Frequently Asked Questions

Neurona 28 is an AI-powered Pocket Auditor that validates Medicare Advantage diagnosis codes against documented clinical evidence under CMS-HCC V28 rules. It is built for individual risk adjustment coders, auditors, CDI specialists, and physicians.

No. Neurona 28 is a human-in-the-loop tool. The professional provides clinical judgment and documented findings. The platform provides V28 regulatory intelligence, evidence validation, and bilingual parsing. Neither functions without the other.

No. This is a deliberate design decision, not a limitation. You enter clinical findings manually in natural language. This keeps the human at the center of the workflow and avoids audit risk from automated data ingestion.

Yes. Neurona 28 is built from inception for bilingual clinical input. It processes the Spanish-English code-switching characteristic of Puerto Rico's clinical workforce without requiring translation.

We describe our posture as HIPAA-aligned security readiness, not HIPAA compliant. Full compliance certification is a later-stage investment. We deploy encryption, audit logging, PHI handling policies, and vendor DPAs within our initial development phase. Users are instructed not to enter protected health information.

We are currently in private early access and are not publishing pricing. If you are a risk adjustment professional in Puerto Rico interested in early access, join the waitlist.

The Golden Record is a proprietary ground truth engine — a curated corpus of expert-built clinical scenarios that govern the platform's AI logic. It is co-built by professionals with direct operational experience in CMS risk adjustment, RADV audit readiness, and payer compliance.

It means the system will not suggest a code without documented clinical support. If the evidence is not in your input, the code is not in the output. The system flags the gap instead of filling it.

Not yet. Puerto Rico is our initial market because every differentiator — V28 impact, bilingual workflow, MA penetration — converges here. Expansion to mainland bilingual markets is planned but contingent on proving the beachhead first.

Yes — we expect it. Early access participants are validation partners, not just users. Your feedback directly shapes the product.

No. Neurona 28 is a decision-support tool that assists coding professionals. It validates codes against clinical evidence and flags documentation gaps, but all final coding decisions are made by the human professional. The Pocket Auditor does not diagnose, treat, or make autonomous coding decisions.

Evidence-gated coding means the system will not suggest a diagnosis code unless it finds documented clinical evidence supporting that code in the medical record. If the documentation is incomplete or ambiguous, the system flags the gap and may suggest a provider query instead of guessing.

ChatGPT is a general-purpose language model trained on internet text. It can generate plausible but incorrect coding suggestions because it has no mechanism to verify clinical evidence. Neurona 28 is purpose-built for CMS-HCC coding with an evidence-gating layer called the Golden Record that prevents unsupported suggestions. ChatGPT guesses confidently. Neurona 28 refuses to guess.

CMS-HCC V28 is the updated Medicare Advantage payment model implemented by the Centers for Medicare and Medicaid Services in January 2026. It eliminated approximately 2,000 diagnosis codes from the HCC payment hierarchy and expanded federal audit capacity from approximately 40 to 2,000 auditors. The transition significantly impacts how risk adjustment coding is performed and audited.

Puerto Rico has over 90 percent Medicare Advantage penetration — one of the highest in the United States. The V28 transition hit Puerto Rico disproportionately hard with a negative 9.1 percent risk score adjustment compared to negative 3.4 percent nationally. Puerto Rico's clinical workforce operates in a bilingual Spanish-English environment that general-purpose tools do not handle well. These factors make Puerto Rico the optimal first market for an evidence-gated coding tool.