Learning From International Comparators of National Medical Imaging Initiatives for AI Development: Multiphase...

Area of Science:

• Medical Imaging
• Artificial Intelligence (AI)
• Health Informatics

Background:

• The COVID-19 pandemic accelerated investment in AI-driven medical imaging for patient management.
• England's National COVID-19 Chest Imaging Database success spurred the creation of a national AI imaging platform.
• The national digital policy body (NHSX) aimed to generalize a national medical imaging platform for AI algorithm development, validation, and deployment.

Purpose of the Study:

• To examine international use cases of medical imaging platforms for AI algorithm development and implementation.
• To inform the strategic design and development of England's national medical imaging platform.

Main Methods:

• A multiphased approach was employed by the National Health Service (NHS) AI Lab Policy and Strategy Team.
• Methods included identifying and prioritizing national AI imaging platforms, conducting Political, Economic, Social, Technological, Legal, and Environmental (PESTLE) analyses, and performing semistructured interviews with stakeholders.
• A workshop synthesized insights and informed policy recommendations.

Main Results:

• Seven international AI imaging platforms were analyzed using PESTLE factors.
• Thirteen international stakeholders were interviewed, yielding eight key themes.
• Twelve policy recommendations were formulated to guide the development of the English national medical imaging platform.

Conclusions:

• AI algorithm development often occurs in silos, necessitating international collaboration.
• Learnings from international counterparts highlight the need for demonstrable platform benefits, robust validation, and tools for non-technical users.
• Policy makers must leverage national and international learnings for successful technological healthcare initiatives.