Deep-Learning Ensembles for Skin-Lesion Segmentation, Analysis, Classification: RECOD Titans at ISIC Challenge 2018

Abstract

This extended abstract describes the participation of RECOD Titans in parts 1 to 3 of the ISIC Challenge 2018 "Skin Lesion Analysis Towards Melanoma Detection" (MICCAI 2018). Although our team has a long experience with melanoma classification and moderate experience with lesion segmentation, the ISIC Challenge 2018 was the very first time we worked on lesion attribute detection. For each task we submitted 3 different ensemble approaches, varying combinations of models and datasets. Our best results on the official testing set, regarding the official metric of each task, were: 0.728 (segmentation), 0.344 (attribute detection) and 0.803 (classification). Those submissions reached, respectively, the 56th, 14th and 9th places.

References in corpus (8)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• TernausNet: U-Net with VGG11 Encoder Pre-Trained on ImageNet for Image Segmentation
• FusionNet: A deep fully residual convolutional neural network for image segmentation in connectomics
• Data Augmentation for Skin Lesion Analysis
• Skin Lesion Synthesis with Generative Adversarial Networks
• RECOD Titans at ISIC Challenge 2017
• Towards Automated Melanoma Screening: Proper Computer Vision & Reliable Results
• Data, Depth, and Design: Learning Reliable Models for Skin Lesion Analysis

Cited by in corpus (3)

• SLSNet: Skin lesion segmentation using a lightweight generative adversarial network
• Segmentation of Skin Lesions and their Attributes Using Multi-Scale Convolutional Neural Networks and Domain Specific Augmentations
• Handling Inter-Annotator Agreement for Automated Skin Lesion Segmentation