The Use of Artificial Intelligence for Skin Cancer Detection in Asia-A Systematic Review

Area of Science:

• Dermatology
• Medical Informatics
• Computer Science

Background:

• Artificial intelligence (AI) demonstrates high performance in skin cancer recognition, comparable or exceeding dermatologists.
• Current AI models are often trained on lighter Fitzpatrick skin types, raising questions about their efficacy in diverse populations.
• Adaptability of AI for skin cancer detection in Asian populations remains an area needing investigation.

Purpose of the Study:

• To systematically review the current applications of AI in detecting skin cancer within Asian populations.
• To summarize the existing research on AI for skin cancer diagnosis in Asia.

Main Methods:

• A systematic literature search was conducted using PubMed and EMBASE databases.
• Studies focusing on AI for skin cancer detection in Asian populations were included.
• Data extraction encompassed study characteristics, AI model details, and performance outcomes.

Main Results:

• Existing studies indicate positive outcomes for AI in skin cancer detection among Asian populations.
• AI's image recognition capabilities show potential, but direct comparison to dermatologists in real-world diagnostics requires further evaluation.
• Optimistic results are reported for AI utilization in Asian skin cancer detection.

Conclusions:

• Current AI models show promise for skin cancer detection in Asian populations.
• Further validation in real-world settings and diverse Asian datasets is essential for effective AI implementation.
• Expanding and diversifying Asian databases is crucial for improving AI model transferability across different genotypes and skin cancers.