Large AI Models in Health Informatics: Applications, Challenges, and the Future

Area of Science:

• Health Informatics
• Artificial Intelligence
• Biomedical Data Science

Background:

• Foundation models, or large AI models, are emerging with unprecedented scale in parameters and data.
• These models demonstrate significant capabilities in various downstream tasks, exemplified by ChatGPT.
• The expanding multi-modal data in biomedical and health domains provide fertile ground for developing large AI models.

Approach:

• This article presents a comprehensive review of large AI models in health informatics.
• It identifies seven key application sectors: bioinformatics, medical diagnosis, medical imaging, medical informatics, medical education, public health, and medical robotics.
• The review examines current challenges and discusses future directions and potential pitfalls.

Key Points:

• Large AI models offer new paradigms for health informatics methodologies.
• Applications span diverse areas from bioinformatics to medical robotics.
• Significant potential exists for transforming health-related fields.

Conclusions:

• Large AI models are poised to revolutionize health informatics.
• Addressing challenges and carefully considering future directions are crucial for realizing their full potential.
• Continued research and validation are essential for ethical and effective implementation.