Health and Biomedicine
Developing research and tools at the intersection of machine learning, AI, biomedicine and health, to deliver lasting impact for patients and communities
New computational biomedicine initiative
UC Berkeley and UCSF launched the Bakar Computational Biomedicine Initiative to develop the frontier of AI and biomedicine and accelerate advances in clinical care. The initiative will create new faculty positions, support postdoctoral researchers and provide funding to build a new open-source computational platform.
"Our goal is to establish a world-leading hub in computational biomedicine by tightly connecting UC Berkeley and UCSF, and to create new opportunities for the best minds to collaborate," said Yun Song, professor and director of the Center for Computational Biology. "By combining advances in AI with breakthroughs in biology and medicine, we hope to accelerate discovery and transform how we understand, predict and treat disease."
Edge Medicine for patients
Edge Medicine, an initiative to bring intelligent, always-on AI tools to frontline healthcare in order to improve patients’ everyday lives, involves researchers from UCSF & UC Berkeley Computational Precision Health and the Berkeley Institute for Data Science. The initiative aims to take health AI beyond traditional healthcare systems directly to patients, delivering intelligent support continuously and connecting with healthcare systems seamlessly. "By definition, chronic disease is 24/7 – yet today’s health AI is not," said Dr. Ida Sim, professor of medicine at UCSF and co-director of Computational Precision Health. "Edge Medicine aims to change that by shifting care upstream to where health outcomes are actually determined: in patients’ daily lives."
Detecting cardiac risk
Ziad Obermeyer and colleagues trained an AI system on hundreds of thousands of EKGs. The resulting cardiac risk predictions are much better than existing methods, paving the way to save lives at scale. “Medical decisions are really hard, and I think that’s why AI is so exciting for me,” said Obermeyer, associate professor at Berkeley’s School of Public Health and UCSF & UC Berkeley Computational Precision Health. “We can not only make better decisions, but also start to understand what’s actually going on with these patients before their heart stops.”
AI for clinical decisions
The large clinical behavioral model developed by Knit Health was trained on data from 130 million patients and 30 U.S. healthcare systems, learning from how clinicians make decisions in the real world to guide patient care.
“If a clinician or provider team knew everything about this patient up to this point, what are the things they might do next, and how will that unfold over time in the patient’s care journey?” asks Kolstad, CEO of Knit Health, economics professor at UC Berkeley and faculty director at the Center for Healthcare Marketplace Innovation.
The company is working to further develop and apply its clinical behavioral model with more healthcare systems.
AI for medical imaging
Voio, a startup founded by UC Berkeley and UCSF researchers, aims to build AI models that help radiologists interpret images faster and more accurately.
“We are empowering individual radiologists to have more impact even with overwhelming workloads – and ultimately, to save more patients’ lives,” said Adam Yala, CEO at Voio and assistant professor of Computational Precision Health at Berkeley and UCSF. The researchers plan to develop advancements through AI across other medical fields as well.
Frontiers in Computational Health
As AI transitions from research to the frontlines of patient care, healthcare faces a critical inflection point where the benefits of AI may be increasingly uneven.
Using the computational power of massive-scale models with clinical ethics and responsibility in mind, the Frontiers in Computational Health conference brought together pioneers, policymakers and clinicians exploring the future of health.
Photo credits — 1: Peakstock via Adobe Stock; 2: Yun Song; 3, 5, 7: Brittany Hosea-Small; 4: Courtesy of Ziad Obermeyer; Joshua Chehov via Unsplash; 6: Courtesy of Bryan Walker Ting/Voio; video: UCSF & UC Berkeley Computational Precision Health