A New Computer Program Can Diagnose Skin Cancer Without a Doctor’s Help

Health News Skin Cancer
A New Computer Program Can Diagnose Skin Cancer Without a Doctor’s Help

A new artificial intelligence program is making skin cancer diagnosis a computer’s job.

Established by a team at Stanford University, the AI algorithm is powered by a database of more than 130,000 images of skin lesions that can detect skin cancer as accurately as a human dermatologist. 

Skin cancer is generally detected by appearance and then tested through a series of biopsies. This computer program will be able to accomplish that first step.

This is a huge break through in the future of science. CNN reported that this computer program brings people one step closer to having a simple cell-phone app that could give patients an initial diagnosis of skin cancer.

Besides the convenience, what are the advantages of this technology? Computer visual systems have extremely detail-oriented “eyes,” and thus are able to pick up on tiny abnormalities that may not be apparent to the human eye.

Another perk? Technology is more easily transportable than people. An app will be able to replace a dermatologist. Those who do not have easy access to a medical clinic or hospital will still be able to receive an expert opinion regarding their health.

Stanford professor of dermatology, Susan Swetter, released a statement after the study was conducted, saying that “advances in computer-aided classification of benign versus malignant skin lesions could greatly assist dermatologists in improved diagnosis for challenging lesions and provide better management options for patients.”

However, Swetter also acknowledged that validation of the algorithm is absolutely necessary before it is released to practitioners or patients. Multiple tests and screenings have yet to be conducted, but the team at Stanford is highly optimistic.

Photo: Dave Gates

Elizabeth Chambers is a health intern with Paste and a freelance writer based out of Athens, Georgia.

Inline Feedbacks
View all comments
Share Tweet Submit Pin