Voio emerges from stealth with Pillar-0, an open source medical imaging AI.
Voio has emerged from stealth as a frontier AI lab dedicated to healthcare, backed by USD 8.6 million in seed funding from Laude Ventures and The House Fund. The company aims to build a unified reading platform that supports radiologists across every scan and modality, addressing some of the biggest bottlenecks in medical imaging workflows.
Central to this launch is Pillar-0, an open source AI model for medical imaging that directly interprets CT and MRI scans to recognise hundreds of conditions.
With demonstrated accuracy improvements of 10%–17% over leading proprietary models from Google, Microsoft, and Alibaba, Pillar-0 represents the world’s most accurate medical imaging AI model. By open-sourcing its flagship technology, Voio intends to democratise frontier vision-language healthcare AI, enabling transparency, global collaboration, and faster adoption especially in underserved healthcare systems.
The Voio team has already proven clinical success. Their earlier AI models have been validated in over 92 hospitals across 30 countries, and their breast cancer tool has supported more than 2 million mammograms worldwide.
Voio was founded by a trio of leading academics: Adam Yala, Assistant Professor of Computational Precision Health at UC Berkeley and UCSF; Dr Maggie Chung, Assistant Professor of Radiology and Biomedical Imaging at UCSF and a practising radiologist; and Trevor Darrell, Professor of Computer Science at UC Berkeley and founder of Berkeley AI Research (BAIR).
With 375 million CT scans performed annually, radiology workloads are increasing faster than workforce capacity. Current reporting demands constant system switching, fuelling burnout and risking diagnostic delays. Voio aims to restore efficiency with a unified reading environment powered by advanced AI that interprets complete exams and drafts high-quality clinical reports for rapid review and finalisation, without compromising accuracy.
