportintervention sequencingadaptive managementexecution systems

Layer E

Evidence & Learning

The final functional layer determines whether the entire system improves over time. Real-world evidence generation extracts evidence from longitudinal data accumulated during ongoing health management — as a systematic, designed-in process.

The discovery loop connects hypothesis generation to observation to evidence to new hypotheses. The Evidence Factory concept represents the design goal: infrastructure that continuously produces evidence as a natural byproduct of its operation. This layer closes the loop — evidence feeds back to refine models, update ontologies, and redefine problems.

real-world evidenceevidence generationhypothesis generationdiscovery loopevidence factory

Foundation

Computational & Scientific Foundations

Supporting all five functional layers is a foundation of computational and scientific methods: ontology and knowledge representation, machine learning, neuro-symbolic AI, agent systems, simulation science, causal inference, biostatistics, complex systems science, systems biology, control theory, and operations research.

These are not applied as isolated techniques. CIM's methodological contribution includes guiding the selection and integration of appropriate computational and scientific methods for specific research questions within the framework.

Architecture

A Closed-Loop Learning System

The defining characteristic of the CIM methodology is that it operates as a closed loop, not a linear pipeline. Evidence generated in Layer E feeds back to every other layer — refining computational models, updating ontologies and knowledge structures, redefining problems, and optimizing decision frameworks.

This is not continuous improvement as a business concept. It is continuous learning as a scientific architecture — a system designed to generate knowledge as a structural property.

Next

See how this methodology becomes operational infrastructure

Explore AEGIS