Fair AI operationalization requires improved data representativeness, auditing, and governance frameworks in healthcare solutionsNew strategies help make healthcare AI fairer for all patients

This study explores the challenges of deploying fair artificial intelligence (AI) solutions for health applications, identifying bias detection and mitigation strategies, along with the policies, governance, and monitoring mechanisms necessary for operationalizing fair AI in healthcare. A scoping review of scientific and grey/regulatory literature was conducted following PRISMA-ScR guidelines. Searches were performed across academic databases and other relevant sources from January 2022 to January 2026. Eligible studies reported AI fairness challenges, bias mitigation strategies, and perspectives on AI governance, auditing and continuous monitoring. The 66 documents included address the challenges in operationalizing fairness throughout the AI lifecycle, including examples of fairness monitoring methodologies considering the regulatory landscape. Barriers and gaps were identified to explore how bias mitigation strategies could be harmonized. The discussion provides an in-depth analysis of fairness-aware measures identified in the healthcare AI literature, while also evaluating the importance of continuous monitoring, AI auditing and governance structures. This resulted in a set of recommendations towards fairness to implement throughout the AI lifecycle for healthcare solutions. This study provides a set of recommendations for enabling AI fairness in healthcare. These include improved data representativeness, fairness-aware measures, auditing and governance frameworks. Their successful implementation will require interdisciplinary collaboration among multidisciplinary stakeholders and coordination between governance bodies to address the fairness concerns that the rapid advancements of AI bring and to ensure its trustworthiness.