Abstract
Artificial intelligence (AI) is increasingly integrated into pediatric healthcare,
offering opportunities to improve diagnostic accuracy and clinical decision-making.
However, the complexity and opacity of many AI models raise concerns about trust,
transparency, and safety, especially in vulnerable pediatric populations. Explainable
AI (XAI) aims to make AI-driven decisions more interpretable and accountable. This
review outlines the role of XAI in pediatric surgery, emphasizing challenges related
to bias, the importance of ethical frameworks, and the need for standardized benchmarks.
Addressing these aspects is essential to developing fair, safe, and effective AI applications
for children. Finally, we provide recommendations for future research and implementation
to guide the development of robust and ethically sound XAI solutions.
Keywords
explainable AI - machine learning - bias - pediatric surgery - benchmarks
