Explainable Artificial Intelligence in Public Health Services: A Review of Interpretable Machine Learning, Deep Learning, and Trustworthy AI

In the field of public health, Explanatory Artificial Intelligence (XAI) is reshaping transparency, interpretability, and trust in data-driven decision-making, making a significant impact. This review investigates the trustworthy AI, explainable deep learning and interpretable machine learning solutions for disease prediction, epidemiological surveillance and health risk assessment. The use of techniques like SHAP, LIME and Grad-CAM can help shed light on model predictions and facilitate evidence-based interventions. There are, however, data quality, algorithmic bias, privacy, explanation reliability and accountability challenges still to be overcome. The responsible implementation of AI requires a balance of fairness, transparency, and human oversight. However, robust explanations, privacy preserving approaches and equitable AI systems are key priorities for future research to advance population health outcomes.