Burna AI

The questions buyers actually ask, answered.

The cross-cutting questions buyers, partners, regulators, and families ask first. Architecture, pilot, pricing, validation, security, patient, investor, and how Burna compares to the systems already in your stack. Audience-specific FAQs live on the audience pages.

What Burna does.

Burna is the safety and data quality platform for clinical research, from early phase trials through postmarket surveillance. It produces CTCAE coding, including TERM and CATEGORY, coupled with intelligence to support determination of severity GRADE, with mandatory citation to source text and multi-drug attribution using WHO-UMC and Kramer algorithms across 42 oncology regimen profiles. A clinician reviews and signs every record.

A scribe writes the note. Burna reads it. There is an ambient recorder built in because it is convenient, and you can sync notes from your EHR instead. The product is the grading and the attribution.

Four buyers told us that, and all four asked for a pilot by the end of the demo. Most often it means there is a place in the EHR to enter a grade, or a rule that fires when a lab value crosses a threshold. Both are real. Burna adds the part that reads the prose, retrieves the criterion, and scores attribution per drug.

The cascading constraint pipeline.

Burna retrieves the governing CTCAE criterion, binds the grade to it, scores attribution with WHO-UMC and Kramer, and records who signed. So you get the same grade twice, you can see which criterion produced it, and you can show a monitor where it came from. That is the difference between an answer and evidence.

Every grade Burna returns is bound to the CTCAE criterion that produced it and to the span of text in the record that supports it. Your clinician can independently review that basis, which is what the term means and why it matters to a regulator.

It surfaces in your dashboard for your team to work. Real-time clinical alerting sits in a different regulatory category, and we have kept a deliberate line there.

The 90-day design partner pilot.

Epic and Oracle Health through SMART on FHIR, iKnowMed, and others on request. A new integration is usually about two weeks of our work. The longer part is your paperwork, and we say that up front.

We are running a concordance study with Mayo Clinic Platform_Accelerate and a pilot with a physician-owned community practice running fifty trials. Both, deliberately. What works at one and what changes at the other is the thing we most want to find out.

How Burna proves it.

December 2026, from the Mayo concordance study: 1,600 cases across three campuses, against an adjudicated two-clinician reference panel, with Andrea Pirzkall as co-principal investigator. We publish a number then, including in a sales meeting.

No. Every grade is a draft until your clinician signs it, and the criterion and the source line sit on screen so they can check the basis themselves. That independent review is what places Burna outside the device definition, as non-device clinical decision support. Our Q-submission with the FDA is in for the written confirmation.

HIPAA, 21 CFR Part 11, GDPR aligned.

HIPAA-aligned architecture, 21 CFR Part 11 signatures, GDPR alignment, and BAAs signed. PHI lives in an isolated plane, and every access to it is audited.

It stays yours. Your credentials stay in your systems, PHI lives in an isolated store, and Protocol Safe gives you an isolated model instance where your protocol needs to stay on infrastructure you hold.

Every question, answered.

The full library, grouped the way buyers ask. Each answer has its own page.

Question not answered? Book a 15-minute call.