Challenges in Implementing Artificial Intelligence in Breast Cancer Screening Programs: Systematic Review and...

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Public Health

Background:

• Artificial intelligence (AI) shows potential to improve accuracy and efficiency in mammographic screening.
• Clinical integration of AI faces hurdles including errors, training needs, and ethical considerations.
• A lack of specific frameworks for AI in breast cancer screening is noted.

Purpose of the Study:

• To identify challenges in implementing AI within breast screening programs.
• To utilize the Consolidated Framework for Implementation Research (CFIR) for a practical AI governance framework.

Main Methods:

• Systematic review of 3 databases (PubMed, Embase, MEDLINE) using keywords: "artificial intelligence," "regulation," "governance," "breast cancer," "screening."
• Inclusion of original studies on AI in breast cancer detection or implementation challenges.
• Narrative synthesis of findings mapped onto CFIR constructs.

Main Results:

• 20 studies included, primarily on AI-enhanced mammography (19 studies).
• Key challenges identified: reproducibility, evidentiary standards, technology, trust, ethical/legal/societal concerns, and post-adoption uncertainty.
• CFIR framework used to develop action plans for identified challenges.

Conclusions:

• AI implementation in breast cancer screening requires consistency, strong evidence, technological progress, user trust, and ethical/legal frameworks.
• Findings provide a blueprint for stakeholders to advance AI adoption in breast cancer screening.
• Addressing identified challenges is crucial for successful AI integration.