A computer system has been trained to use AI to identify skin cancer lesions with the same accuracy as a dermatologist.
Researchers from Stanford University trained the computer using a deep learning algorithm, which is a type of artificial intelligence. Using a database of 130,000 images of skin disease and applying the algorithm, the computer then diagnosed skin cancer using only the images, and proved to be as accurate as 21 board-certified dermatologists, CNN reported.
According to The Verge, the AI computer identified 1 percent more malignant lesions than the human dermatologists, and 14 percent more benign moles than human dermatologists, with at least 90 percent accuracy on both kinds of identifications.
The scientists said they are not seeking to replace dermatologists with computers.
“Our objective is to bring the expertise of top-level dermatologists to places where the dermatologist is not available,” senior study author and founder of GoogleX Sebastian Thrun told CNN.
Eventually, a mobile app may be able to diagnose skin cancer, the research suggested.
Although melanomas make up only 5 percent of skin cancers, they cause almost three quarters of skin cancer deaths, CNN reported. Early detection is key, with up to a 97 percent survival rate compared to only 14 percent when detected in the latest stage.
Dermatologists identify skin cancer primarily by looking at moles and lesions, which is exactly what the computer system does.
According to The Verge, 5.4 million new cases of skin cancer are diagnosed each year in the U.S.
© 2026 Newsmax. All rights reserved.