RECOD Titans at ISIC Challenge 2017

TL;DR

The paper details RECOD Titans' participation in the ISIC 2017 Skin Lesion Analysis Challenge, highlighting their segmentation and classification models' design, training strategies, and achieved validation scores.

Contribution

It presents a comprehensive approach combining multiple deep learning models and ensemble techniques for skin lesion segmentation and classification in the ISIC 2017 challenge.

Findings

Segmentation models achieved validation scores around 0.78-0.79.

Ensemble of four models improved segmentation score to 0.793.

Meta-model with seven deep-learning models enhanced classification results.

Abstract

This extended abstract describes the participation of RECOD Titans in parts 1 and 3 of the ISIC Challenge 2017 "Skin Lesion Analysis Towards Melanoma Detection" (ISBI 2017). Although our team has a long experience with melanoma classification, the ISIC Challenge 2017 was the very first time we worked on skin-lesion segmentation. For part 1 (segmentation), our final submission used four of our models: two trained with all 2000 samples, without a validation split, for 250 and for 500 epochs respectively; and other two trained and validated with two different 1600/400 splits, for 220 epochs. Those four models, individually, achieved between 0.780 and 0.783 official validation scores. Our final submission averaged the output of those four models achieved a score of 0.793. For part 3 (classification), the submitted test run as well as our last official validation run were the result from a meta-model that assembled seven base deep-learning models: three based on Inception-V4 trained on our largest dataset; three based on Inception trained on our smallest dataset; and one based on ResNet-101 trained on our smaller dataset. The results of those component models were stacked in a meta-learning layer based on an SVM trained on the validation set of our largest dataset.