The Effect of Explainable AI and Uncertainty Quantification on Medical Students’ Perspectives of Decision-Making AI: A Cancer Screening Case Study

Artificial intelligence (AI) offers potential to enhance healthcare decision-making but is limited by its 'black box' nature. Explainable AI (XAI) and Uncertainty Quantification (UQ) address these challenges by improving interpretability and reliability. Despite their potential, the impact of XAI and UQ on medical students’ perception of AI in healthcare remains unclear. This study explores the impact of XAI and UQ on medical students’ perceptions of AI in healthcare. A mixed-method study with 131 medical students from Singapore and China assessed the effects of varying AI methods on trust, usability, and decision-making. Results show that XAI and UQ enhance AI usability but highlight the need for clinically relevant explanations and contextualised uncertainty reasoning to optimise AI adoption in healthcare.