Automatic Skin Lesion Segmentation Using Deep Fully Convolutional Networks with Jaccard Distance

Abstract

Automatic skin lesion segmentation in dermoscopic images is a challenging task due to the low contrast between lesion and the surrounding skin, the irregular and fuzzy lesion borders, the existence of various artifacts, and various imaging acquisition conditions. In this article, we present a fully automatic method for skin lesion segmentation by leveraging a 19-layer deep convolutional neural networks (CNNs) that is trained endto-end and does not rely on prior knowledge of the data. We propose a set of strategies to ensure effective and efficient learning with limited training data. Furthermore, we design a novel loss function based on Jaccard distance to eliminate the need of sample re-weighting, a typical procedure when using cross entropy as the loss function for image segmentation due to the strong imbalance between the number of foreground and background pixels.

Existing System

Thresholding Method, Wavelet Network, Genetic Algorithm.

Proposed System

In this work, we proposed a fully automatic framework based on deep convolutional neural network for skin lesion segmentation on dermoscopic images. Several effective training strategies were implemented to tackle the challenges that training a deep network may face when only limited training data is available. We designed a novel loss function based on the Jaccard distance to further boost the segmentation performance. Compared to the conventionally used cross entropy, the Jaccard distance based loss function directly maximizes the overlap between the foreground of the ground truth and that of the predicted segmentation mask, and thus eliminates the needs of data re-balancing when the numbers of foreground and background pixels are highly unbalanced, such as binary medical image segmentation. Our contributions in this paper are three fold. Firstly, we introduce a fully automated method for skin lesion segmentation by leveraging the discriminative power of a 19-layer deep FCN. To the best of our knowledge, our work is among the first few attempts to use deep neural networks to tackle this challenging problem. This model does not rely on the prior knowledge of data and is trained in an end-to-end fashion. We investigate a set of training strategies to ensure effective and efficient learning with limited training data. Secondly, we design an appropriate loss function that naturally handles the lesion-background imbalance of pixel wise classification for medical image segmentation.

Architecture

BLOCK DIAGRAM