Voio, a new frontier AI company focused on healthcare, has emerged from stealth with an $8.6 million seed round backed by Laude Ventures and The House Fund. The Berkeley- and UCSF-affiliated team is developing a unified AI-powered reading platform designed to support radiologists across all scan types and clinical tasks, aiming to improve diagnostic accuracy, reduce burnout, and address the growing capacity challenges in medical imaging.
The company introduced its first major release, Pillar-0, an open-source AI model capable of interpreting CT and MRI scans and detecting hundreds of medical conditions directly from imaging data. According to the company, Pillar-0 represents the most accurate publicly available AI model for radiology to date, achieving 10% to 17% higher accuracy than leading models from Google, Microsoft, and Alibaba.
Voio’s founding team includes Adam Yala, Assistant Professor of Computational Precision Health at UC Berkeley and UCSF; Dr. Maggie Chung, Assistant Professor in Radiology and Biomedical Imaging at UCSF and practicing radiologist; and Trevor Darrell, Professor of Computer Science at UC Berkeley and founder of Berkeley AI Research (BAIR). Collectively, their prior AI models have been deployed in more than 92 hospitals across 30 countries and have supported over 2 million breast cancer mammogram assessments globally.
The company’s mission is to address both the clinical and logistical challenges radiologists face. With approximately 375 million CT scans conducted annually, workforce shortages have led to longer turnaround times, heavier workloads, and increased diagnostic delays. Voio aims to shift the balance back toward clinical decision-making by integrating image viewing, reporting, patient history, and AI-assisted interpretation into a single environment powered by vision-language models.
The unified platform drafts complete radiology reports, interprets exam findings, and surfaces relevant patient context, allowing radiologists to finalize analyses more efficiently while maintaining accuracy. This approach is intended to reduce the time lost to switching between disparate systems and help clinicians devote more attention to patient care.
Early benchmarks for Pillar-0 demonstrate strong performance. The model achieved a .87 AUC across more than 350 findings in CT and MRI modalities, surpassing Google’s MedGemma (.76 AUC), Microsoft’s MI2 (.75 AUC), and Alibaba’s Lingshu (.70 AUC) on the same dataset. The system also showed improvements in predictive modeling; by fine-tuning Pillar-0, Voio exceeded the state of the art for future lung cancer risk prediction by 7% in an external study at Massachusetts General Hospital.
Voio plans to continue scaling its technology to support multi-modal clinical workflows and expand its work in predictive and preventative healthcare. The company is also committed to open-source development, enabling researchers to replicate, evaluate, and extend its models. It aims to support the creation of independent benchmarks that could increase transparency in radiology AI performance assessments.
The founders bring a long history of translating academic research breakthroughs into clinically validated AI systems. Previous projects include Mirai and Sybil, two prominent risk-prediction models used in cancer imaging research and care settings, as well as contributions to Caffe. This deep learning framework played a foundational role in modern computer vision.
Voio’s investors expressed confidence in the company’s potential to shape the future of medical imaging and help radiology transition from reactive diagnosis to more preventative, proactive care. With new funding and a publicly released model, the company is positioning itself as an emerging infrastructure provider for next-generation radiology workflows.
KEY QUOTES:
“Radiologists shouldn’t have to choose between speed and quality. Our goal is to make radiology reporting seamless by drafting full reports and connecting images, history, and prior exams into one intelligent platform that feels natural to use. By re-designing the reporting experience from the ground up, we can reduce unsatisfying grunt work and let radiologists focus on patient care.”
Dr. Maggie Chung, Co-Founder and Medical Lead of Voio
“The Voio team is showing that AI research can change healthcare outcomes. This work represents a fundamental shift in healthcare from reactive diagnosis to predictive medicine. The ability to identify future health risks from current imaging before symptoms appear could transform how we approach preventive care, making radiology central to proactive health management rather than just documenting what’s already wrong.”
Andy Konwinski, Co-Founder of Laude Ventures, Databricks, and Perplexity
“Voio is the culmination of nearly a decade pushing the frontier of AI for health, building leading models for cancer imaging and validating them globally. With Voio, we’re scaling our impact across radiology; we’re building a frontier AI lab and a unified reading platform that supports radiologists across every study and task. Over time, this foundation will enable richer AI systems that collaborate seamlessly across modalities and specialties.”
Adam Yala, Co-Founder and CEO of Voio
“Adam, Maggie, and Trevor represent exactly what we look for at The House Fund – domain expert Berkeley founders combining world-class AI research with proven clinical impact. Voio is building the infrastructure layer that will define the next generation of radiology, and we’re thrilled to support their mission.”
Jeremy Fiance, Founder of The House Fund
