This Deep Learning Algorithm is as Effective as a Dermatologist

 - Jan 27, 2017
References: news.stanford.edu & engadget
As Engadget reports, those worried about skin cancer may no longer have to only depend on a dermatologist to recognize the signs of it for much longer, as a group of Stanford researchers have developed a deep learning algorithm that works to much of the same effect.

The team of researchers managed this by "training an existing Google image recognition algorithm using over 130,000 photos of skin lesions representing 2,000 diseases." By using these images, they're able to detect different types of cancers and benign lesions with impressive accuracy.

Although early tests have proven the efficiency of the deep learning algorithm, it must go through more extensive tests before it can be implemented for actual medical use. Due to the uncertainty that's currently associated with it, the AI system's true potential isn't fully known, however it's possible that it could be used with a smartphone camera once fully developed.