Early diagnosis is a key component of melanoma prevention, but diagnostic accuracy varies considerably among clinicians managing patients with potentially suspicious skin lesions. Several tools have been developed to objectively enhance diagnostic assessment and accuracy. Herein, we discuss the potential diagnostic value and limitations of GoogleNet Inception, a new tool for image-based classification of skin lesions.
Refers to Esteva, A. et al. Dermatologist-level classification of skin cancer with deep neural networks. Nature542,115-118 (2017)