UC Berkeley and UCSF Researchers Use AI to Transform Medical Imaging

Researchers from UC Berkeley and UC San Francisco are developing a startup called Voio, which aims to make medical imaging more efficient and accurate amid a growing shortage of radiologists. With an aging population and an increasing number of patients requiring diagnostic images, the number of radiologists is not keeping pace with the rising demand. This trend leads to practitioner burnout and delays in patient results.

Voio is developing artificial intelligence models to assist radiologists in interpreting images faster and more accurately. The startup's tools are designed to generate draft reports, allowing radiologists to focus on patient care, and to predict the risk of serious conditions such as cancer and osteoporosis years in advance.

Voio's CEO Adam Yala emphasizes that their goal is to empower individual radiologists to have a greater impact even with overwhelming workloads. The startup also plans to develop similar advancements in other medical fields. Yala and his team previously created Pillar-0, an open-source AI model trained on UCSF medical images to detect current and potential health issues.

Voio is currently working on Pillar-1, a new AI model that will assess patient risks based on an even broader range of images. This model will assist in interpreting complex cases and provide radiologists with insights into disease progression that they cannot detect on their own.

Dr. Maggie Chung from UCSF expresses hope that AI will help radiologists spend more time on patient care. She believes that automating routine tasks will restore joy to the work of radiologists. Yala also hopes that AI can revolutionize clinical guidelines for radiologists to enhance the quality of healthcare.