
Published in Nature Medicine: peer-reviewed research behind MedGemma, our collection of open medically-tuned foundation AI models for developers. ➡️ goo.gle/3WSHa3Y

Official Google for Health news and updates on how we’re progressing our mission to help billions of people be healthier.

Published in Nature Medicine: peer-reviewed research behind MedGemma, our collection of open medically-tuned foundation AI models for developers. ➡️ goo.gle/3WSHa3Y

We're heading to #HLTHUSA! Join us in Las Vegas for conversations on scaling new technology, turning health data into clinical insights, and building sustainable paths to growth in health tech. Find us at Booth 2132, Nov. 15–18.

Today, we’re announcing Gemini for healthcare. This newest addition to our suite of purpose-built industry solutions extends the Gemini agent to health systems and health plans. Learn more → goo.gle/4zf0s11

Most of your health happens outside the clinic. In Episode 3 of The @GoogleResearch Podcast, @sineadbovell joins Katherine Chou & Dale Webster to explore MedGemma, multimodal diagnostic research (AMIE), & the potential for predictive healthcare. Watch: goo.gle/4z42YXG

More than half of adults with hypertension don’t know they have it. New Google research introduces an AI system that uses smartwatch sensors to passively detect signs of hypertension, without cuff calibration. Learn more: goo.gle/4BG1up7

Evaluating generative technology for health functions takes more than multiple-choice tests. Our new whitepaper details our framework combining benchmarks, automated assessments, and expert human review to assess accuracy, safety, and usefulness. goo.gle/3UxGBfg

Read the full paper here: arxiv.org/abs/2610.40476

In a new research paper, we show how ResidencyRL in simulation studies helps improve a model's ability to assist with routine psychiatric intake – building skills like trust-building and structured inquiry to augment and assist clinicians, rather than replace human care.

In psychiatry, conversation is central to diagnosis. Yet with mental health care constrained worldwide, tools that support front-line clinicians are urgently needed.

The highlights: 🩺 Taught by our top medical experts 🗓️ Self-paced & 100% online ✅ Approved for 5 CME credits 🏥 Keeps patient privacy and clinical judgement front and center Get started: grow.google/healthcare-profe…

Introducing the Google AI for Healthcare Professionals Certificate, practical training developed by clinicians at Google to help healthcare professionals use AI responsibly in their daily work.

From early research in 2016 to more than 1M eyes screened in real-world clinical practice. Learn how key partnerships helped scale ARDA, a diabetic retinopathy screening tool, across health systems in India, Thailand & Australia. New in @NatureMedicine: goo.gle/4iloKQv

Tuberculosis (TB) remains one of the world's deadliest infectious diseases, claiming an estimated 1.23 million lives each year. This week at IDWeek 2026, we’re highlighting how open foundation models like MedGemma are bringing AI to the front lines of TB screening to broaden early detection. Learn more: goo.gle/4i7HS4o

Our work presents the most comprehensive assessment of a personal health agent to date, and we're open-sourcing the framework and benchmarks to help the research community build the future of personalized health AI. Read our paper to learn more: goo.gle/4iCnD04

Monolithic LLMs have struggled to crunch wearable data, ground insights in clinical evidence, and guide behavior change. PHA solves this by orchestrating three specialized sub-agents, a Data Scientist, a Domain Expert, and a Health Coach, to reason over multi-modal health data.

Today in @Nature, we’re sharing the breakthrough research behind the Personal Health Agent (PHA)—the foundational science that was fundamental to how we built the Google Health Coach in the Google Health App.