Toward a Conceptual Framework for Trustworthy and Explainable Artificial Intelligence in Medical Decision-making
Type: article
Abstract: Artificial intelligence (AI) is increasingly embedded in clinical decision-support workflows, from imaging interpretation to risk stratification and postoperative monitoring. Yet, many high-performing models remain difficult to interpret, which limits clinicians’ ability to justify decisions, assign responsibility, and evaluate safety under real-world variability. This article introduces a conceptual framework – TE-AIM (Trustworthy and Explainable Artificial Intelligence in Medicine) – that consolidates five complementary pillars: explainability, reliability, fairness, accountability, and human oversight. The framework is developed through an interdisciplinary synthesis spanning computer science, medical ethics, cognitive psychology, and AI governance. Within TE-AIM, trust is treated not as a single feature of an algorithm but as an emergent outcome of how these pillars interact in clinical use. The model offers a structured lens for designing and evaluating medical AI systems that are interpretable, technically robust, and aligned with ethical and professional obligations
Source: CyberLeninK (CyberLeninka)
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7165237111