Getting access to quality healthcare often depends on where you live and whether a specialist is available nearby. In busy city hospitals and remote village clinics around the world, healthcare workers face challenges in treating a high volume of patients with limited specialized help. To help address these challenges, we developed MedGemma, a collection of open-weight AI models designed to understand medical text and images. Built on our Gemma models, MedGemma provides developers, researchers, and public health organizations with an adaptable foundation to build specialized tools to support triage, diagnostic screenings, and more. Since its release, MedGemma has been downloaded more than 10 million times worldwide.

Published in Nature Medicine: @Google researchers and collaborators share lessons from prospective studies of AMIE, highlighting why real-world clinical evidence is critical to evaluating conversational AI in health. Read more: goo.gle/4Ac1b4P

Today, we're introducing AlphaGenome Atlas, a database that predicts the effects of every possible single nucleotide variant in the human genome. Learn more ⬇️ blog.google/innovation-and-a…

Gemma, @Google’s family of open models, has surpassed 1 billion downloads. In health, developers & researchers are using Gemma to build solutions for managing health data and pioneer breakthroughs in medical research. See the impact in health & beyond ➡️ goo.gle/3SOZKZ7

Backed by rigorous clinical research, Health Guardian features track subtle physiological shifts and help you catch changes in your health and wellness that need attention. See how Health Guardian features on Google #PixelWatch 5 helps you stay informed. ⬇️

NEW: @Zocdoc is coming to @GeminiApp, bringing a first-of-a-kind agentic experience for booking healthcare appointments. The integration will make it easier to find providers, book appointments and connect to care. Learn more: goo.gle/4wV7Q0Z

New from @GoogleResearch and @GoogleDeepMind: AMIE can now conduct real-time video consultations. In a randomized study using simulated consultations, it met clinical performance benchmarks, showing AI’s potential to expand telehealth access. Learn more: goo.gle/4bHRK2v

Google researchers joined the @AuntMinnie podcast to discuss a new study in JAMA Network that demonstrates how AI and blind sweep ultrasound has the potential to expand access to gestational age assessment and improve global maternal health. Listen here: goo.gle/4q2wSsv

Access to prenatal ultrasound remains limited in many settings. A new JAMA Network Open study found an AI system interpreting blind sweep ultrasound by novice operators achieved accuracy comparable to clinical standards, showing potential to expand access: goo.gle/4fSDXao

Dr. Michael Howell, @Google's Chief Health Officer, joins the latest episode of “A New Era of American Innovation” from @WSJ to discuss how collaborative AI research, including Project AMIE, can help connect medical research with clinical care. Watch: goo.gle/3U9lP4O

Taiwan’s National Health Insurance Administration, in partnership with Google, is transforming public health by integrating their predictive AI risk model into a universal clinician system and utilizing a Gemini assistant in a population-scale health app: goo.gle/4xbs32x

New from @GoogleResearch: A study exploring how conversational AI might conduct end-to-end symptom interviews and differential diagnostic assessment for research benchmarking purposes. Learn more about this first-of-its-kind, national-scale study: goo.gle/44NY2K3

In Nature Medicine, a new perspective from researchers at @Stanford, @Google, and leading experts across major AI labs and academic institutions, explores the need for a rigorous, task-based framework to define and measure medical AI: goo.gle/3REZ4VJ

See how Seattle Children’s is using an AI Pathway Assistant built with @GoogleCloud's Gemini Enterprise to access trusted guidance in real time—helping support more consistent care across teams. Learn more: goo.gle/4pxizfh

Today, we’re announcing that @Google is transferring the Open Health Stack code and assets to @LinuxFoundation, which will launch the Open Health Stack Software Foundation as a new, community-led home for the project. Learn more: goo.gle/3RrU8mT

In as little as 45 seconds, Nexus AI can help clinicians screen chest X-rays for lung abnormalities, including findings suggestive of pulmonary TB. Learn how a team in Pretoria leveraged Google’s CXR model to help make screenings more timely & accessible. goo.gle/4f0KReN