Explore our latest articles and insights
Three experiments tested Jev as a review signal for RH Skills. Routing fewer cases to Luna reduced modeled cost, but Jev also falsely flagged clean cases.
At the September 2026 HL7 Connectathon, we are presenting a proposal and reference implementation for modeling patient-specific oncology medication regimens to support Da Vinci CRD, DTR, and PAS prior-authorization workflows.
At the September 2026 HL7 Connectathon, a proposed AI-assisted development track will test whether guidelines, assessments, and measures can be turned into deterministic FHIR and CQL artifacts with shared fixtures, traceable source material, and human review.
Deploy clinical terminology without building a runtime data pipeline: one immutable image for ICD-10-CM, LOINC, RxNorm, SNOMED CT, search, and FHIR ValueSet expansion.
CMS's new ACCESS Model pays for outcomes, not activities. The orgs that win will be the ones whose clinical data is structured, coded, and validatable enough to prove the outcome. That's a terminology and conformance problem, and a problem we know how to solve.
A DevDays follow-up on rh-validator, the Rust-based FHIR validator in the rh toolkit, and how its internals keep validation fast, cached, and conformance-aware.
Notes from FHIR DevDays 2026 on patient-directed exchange, SMART Health Links, TEFCA, the rh Rust FHIR toolkit, and conversational interoperability for workflows that do not fit pre-negotiated APIs.
Our open-source framework orchestrates AI reasoning over raw clinical source material and produces deterministic, computable rules — structured artifacts you can embed directly into EHRs, quality programs, and clinical decision support systems.
We just shipped three terminology skills built around a different model: every code the agent returns is looked up against a live terminology server — SNOMED CT, LOINC, ICD-10-CM, RxNorm. Not retrieved from training weights. Not guessed. Resolved.
CodeX's Prior Authorization in Medical Oncology use case is one of the most important healthcare interoperability efforts happening right now. We're excited to share that Vermonster is part of the working group — and that we're contributing to the spec and an initial reference implementation.
Anthropic has a recent post on the unreasonable effectiveness of HTML as an output format. It's also selling Claude Code — but the underlying point about richer structured content is worth thinking about. Here's the version of that argument as it lands for us, and the convention we've landed on.
A small team. 15+ PRs a day. Most of them agent-generated. Code was landing faster than people could review. Here's the experiment we built to fix it — and the open-source skill we're releasing to make it work.
AI coding assistants are great at generic software — and quietly wrong about healthcare. We just published the initial set of Reason healthcare tech skills, a domain-aware workflow stack for the people actually building this stuff. Try it and tell us what's missing.
LessPrompt is a prompt pre-processor that rewrites prompts to be clearer, smaller, more structured, and more deterministic — without changing their intent.
At CMSS 2025, Vermonster connected with leaders from medical specialty societies to explore how FHIR Clinical Reasoning can translate guidance into practice, strengthen learning health systems, and link evidence to outcomes.
Turn disconnected information into computable, standards-aligned knowledge. ReasonHub brings CRMI-native authoring, terminology services, MCP-enabled APIs, and secure syndication together in one place.
ReasonHub provides the FHIR-native infrastructure for your clients asking for next-generation use cases like knowledge-based quality measurement, prior authorization, and care management. Focus on differentiation, not plumbing.
Payers have heard about the digital transition for years, but standards, infrastructure, and content have been immature. But there's been a shift. FHIR and practical FHIR knowledge systems are here, and ReasonHub is the missing piece of the puzzle
FHIR has made clinical and policy knowledge something you can build, validate, and publish like software — versioned, testable, and ready to plug into real workflows. For healthcare consultants and clinical informaticists, that shift turns expertise into durable, interoperable assets.
Leading medical societies aren't just publishing anymore, they're building learning health systems. With FHIR and ReasonHub the continuous improvement vision of 21st-century medicine is here.
Leading provider organizations aren't just exchanging data, they're building intelligent clinical systems. With FHIR and ReasonHub, executable quality measurement, evidence based decision support, and clinically aware AI become realities, not future promises.