Essential Informatics Tools and Computing Infrastructure for Big Data to Advance Artificial Intelligence in...

Area of Science:

• Rheumatology
• Artificial Intelligence
• Health Informatics

Background:

• Rheumatic diseases are chronic and require diverse data for effective study.
• Assembling real-world evidence for treatment effectiveness and safety is challenging.

Purpose of the Study:

• To describe the necessary infrastructure for scalable and trustworthy artificial intelligence (AI) in rheumatology.
• To outline computing infrastructure considerations for AI in rheumatology.

Main Methods:

• Emphasis on data acquisition, harmonization, linkage, and privacy protection.
• Discussion of hybrid on-premises and cloud computing architectures.
• Consideration of governance models for AI development.

Main Results:

• Scalable, trustworthy AI in rheumatology necessitates integrated data and secure computational environments.
• Hybrid cloud architectures are relevant for rheumatology computing needs.
• Shared infrastructure and longitudinal data ecosystems are vital.

Conclusions:

• Sustained progress in AI for rheumatology depends on investment in shared infrastructure.
• Effective governance models are needed to balance innovation, privacy, and clinical value.
• Longitudinal data ecosystems are essential for advancing AI in rheumatic diseases.