Secure EHR Interoperability with Logit-Boundary Interfaces.

Alvin Spivey, Thomas Huang· August 12, 2026 View original

Key takeaways

  • Logit boundaries provide a secure interface for EHR interoperability.
  • The Geometric Belief Interface (GBI) ensures deterministic judgment of AI outputs.
  • The system defines certificate-producing checks at the model-to-system boundary.
  • Initial benchmarks show strict quarantine for non-compliant generative model outputs.

Who benefits

HealthcareBFSIGovernmentLegalCybersecurity

Summary

This paper proposes a mathematical and engineering architecture for secure Electronic Health Record (EHR) interoperability using "logit boundaries" and a Geometric Belief Interface (GBI). The system defines certificate-producing checks at the model-to-system boundary, ensuring that generative model outputs are admissible, require review, or are quarantined before any FHIR transaction.

This research introduces a novel mathematical and engineering architecture designed to enhance secure interoperability for Electronic Health Records (EHR). The core concept revolves around "logit boundaries," which act as a critical interface between diverse systems like legacy EHRs, generative models, and human reviewers. This boundary ensures that any proposed categorical decision from a discovery model, even with pre-threshold scores, is subjected to a deterministic judgment substrate. This substrate decides whether the proposal is admissible, requires human review, or must be quarantined before any Fast Healthcare Interoperability Resources (FHIR) transaction can be constructed. The resulting Geometric Belief Interface (GBI) integrates finite boundary semantics, local Dirichlet evidence, and advanced diagnostic tools, alongside a Decentralized Cryptographic Sheaf-Enclave (DCSE) protocol for fail-closed deployment. While the framework does not aim to establish clinical truth or end-to-end safety, it rigorously defines certificate-producing checks at the model-to-system boundary. An initial benchmark using a Qwen3-4B-Instruct model demonstrated that all executions were either rejected during safe parsing or schema validation, resulting in zero accepted outputs and deterministic quarantine, highlighting the strictness of the admission boundary.

Why it matters

This framework offers a robust, secure, and auditable approach to EHR interoperability, critical for safely integrating AI and diverse systems in healthcare while maintaining data integrity and patient safety.

How to implement this in your domain

  1. 1Evaluate the "logit boundary" concept for integrating generative AI outputs into sensitive enterprise systems, beyond healthcare.
  2. 2Design and implement strict admission boundaries and deterministic judgment substrates for AI-generated content in regulated environments.
  3. 3Explore the application of Geometric Belief Interfaces (GBI) for managing uncertainty and ensuring data provenance in complex data exchanges.
  4. 4Develop cryptographic sheaf-enclave protocols for fail-closed deployment in highly sensitive network interoperability scenarios.

Original post by Alvin Spivey, Thomas Huang

"arXiv:2608.10300v1 Announce Type: new Abstract: Electronic health-record interoperability is a boundary problem: legacy systems, generative models, terminology services, identity systems, and human reviewers may each expose rich internal states, while operational exchange require…"

View on X

Originally posted by Alvin Spivey, Thomas Huang on X · view source

Want to go deeper?

Turn these trends into skills with Learnijoy's hands-on AI & tech courses.

Explore courses