Exploring Explainability in Biomedical AI: A Bird’s-Eye View of Explainable AI and Extended Explainable AI

Integrating Artificial Intelligence (AI) into biomedical applications has revolutionized diagnostics, treatment planning, and drug discovery. However, the opacity of many AI models particularly deep learning architectures raises significant concerns in high-stakes biomedical contexts where interpretability and trust are paramount. This chapter provides a comprehensive overview of Explainable AI (XAI) and extended Explainable AI (xXAI) within the biomedical domain, examining current methodologies, challenges, and opportunities. A survey of foundational techniques such as saliency maps, SHAP, and LIME, alongside emerging paradigms like concept-based explanations, causal inference models, and human-in-the-loop frameworks, is carried out. Furthermore, we analyze how regulatory, ethical, and usability considerations shape the development and deployment of explainable systems in biomedicine. By offering a holistic perspective, we aim to clarify the landscape of biomedical XAI and xXAI, highlight critical gaps, and propose future directions for achieving transparent, trustworthy, and clinically actionable AI systems.