Bracco and Avicenna.AI's UPTUNITY Aneurysm clears FDA 510(k) review
Bracco and Avicenna.AI announced on Nov. 13 that UPTUNITY Aneurysm, AI software that flags suspected brain aneurysms on head CT angiography, received FDA 510(k) clearance, the first product in a new interpretive-AI portfolio built from the companies' March 2026 partnership.
Bracco and Avicenna.AI announced on Nov. 13 that the FDA granted 510(k) clearance to UPTUNITY | Aneurysm, interpretive AI software that detects and assesses suspected saccular brain aneurysms larger than 3 mm on head CT angiography (CTA) images, according to the companies' press release. Imaging-informatics outlet AuntMinnie independently confirmed the clearance the same day. The software also includes a follow-up mode for known aneurysms, producing quantitative measurements and exam-to-exam comparisons to track change over time. UPTUNITY Aneurysm cleared under 510(k) numbers K262062 and K262147 and is the first commercial product to come out of the strategic partnership Bracco and Avicenna.AI announced in March 2026, with Avicenna.AI leading U.S. commercial distribution.
The clearance fits a pattern this site has tracked through most of 2026: a contrast-media and imaging incumbent licenses detection algorithms from an AI specialist and ships the pairing as a software product, rather than building a model in-house or bolting AI onto a single scanner line. That structure only works commercially if the resulting software behaves like one coherent platform across every site that licenses it, with consistent triage logic, consistent measurement output, and consistent uptime, rather than a research model wrapped around a demo. Getting there reliably is a SaaS development problem as much as a radiology one: multi-tenant deployment, versioned model updates that don't silently change prior-exam comparisons, and an integration layer PACS and RIS vendors can certify against without a custom build for every hospital network.
Why portfolio AI raises the integration bar
Avicenna.AI already ships stroke- and pulmonary-embolism-detection tools cleared under its CINA suite, and UPTUNITY Aneurysm is explicitly framed as the opening product in a broader Bracco portfolio rather than a one-off. That changes the engineering target: a radiology department evaluating the Uptunity line isn't integrating one algorithm once, it's committing to a pipeline that will keep adding cleared modules on the same DICOM worklist and the same notification path. Software built for a single product tends to hardcode assumptions, such as one result schema, one alerting threshold, one reviewer workflow, that break the moment a second or third cleared module needs to share the same queue without clinicians losing track of which finding came from which algorithm.
What this means for teams building similar imaging-AI platforms
For a device maker or health system standing up a comparable interpretive-AI offering, the FDA submission is usually the part with the clearest playbook; the part that is harder to get right twice is the platform layer underneath it: the piece that has to onboard new hospital tenants, roll out model updates without breaking historical comparisons, and stay auditable as more cleared algorithms get added to the same line. That's infrastructure work, and it rarely survives being treated as an afterthought to the regulatory submission.
Scaling a cleared imaging AI model into a multi-site platform? Talk to Epica →
More from the blog.
Building regulated clinical software?
From EHR integration to agentic workflows, we build the software layer that holds up under real hospital IT constraints.