Smartphone apps for skin cancer diagnosis: Implications for patients and practitioners

Abstract

A research team at Stanford recently reported that their deep convolutional neural network had learned to classify skin cancer with a level of competence equivalent to that of board-certified dermatologists. It is possible that in time, and using larger datasets, such software may surpass the average doctor in diagnostic ability, and that highly accurate technology may be available to both clinicians and patients via smartphones. This technology is poised to change the landscape of skin cancer diagnosis for both physicians and patients, but whether such changes are beneficial will depend on how they are regulated and implemented.