A production-focused engineering reference for clinical laboratory data pipelines.
Architecture patterns, typed Python implementations, and compliance mappings for the teams that move instrument results from the analyzer to the patient chart — safely, deterministically, and auditably.

Read it live → www.clinicallims.org

Clinical result pipelines are where software meets patient safety. A single mis-mapped unit, a dropped HL7 segment, or an unverifiable audit trail can put a wrong value in front of a clinician. This site is a free, deeply technical handbook for the people who build those pipelines — lab directors, clinical data engineers, LIMS integrators, and Python automation builders — pairing architectural rationale with copy-pasteable code and the exact CLIA, CAP, HIPAA, and 21 CFR Part 11 clauses each design decision satisfies.

Every page is grounded in real standards (HL7 v2, ASTM E1394, FHIR R4, LOINC, SNOMED CT, UCUM), ships async-first, typed pydantic v2 Python, and carries hand-drawn architecture diagrams and structured data — no fluff, no stock screenshots.

The reference is organized into four connected areas, spanning 50+ in-depth pages:

• LIMS Architecture & Regulatory Compliance Foundations — CLIA/CAP data boundaries, HL7 v2 segment mapping, test-code taxonomy, patient identity & order matching, and security and access controls.
• Instrument Data Ingestion — HL7 / CSV Pipelines — serial/FTP polling architectures, CSV→HL7 transformation, schema validation and error handling, and async batch processing of analyzer output.
• Clinical Result Validation & Rule Engine Architecture — reference-range checks, delta validation, critical-value routing, reflex-testing logic, threshold tuning, and auto-verification rule configuration.
• Result Reporting, Release & Interoperability — 21 CFR Part 11 signed report generation, FHIR R4 result publishing, tamper-evident audit-trail export, and LOINC/SNOMED CT code-mapping references.

• Decision guides for the choices integrators actually face — HL7 v2 vs FHIR R4 and Pydantic v2 vs Cerberus — with side-by-side comparisons and runnable code for each option.
• Runnable Python throughout: pydantic v2 models, asyncio concurrency patterns, hl7apy parsing, FHIR resource projection, hash-chained audit trails, and property-based tests with hypothesis.
• Compliance mapping tables that tie each regulatory clause to a concrete architectural control.
• Original, accessible diagrams authored as inline SVG, plus structured JSON-LD on every page.

• Eleventy — static site generator
• Hand-authored CSS (no framework), inline SVG diagrams, and KaTeX for math
• Deployed on Cloudflare static assets

npm install        # install dependencies
npm run build      # build the static site into ./_site
npm run serve      # local dev server with live reload
npm run deploy     # build and deploy to Cloudflare

• Live site: https://www.clinicallims.org
• Source: https://github.com/clinicallims/clinicallims

_{© 2026 clinicallims.org — an open engineering reference for clinical laboratory data pipelines.}