Published, and open to scrutiny
Every framework and finding behind oDIX8 is in the public record. These are the core works; the full list is on ORCID 0000-0002-6273-8014.
Ontology-Aware Design Patterns for Clinical AI Systems
Seven Gang-of-Four-style patterns for building clinical-AI pipelines resilient to ontological distortion: Ontological Checkpoint, Dormancy-Aware Pipeline, Drift Sentinel, Dual-Ontology Layer, Reification Circuit Breaker, Terminology Version Gate, Regulatory Compliance Adapter.
Read on arXiv →Artificial Epidemiology
How self-evolving clinical AI can manufacture disease prevalence from administrative coding artifacts, and a formal measure of the divergence between administrative and clinical prevalence.
Read on Cambridge Open Engage →Cascade Amplification in Administrative AI Pipelines
When several AI tools share one workflow, feedback loops turn small additive errors into system-level failures, the chain-level risk most single-tool assessments never test.
Read on Cambridge Open Engage →The Pipeline Gap
Why the EU AI Act, the Product Liability Directive and the Medical Device Regulation do not reach multi-tool clinical-AI pipelines. Posted fourteen days before Healthwatch England reported the same failure mode in NHS AI scribes.
Read on Cambridge Open Engage →The Cost of Ontological Drift
A health-economic analysis of how coding distortion propagates through clinical-AI systems and where it becomes expensive.
Read on Cambridge Open Engage →The Ontological Attack Surface
Measured distortion channels reframed as adversarial primitives, the security view of clinical-AI data integrity.
Read on Cambridge Open Engage →The Symptom and the Billing Code
How minimal medical data sets are constructed through negotiation among clinical, billing and documentation pressures. Introduces the reification feedback loop, demonstrated in the German FDZ Gesundheit.
Read on SSRN →Feedback-Loop Amplified Operational Risk in Clinical AI Pipelines
Correlated failure risk across clinical-AI pipelines, and why standard operational-risk models under Basel II/III and Solvency II understate it.
Read on SSRN →Preserve, Don’t Prune
Why strategic dormancy is an architectural necessity for clinical-AI governance: pruning rare vocabulary severs the diagnostic pathways that rare conditions depend on.
Read on Research Square →Enhanced Mycelium of Thought (EMoT)
A bio-inspired hierarchical reasoning architecture combining strategic thought dormancy with reactivation and mnemonic memory encoding. An experimental research prototype, strongest on cross-domain synthesis.
Read on arXiv →These are the core works. The complete record, including peer-reviewed journal articles, a monograph, a book chapter and further preprints, is on ORCID 0000-0002-6273-8014.