This repository curates a comprehensive list of large AI models (LAMs), also known as foundation models, specifically applied to the healthcare domain. It categorizes these models into LLMs, LVMs, audio models, and multi-modal models, covering applications from bioinformatics to medical robotics. The project serves as a valuable resource for researchers, engineers, and practitioners exploring or developing AI solutions in healthcare.

How It Works

The repository organizes a vast collection of research papers and code repositories related to healthcare foundation models. It categorizes models by modality (language, vision, audio, multi-modal) and application area within healthcare. This structured approach allows users to quickly identify relevant models and research directions, facilitating knowledge discovery and adoption.

Quick Start & Requirements

This is a curated list of research papers and models, not a runnable software package. To utilize any of the listed models, users must refer to the individual project pages linked within the repository for specific installation and usage instructions.

Highlighted Details

• Comprehensive categorization of foundation models across LLMs, LVMs, audio, and multi-modal types.
• Extensive coverage of applications within healthcare, including bioinformatics, medical diagnosis, imaging, informatics, education, public health, and robotics.
• Includes links to papers, code repositories, and HuggingFace models for many entries.
• Features a dedicated section on AI legislation and large-scale biomedical datasets.

Maintenance & Community

The repository is actively maintained and welcomes community contributions via pull requests. It highlights recent publications and news, such as a special issue on Biomedical and Health Foundation Models for IEEE J-BHI and an article on LLM agents in medicine published in Nature Machine Intelligence.

Licensing & Compatibility

The repository itself is a list of links and does not have a specific license. However, the underlying models and code linked within the repository will have their own licenses, which users must adhere to. Compatibility for commercial use or closed-source linking depends entirely on the licenses of the individual projects referenced.

Limitations & Caveats

This repository is a curated list and does not provide a unified interface or execution environment for the models. Users are responsible for navigating to individual project pages to access, install, and run the models, which may have varying dependencies, hardware requirements, and licensing terms.