Artificial intelligence may be just as good at spotting skin cancer as your MD, according to new research from Stanford University.
The researchers pitted their AI program against 21 trained doctors and found that the AI was as accurate at detecting skin cancer as the experts, according to the study published in the journal Nature. (The program used an existing Google algorithm that was “fed” almost 130,000 images of around 2,000 different skin diseases, and then essentially learned how to tell the difference between them.)
While the program isn’t quite ready to be rolled out to doctor’s offices nationwide, the potential impact is big: there are 5.4 million new cases of skin cancer in the U.S. each year, and melanoma — considered the most dangerous form of skin cancer — has a five-year survival rate of 97% when it’s caught early. That number drops to about 14% if it’s found in its latest stages.
Read more about the study on Stanford News.
Originally published at medium.com