Patient and dermatologists' perspectives on augmented intelligence for melanoma screening: A prospective study.

Area of Science:

• Dermatology
• Artificial Intelligence
• Medical Imaging

Background:

• Artificial intelligence (AI) offers potential for augmented intelligence (AuI) to improve decision-making.
• Critical evaluation of AI in real-world skin cancer screening is necessary.

Purpose of the Study:

• To investigate patient and dermatologist perspectives on skin cancer screening using human, artificial, and augmented intelligence.
• To compare patient safety and dermatologist trust across different screening methods.

Main Methods:

• Prospective comparative cohort study with 205 high-risk patients and 8 dermatologists.
• Inclusion of 2D/3D total-body photography (TBP) and AI-powered dermoscopic lesion analysis.
• Patient and dermatologist surveys assessing experiences and safety (Subjective Safety Score - SSS).

Main Results:

• 95.5% of patients believed AI could enhance diagnostic performance.
• 83.4% of patients preferred AuI screening over AI or dermatologist alone; AuI yielded higher SSS (9.5) vs. AI (7.7) or dermatologist (9.1).
• Dermatologists showed high trust in AI results (90.2%-96.1%), with 68.3% feeling diagnostic accuracy improved; beginners benefited more than experts.

Conclusions:

• Patients demonstrated a clear preference for AuI with 3D-TBP in melanoma detection.
• Dermatologists, despite confidence in AI, maintained reliance on their clinical judgment for final decisions.