Vara Wins World-First CE Mark for Autonomous AI in Breast Screening

Vara has earned CE mark certification under the EU MDR for autonomous triage in organized breast cancer triage, which the company calls ATMON (Autonomous Triage Monitoring).

In autonomous triage, AI can report examinations classified as clearly normal without radiologist review, while all other cases remain subject to at least one radiologist read. The approach is designed for population screening programs facing radiologist shortages and the demands of double-reading mammograms.

The certification builds on the Berlin, Germany-based company’s safety system, which monitors the autonomous use of its AI models.

The system continuously monitors each site’s operating point, mammography hardware, system health, and daily performance, automatically returning sites to full radiologist reading if predefined limits are exceeded. This provides the human oversight required for high-risk AI systems under the EU AI Act.

The CE mark follows Vara’s focus on real-world AI performance. Last year, the company published PRAIM in Nature Medicine, the largest prospective study of AI in breast screening to date, covering 461,818 women across 461,818 examinations nationwide. Its real-world evidence helped support Vara’s autonomous-triage certification.

Autonomous triage is now available to screening programs across Europe, subject to national guidelines and clinical readiness. Vara is also offering ATMON to existing and new customers across all deployment modes, including decision support, independent second reading, and autonomous triage.

Comments from Vara

Stefan Bunk, chief technology officer and co-founder: The certification has only been possible because Vara has developed a safety system that supervises the autonomous use of its AI models. “What made this possible is not the AI model alone. It is based on seven years of continuous monitoring of how our AI performs in the real world, on every single case. That experience is what allowed us to build a system that can be trusted to read without a second pair of eyes.”

Jonas Muff, co-founder and CEO: “Studies have shown that AI can improve cancer detection. But even the best AI model’s performance can shift once it operates outside the controlled conditions of a study. For AI to deliver its positive impact on patient outcomes, we need to notice when this happens. We believe our ATMON system can be a template for deploying AI safely across healthcare.”