Google has released Health AI Developer Foundations (HAI-DEF), a new open resource designed to enable developers to efficiently build and implement AI models for healthcare att every stage of development, from early research to commercial ventures, according to Google.
The initial launch of HAI-DEF is focused on imaging applications in radiology, dermatology, and pathology. In a November 25 blog, Google said HAI-DEF builds on the research-only tool released in 2023 and supplements the Open Health Stack (released in 2023) and Population Dynamics Foundation Model (launched in 2024). Over the past two years, researchers across academia, healthcare institutions, and pharmaceutical companies have been building with these models through a Google Research-hosted application programming interface (API).
HAI-DEF provides foundations for chest x-rays, pathology for digital pathology, and dermatology for skin images, according to Google.
Each of these is an embedding model specialized for a specific medical imaging modality, according to Google. The models are developed from extensive, self-supervised training on large amounts of diverse, deidentified data for their respective modalities. As a result, the embeddings the models produce provide a powerful starting point for developers to build high-performing AI models for their own use cases, with a very small amount of additional data and computing.
