RECOD Titans at ISIC Challenge 2017

Our team has worked on melanoma classification since early 2014 [1], and has employed deep learning with transfer learning for that task since 2015 [2]. Recently, the community has started to move from traditional techniques towards deep learning, following the general trend of computer vision [3]. Deep learning poses a challenge for medical applications, due to the need of very large training sets. Thus, transfer learning becomes crucial for success in those applications, motivating our paper for ISBI 2017 [4]. Our team participated in Parts 1 and 3 of the ISIC Challenge 2017, described below in that order. Although our team has a long experience with skin-lesion classification (Part 3), this Challenge was the very first time we worked on skin-lesion segmentation (Part 1).