A research team from CSIRO’s Australian e-Health Research Centre, The University of Queensland, and international collaborators introduce CLIX-M, a clinician-informed 14-item evaluation checklist for explainable AI in clinical decision support systems. CLIX-M spans four categories—Purpose, Clinical, Decision, and Model attributes—offering expert-derived metrics, Likert-scale assessments, and guidance on reporting development and clinical evaluation phases.

Key points

• Introduces CLIX-M, a 14-item checklist covering Purpose, Clinical, Decision, and Model attributes for XAI evaluation.
• Incorporates expert-informed metrics such as domain relevance, coherence, actionability, correctness, confidence, and consistency.
• Utilizes quantitative methods like bootstrapping confidence intervals, feature agreement analysis, and bias assessment tools.

Why it matters: Standardized XAI evaluation enhances transparency and trust, accelerating safe integration of AI-driven decision support into clinical practice.