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

<b>Background</b>: Artificial intelligence (AI) developed for skin cancer recognition has been shown to have comparable or superior performance to dermatologists. However, it is uncertain if current AI models trained predominantly with lighter Fitzpatrick skin types can be effectively ad...

Full description

Saved in:
Bibliographic Details
Main Authors: Xue Ling Ang, Choon Chiat Oh
Format: Article
Language:English
Published: MDPI AG 2025-04-01
Series:Diagnostics
Subjects:
Online Access:https://www.mdpi.com/2075-4418/15/7/939
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:<b>Background</b>: Artificial intelligence (AI) developed for skin cancer recognition has been shown to have comparable or superior performance to dermatologists. However, it is uncertain if current AI models trained predominantly with lighter Fitzpatrick skin types can be effectively adapted for Asian populations. <b>Objectives</b>: A systematic review was performed to summarize the existing use of artificial intelligence for skin cancer detection in Asian populations. <b>Methods</b>: Systematic search was conducted on PubMed and EMBASE for articles published regarding the use of artificial intelligence for skin cancer detection amongst Asian populations. Information regarding study characteristics, AI model characteristics, and outcomes was collected. <b>Conclusions</b>: Current studies show optimistic results in utilizing AI for skin cancer detection in Asia. However, the comparison of image recognition abilities might not be a true representation of the diagnostic abilities of AI versus dermatologists in the real-world setting. To ensure appropriate implementation, maximize the potential of AI, and improve the transferability of AI models across various Asian genotypes and skin cancers, it is crucial to focus on prospective, real-world-based practice, as well as the expansion and diversification of existing Asian databases used for training and validation.
ISSN:2075-4418