Google Earth AI: Improving Public Health with Predictive Modeling & Geospatial Data

The fight against global disease is entering a new era, powered by artificial intelligence and increasingly detailed planetary-scale data. Google Earth AI, leveraging models like its Planetary-scale Foundation Model (PDFM) and AlphaEarth, is emerging as a critical tool for public health officials, offering unprecedented insights into disease patterns, forecasting capabilities, and resource allocation. From predicting clinic utilization in Malawi to improving vaccination coverage estimates for measles, these technologies are demonstrating the potential to proactively address health crises and improve outcomes worldwide.

This isn’t simply about mapping disease; it’s about understanding the complex interplay of factors that influence health, including weather patterns, population density, and access to healthcare. By combining satellite imagery, weather data, and epidemiological information, researchers are building models that can anticipate outbreaks, identify vulnerable populations, and optimize the delivery of essential medical resources. The implications are particularly profound for low- and middle-income countries, where healthcare systems are often strained and data collection can be challenging.

The core of this advancement lies in Google’s Earth AI initiatives. PDFM, a foundation model for geospatial inference, allows for the creation of detailed representations of the Earth’s surface, while AlphaEarth focuses on generating high-resolution satellite embeddings. These technologies, coupled with time-series forecasting models like TimesFM, are enabling a new level of precision in public health interventions. The ability to analyze vast datasets and identify subtle patterns is proving invaluable in a world facing increasingly complex health challenges.

Predicting Healthcare Needs and Combating Measles

In Malawi, a collaborative effort led by Google.org grantee Cooper/Smith is demonstrating the practical application of Earth AI. By integrating PDFM and AlphaEarth satellite embeddings with local data, the organization is predicting health service utilization at local clinics. This predictive capability allows decision-makers to anticipate surges in demand and allocate resources more efficiently, ensuring that clinics are adequately staffed and supplied to meet the needs of the population. According to a research paper available on arXiv (https://arxiv.org/abs/2510.25954), this approach can significantly improve the responsiveness of healthcare systems to emerging health threats.

The fight against measles, a highly contagious and potentially deadly disease, is also benefiting from these advancements. Researchers at Mount Sinai and a joint effort from Boston Children’s Hospital and Harvard University are utilizing Earth AI’s PDFM to generate “superresolution” estimates of vaccination coverage. Published in Nature (https://www.nature.com/articles/s44360-025-00031-8), this technique allows for the creation of detailed maps showing vaccination rates down to the ZIP-code level, without compromising patient privacy. This granular data is crucial for identifying localized clusters of undervaccination that may be contributing to recent outbreaks, enabling targeted interventions to increase immunization rates.

Forecasting Disease Outbreaks with Enhanced Accuracy

The relationship between weather and disease is well-established, with specific weather patterns often serving as precursors to outbreaks. For example, summer rains can create breeding grounds for mosquitoes, leading to spikes in dengue fever cases, while flooding can significantly increase the risk of cholera. By combining population dynamics with predictive weather models, public health officials can improve their ability to forecast health emergencies weeks or months in advance.

A collaboration between Google and the WHO Regional Office for Africa has yielded promising results in cholera forecasting. Researchers found that integrating Google’s TimesFM time-series model with PDFM and weather data improved the accuracy of cholera case forecasts by over 35% compared to standard models. This enhanced forecasting capability allows public health officials to proactively prepare for outbreaks, ensuring that life-saving rehydration supplies are readily available in affected areas. The WHO (https://www.afro.who.int/) continues to emphasize the importance of early warning systems in mitigating the impact of infectious diseases.

Similar success has been achieved in forecasting dengue fever. Researchers at the University of Oxford have successfully used Earth AI models and datasets to improve six-month forecasts for dengue fever in Brazil. The inclusion of PDFM embeddings significantly increased the predictive accuracy of these forecasts, providing local authorities with more time to implement preventative measures, such as mosquito control programs and public awareness campaigns.

Addressing Chronic Disease Needs in Australia

Earth AI’s applications extend beyond infectious diseases, offering valuable insights into non-communicable diseases as well. In Australia, Google has partnered with the Victor Chang Cardiac Research Institute, Wesfarmers Health, and Latrobe Health Services to deploy Population Health AI (PHAI). Currently available as a proof-of-concept to select partners, PHAI leverages Earth AI’s PDFM embeddings alongside key datasets, including air quality, pollen levels, and place insights, to uncover the health needs of communities in rural Australia. The Victor Chang Cardiac Research Institute (https://www.victorchang.edu.au/?gad_source=1&gad_campaignid=1075861874&gbraid=0AAAAADCIdnWsAFpD3oekpHKSCJM4QnXTF&gclid=Cj0KCQiA6sjKBhCSARIsAJvYcpMxQebX9hT1HNAGqYffK7WrSsfkcKOfJ2Nd1sn1TmzKeq9g3RmV1ykaAjHyEALw_wcB) is a leading research institution dedicated to preventing and treating heart disease.

Wesfarmers Health (https://www.wesfarmers.com.au/our-businesses/wesfarmers-health) and Latrobe Health Services (https://www.latrobehealth.com.au/cover-selector/?state=VIC&type=single&gad_source=1&gad_campaignid=22018179055&gbraid=0AAAAADKB5fboTA4xsAGQx3dI8b8JVF9rq&gclid=Cj0KCQiA6sjKBhCSARIsAJvYcpMxsDGRI4Trf-mjUX1DBzxp9RCtMPqB9xhKPGXspUQDuOgHHLlhjOgaAt28EALw_wcB) are key partners in this initiative, providing valuable expertise, and resources. The goal is to support chronic disease prevention efforts and address the unique health needs of rural communities, where access to healthcare can be limited.

Mount Sinai International School’s Role in Lilongwe, Malawi

While not directly involved in the AI-driven health initiatives described above, Mount Sinai International School (http://www.mountsinaiinternationalschools.com/), located in Area 14, Lilongwe, Malawi, plays a vital role in educating the next generation of healthcare professionals and leaders. Established in 2004, the school provides a co-educational learning environment for both Malawian and international students, fostering intellectual, social, and moral development. The school’s commitment to education is crucial for building a skilled workforce capable of addressing the country’s health challenges. According to their Facebook page (https://www.facebook.com/MountSinaiInternationalSchoolLlw/), the school caters to students from Nursery through IGCSE Level.

A Proactive, Healthier Future

Technology’s true power lies in its ability to translate data into actionable insights. By fusing Google Earth AI’s planetary intelligence with the deep health expertise of its partners, Google is paving the way for a future where health systems worldwide possess the data-driven insights needed to protect and improve public health. This collaborative approach, combining cutting-edge technology with local knowledge and expertise, is essential for tackling the complex health challenges facing our planet.

Looking ahead, the focus will be on expanding the reach of these technologies and integrating them into existing public health infrastructure. The next steps include refining forecasting models, improving data collection methods, and ensuring equitable access to these tools for all communities. Further research and development are planned to explore the potential of Earth AI in addressing a wider range of health issues, from non-communicable diseases to mental health. Stay tuned for updates on these initiatives from Google AI and its partners in the coming months.