UCSF AI News

  • As Health Care Goes Digital, Patients Are Being Left Behind

    By Victoria Colliver on

    An NIH-funded survey by UCSF researchers found that despite the fact that digital health will continue to be core in health delivery, less than half of surveyed health systems are checking for barriers to using patient portals, telemedicine, and other digital health tools.

  • AI Bots Accurately Analyze Big Data Set to Predict Preterm Birth

    By Levi Gadye on

    A team of UCSF researchers successfully tested several mainstream AI agents for the ability to analyze big data on women's reproductive health, with the goal of predicting pre-term births.

  • For Global Health Care Providers, the Stakes Have Never Been Higher

    By Talya Sanders on

    A Q&A with the executive director of the UCSF Institute for Global Health Sciences, Payam Nahid, that checks in on the state of global health in the wake of USAID and other cuts.

  • Is AI the Cure for Health Care’s Biggest Challenges?

    By Brandy Ford on

    UCSF’s Robert Wachter, MD, a leading voice on the impact of technology on medicine, explores what generative Al can — and can't yet — do for patient care.

  • Five Questions for Kole Roybal

    By Ashley Han on

    At the new Weill Cancer Hub West, Kole Roybal, PhD, is reimagining how immune cells fight solid tumors.

  • Psychiatrists Hope Chat Logs Can Reveal the Secrets of AI Psychosis

    By Laura López González on

    UCSF psychiatrists may have documented what is likely the first case of AI psychosis in a peer-reviewed journal. He tells us what to know as media reports about cases mount.

  • A New Way to Diagnose Deadly Lung Infections and Save Lives

    By Robin Marks on

    UCSF researchers have discovered a way to identify deadly lung infections with improved accuracy by pairing AI with a unique genetic signal. This could revolutionize ICU care and prevent thousands of patients from receiving unnecessary antibiotics.

  • Wearable Tech Can Help Heart Surgery Patients Stick with Rehab

    By Lorna Fernandes on

    Nurse scientist Linda Park, PhD, uses NIH funding to study cardiac rehab and digital tools—like pedometers and apps—to motivate post-surgery patients and prevent health decline.

  • Texting Helps UCSF Reach More Patients with Needed Care

    By Chad Burns on

    Using a mix of outreach methods – texts, automated calls, and live phone calls – can improve follow-up care for discharged patients, especially hard-to-reach groups. UCSF Health’s study found this approach significantly boosts engagement and reduces disparities in care, particularly among African American patients.

  • How Can AI Sentiment Analysis Apply to Complex Medical Diagnoses?

    By Chad Burns on

    UCSF researchers are studying whether large language models, a type of artificial intelligence (AI), can analyze the clinical notes of multiple physicians and other care providers to improve the diagnosis of complex liver conditions like hepatorenal syndrome (HRS).