SLSDeep: Skin Lesion Segmentation Based on Dilated Residual and Pyramid Pooling Networks

TL;DR

This paper introduces SLSDeep, a novel encoder-decoder deep learning model with dilated residuals and pyramid pooling for accurate skin lesion segmentation, outperforming existing methods on public datasets.

Contribution

The paper proposes a new skin lesion segmentation model combining dilated residual layers and pyramid pooling, with a novel loss function for improved boundary accuracy.

Findings

Outperforms state-of-the-art segmentation methods.

Capable of segmenting over 100 images per second.

Achieves high accuracy on ISBI 2016 and 2017 datasets.

Abstract

Skin lesion segmentation (SLS) in dermoscopic images is a crucial task for automated diagnosis of melanoma. In this paper, we present a robust deep learning SLS model, so-called SLSDeep, which is represented as an encoder-decoder network. The encoder network is constructed by dilated residual layers, in turn, a pyramid pooling network followed by three convolution layers is used for the decoder. Unlike the traditional methods employing a cross-entropy loss, we investigated a loss function by combining both Negative Log Likelihood (NLL) and End Point Error (EPE) to accurately segment the melanoma regions with sharp boundaries. The robustness of the proposed model was evaluated on two public databases: ISBI 2016 and 2017 for skin lesion analysis towards melanoma detection challenge. The proposed model outperforms the state-of-the-art methods in terms of segmentation accuracy. Moreover, it is capable to segment more than $100$ images of size 384x384 per second on a recent GPU.