Developing, implementing and governing artificial intelligence in medicine: a step-by-step approach to prevent an...

Area of Science:

• Medical Informatics
• Clinical Artificial Intelligence
• Healthcare Technology Implementation

Background:

• Artificial intelligence (AI) in medicine shows great potential but faces challenges in clinical integration.
• Most AI models remain in development, limiting patient benefits and clinical application.
• A structured overview for developing and implementing clinical AI models is currently lacking.

Purpose of the Study:

• To provide a structured, step-by-step overview for the development and implementation of AI in medicine.
• To enhance clinician understanding of the AI implementation trajectory.
• To promote higher quality medical AI research and development.

Main Methods:

• Summarized key elements required for AI development and safe implementation in medicine.
• Incorporated current guidelines, regulatory documents, and best practices.
• Focused on creating an accessible overview for stakeholders without prior AI expertise.

Main Results:

• A comprehensive, step-by-step overview for clinical AI implementation is presented.
• The overview integrates essential elements and current guidelines for safe and effective deployment.
• It serves as a practical guide to transition AI from research to clinical practice.

Conclusions:

• The provided overview facilitates the practical implementation of AI in clinical settings.
• It aims to move AI from research (bytes) to patient care (bedside).
• This framework enhances accessibility for diverse stakeholders, promoting wider AI adoption in healthcare.