Voio, an AI-focused healthcare lab, emerged from stealth with $8.6 million in seed funding from Laude Ventures and The House Fund. The company’s primary goal is to develop a unified reading platform that assists radiologists in interpreting scans across all modalities. Voio is spun out of research labs at the University of California, Berkeley, and the University of California, San Francisco. The founding team has also introduced Pillar-0, an open-source AI model capable of analyzing medical images to detect hundreds of conditions from CT and MRI scans with exceptional accuracy. The initial version of Pillar-0 has demonstrated a 10 to 17 percent improvement in accuracy compared with leading AI models from Google, Microsoft, and Alibaba.
Health Technology Insights: Hyperfine Awarded $3.7 Million Grant to Advance Global Brain Health
Voio was founded by 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 a practicing radiologist, and Trevor Darrell, Professor of Computer Science at UC Berkeley and founder of the Berkeley AI Research Lab. Their prior AI models have been validated in more than 92 hospitals across 30 countries, and their breast cancer tool has been applied to over two million mammograms worldwide.
Radiologists play a critical role in driving better patient outcomes by providing faster, more precise diagnoses, proactively managing care based on subtle imaging indicators, and ensuring patients have access to high-quality care. Voio’s mission is to empower radiologists with advanced technology that improves care standards and addresses workforce shortages. With over 375 million CT scans performed annually, the gap in radiology capacity often leads to delays, burnout, and diagnostic errors.
Current radiology reporting requires radiologists to juggle multiple systems, including image viewers, reporting software, electronic health records, and separate AI tools. This fragmentation reduces efficiency and contributes to burnout. Voio addresses these challenges with a unified reading environment powered by vision-language AI models that process entire exams and generate high-quality draft reports, allowing radiologists to review and finalize faster while maintaining accuracy.
Health Technology Insights: Opti9 Acquires Aptible to Expand Cloud Services
Dr. Maggie Chung, Co-Founder and Medical Lead at Voio, said that radiologists should not have to compromise between speed and quality. She emphasized that the company’s goal is to make reporting seamless by connecting images, prior exams, and patient history in one intelligent platform, reducing repetitive tasks and allowing clinicians to focus on patient care.
Pillar-0 achieved an area under the curve score of 0.87 across more than 350 findings in chest, abdomen, and brain CT scans, as well as breast MRI scans. This outperforms all publicly available radiology AI models, including Google’s MedGemma, Microsoft’s MI2, and Alibaba’s Lingshu. The model can be extended to tackle complex clinical challenges, and when fine-tuned, it improved prediction of future lung cancer with Sybil-1 by 7 percent in an external study at Massachusetts General Hospital. Voio is now expanding the platform to support multi-modal workflows in radiology and enhance preventive care.
Andy Konwinski, Co-Founder of Laude Ventures, highlighted that Voio’s work represents a major shift from reactive diagnosis to predictive healthcare. He stated that the platform could transform preventive care by enabling radiology to identify health risks before symptoms appear.
Voio builds on a proven track record of translating AI research into clinical practice. Adam Yala developed Mirai, a breast cancer risk prediction model validated on two million mammograms, and Sybil, which predicts lung cancer risk from screening scans. Dr. Maggie Chung has led studies demonstrating AI’s ability to reduce diagnostic workup times for high-risk patients. Trevor Darrell founded Berkeley AI Research and led the team that created the Caffe deep learning framework, revolutionizing image processing.
Adam Yala said that Voio represents nearly a decade of work at the frontier of AI for health. He explained that the platform supports radiologists across all studies and tasks and will enable advanced AI systems to collaborate seamlessly across specialties. Jeremy Fiance, Founder of The House Fund, added that the team’s expertise in AI research and clinical impact positions Voio to define the next generation of radiology infrastructure.
Voio’s open-source approach allows academic researchers to validate and build on its models, promoting transparency in an industry where AI performance claims are often unverified. The company also plans to support the creation of independent benchmarks to establish evidence-based standards for AI in radiology.
Health Technology Insights: Fujifilm Sonosite Launches AI-Powered PIV Assist for IVs
To participate in our interviews, please write to our HealthTech Media Room at info@intentamplify.com
Source- businesswire
