I
Mechanistic foundation
Understand physiology
Begin with the mechanisms that support muscle function and movement, expressed through clinically meaningful educational models.
Explore PhysiologResearch program
From mechanism to evidence to inference.
Three connected layers examine how computational systems can strengthen clinical inquiry without obscuring uncertainty or replacing human judgment.
II
Population knowledge
Represent the evidence
Turn literature into curated causal knowledge while preserving provenance, uncertainty, disagreement, and scientific review.
Explore Models4PTIII
Individual reasoning
Ask what follows here
Explore causal models, Bayesian inference, and iterative belief revision for inspectable, patient-specific explanations.
Explore the Inference EngineWorking principles
Epistemic safeguards belong in the architecture.
Population knowledge is not a patient conclusion.
General evidence constrains individual reasoning; it does not determine it without patient findings, context, and uncertainty.
Association is not causation.
Evidence types, causal claims, concepts, measurements, and curation decisions remain explicit rather than being collapsed into one record.
AI output is candidate knowledge.
Extraction can assist scientific work, but ontology resolution and human curation remain necessary before a claim enters the knowledge base.
Explanations should remain inspectable.
Inference systems should expose evidence, assumptions, model versions, uncertainty, and reasoning steps to scientific and clinical review.
The boundary
One program, not one monolithic product.
The projects are conceptually and architecturally aligned, but they are not yet technically integrated.
Physiolog is a deployed educational platform. Models4PT is an early research prototype for population-level causal knowledge. The Clinical Inference Engine defines a future patient-specific inference layer. Keeping those boundaries visible is part of the scholarly argument.
See project detailsThe aim is not to automate clinical judgment. It is to make the knowledge, assumptions, uncertainty, and inferential steps supporting that judgment more explicit.