Dense pooling layers in fully convolutional network for skin lesion segmentation

•Producing dense feature maps and eliminating the need for a decoder phase to reconstruct missing features.•Designing a network that is as fast as FCN and outperforms state-of-the-art methods.•Segmenting skin lesions for melanoma detection accurately. One of the essential tasks in medical image anal...

Full description

Saved in:
Bibliographic Details
Published in:Computerized medical imaging and graphics 2019-12, Vol.78, p.101658-101658, Article 101658
Main Authors: Nasr-Esfahani, Ebrahim, Rafiei, Shima, Jafari, Mohammad H., Karimi, Nader, Wrobel, James S., Samavi, Shadrokh, Reza Soroushmehr, S.M.
Format: Article
Language:English
Subjects:
Algorithms
Deep neural networks
Dense pooling layer
Image analysis
Image detection
Image processing
Image segmentation
Lesions
Medical imaging
Melanoma
Skin cancer
Skin diseases
Skin lesions
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:•Producing dense feature maps and eliminating the need for a decoder phase to reconstruct missing features.•Designing a network that is as fast as FCN and outperforms state-of-the-art methods.•Segmenting skin lesions for melanoma detection accurately. One of the essential tasks in medical image analysis is segmentation and accurate detection of borders. Lesion segmentation in skin images is an essential step in the computerized detection of skin cancer. However, many of the state-of-the-art segmentation methods have deficiencies in their border detection phase. In this paper, a new class of fully convolutional network is proposed, with new dense pooling layers for segmentation of lesion regions in skin images. This network leads to highly accurate segmentation of lesions on skin lesion datasets, which outperforms state-of-the-art algorithms in the skin lesion segmentation.
ISSN:0895-6111
1879-0771
DOI:10.1016/j.compmedimag.2019.101658