Automatic segmentation of skin lesions using deep learning
This work addresses the need for automated melanoma detection in dermatology, but it is incremental as it builds on existing U-net methods with added processing steps.
The paper tackled the problem of automatically segmenting skin lesions from dermoscopic images, achieving accurate lesion boundary segmentation using a U-net deep learning network with pre- and post-processing enhancements.
This paper summarizes the method used in our submission to Task 1 of the International Skin Imaging Collaboration's (ISIC) Skin Lesion Analysis Towards Melanoma Detection challenge held in 2018. We used a fully automated method to accurately segment lesion boundaries from dermoscopic images. A U-net deep learning network is trained on publicly available data from ISIC. We introduce the use of intensity, color, and texture enhancement operations as pre-processing steps and morphological operations and contour identification as post-processing steps.