fi-fhir
fi-fhir is the sensitive healthcare integration lane in the FlexInfer stack. It turns legacy clinical formats into semantic events and routable workflows, with AI-assisted development paths that can run against private FlexInfer models instead of moving healthcare data into a shared SaaS control plane.
From legacy feeds to explainable events
Healthcare integration work usually fails in the gaps between source quirks, validation, terminology, and operations. fi-fhir makes those boundaries explicit so they can be tested in a modern AI-first IDE loop.
Source Profiles
Every feed carries its own configuration, mappings, validation expectations, and routing behavior, with vendor starter templates for Epic, Cerner, Meditech, and Allscripts that fork per interface.
Format-agnostic ingestion
HL7v2, FHIR, CSV, EDI X12 with payer companion guides, and CDA/CCDA inputs all land in one semantic event layer that is easier to test, explain, and operate.
Workflow routing
YAML workflow definitions with CEL filters route events to FHIR APIs, webhooks, databases, queues, or validation-only test paths, with a GraphQL API and TypeScript SDK around the platform.
FHIR and terminology output
US Core FHIR R4 generation across 24+ resource types plus Da Vinci profiles, with terminology mapping and configurable warn-versus-error validation.
From library to production feed runtime
The engine program assembles the capability kernel into a headless feed runtime: secure ingress, profile selection, parsing, durable receipt, persistence, and workflow delivery on one execution path, with a pinned 1.0 evidence baseline instead of untested support claims.
Authenticated ingress
HTTP HL7 ingress authenticates every submission with service, source, and tenant identity, so no production record is ever anonymous.
Durable acceptance
Atomic durable submission gives every accepted message a receipt, idempotent duplicate handling, and restart-safe persistence before delivery.
Deterministic preview
Preview mode runs the same parsing and routing semantics as production without persisting or delivering, so feed tests exercise the real execution path.
Versioned, governed artifacts
Immutable profile and workflow revisions, typed secret references, and PHI-aware data classification keep integration changes auditable.
Keep assistance close to the data boundary
The AI angle is specific: help engineers explain warnings, draft workflows, and search terminology while keeping private healthcare context aligned with the runtime and governance boundary.
Warning explanations
AI-assisted explanations turn parse warnings and opaque healthcare field names into reviewable reasoning instead of tribal knowledge.
Workflow generation
Natural-language workflow drafting and explain-back run against live LLM operations in the Mapping Studio, for operators who know the integration behavior but not the DSL shape yet.
Semantic terminology search
Meaning-based terminology search helps bridge local code systems, LOINC, RxNorm, and downstream FHIR expectations.
Entity extraction and quality scoring
Clinical entity extraction and data-quality scoring run against any OpenAI-compatible endpoint, including private FlexInfer models, behind runtime capability checks.
Healthcare is the concrete private-AI proof path
FlexInfer can host the private models used for explanations or workflow drafting. Loom governs MCP, context, and agent access. MentatLab can orchestrate multi-step integration workflows as DAGs. fi-fhir supplies the domain-specific healthcare contracts that make the platform story real.