A Comprehensive Framework for Transparent and Explainable AI Sensors in Healthcare
Rabaï Bouderhem
Abstract
This research proposes a comprehensive framework for implementing explainable and transparent artificial intelligence (XAI) sensors in healthcare, addressing the challenges posed by AI “black boxes” while adhering to the European Union (EU) AI Act and Data Act requirements. Our approach combines interpretable machine learning (ML), human–AI interaction, and ethical guidelines to ensure AI sensor outputs are comprehensible, auditable, and aligned with clinical decision-making. The framework consists of three core components: First, interpretable AI model architecture using techniques like attention mechanisms and symbolic reasoning. Second, an interactive interface facilitating collaboration between healthcare professionals and AI systems. And third, a robust ethical and regulatory framework addressing bias, privacy, and accountability. By tackling transparency and explainability challenges, our research aims to improve patient outcomes, support informed decision-making, and increase public acceptance of AI in healthcare. The proposed framework contributes to the responsible development of AI technologies in full compliance with EU regulations, ensuring alignment with the vision for trustworthy and human-centric AI systems. This approach paves the way for the safe and ethical adoption of AI sensors in healthcare, ultimately enhancing patient care while maintaining high standards of transparency and accountability.