Physicians value AI explanations in medical imaging, but trust false diagnoses over XAI

A new study analyzing explainability in AI-based medical image diagnosis surveyed 33 physicians, finding that 88% believe AI explanations are important for diagnoses. Participants rated a combination of bounding boxes and textual reports as the most effective explainability method. Notably, 50% of physicians trusted false AI diagnoses over all tested explainability techniques, highlighting a significant gap in trust and understanding. AI

Summary written by gemini-2.5-flash-lite from 1 source. How we write summaries →

IMPACT Highlights the critical need for effective explainability in medical AI to ensure trust and proper adoption by physicians.

RANK_REASON This is a research paper published on arXiv detailing a user-centric analysis of explainability methods in AI for medical image diagnosis. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CV →

• arXiv
• XAI

• paper
• safety
• other