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

Sarker, Md. Mostafa Kamal; Rashwan, Hatem A.; Akram, Farhan; Banu, Syeda Furruka; Saleh, Adel; Singh, Vivek Kumar; Chowdhury, Forhad U.H.; Abdulwahab, Saddam; Romani, Santiago; Radeva, Petia; Puig, Domenec

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 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 formulated a new loss function by combining both Negative Log Likelihood (NLL) and End Point Error (EPE) to accurately segment the boundaries of melanoma regions. 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 the segmentation accuracy. Moreover, it is capable of segmenting about 100 images of a 384×384 size per second on a recent GPU.

Citation

SARKER, M.M.K., RASHWAN, H.A., AKRAM, F., BANU, S.F., SALEH, A., SINGH, V.K., CHOWDHURY, F.U.H., ABDULWAHAB, S., ROMANI, S., RADEVA, P. and PUIG, D. 2018. SLSDeep: skin lesion segmentation based on dilated residual and pyramid pooling networks. In Frangi, A.F., Schnabel, J.A., Davatzikos, C., Alberola-López, C. and Fictinger, G. (eds.) Medical image computing and computer assisted intervention (MICCAI 2018): proceedings of 21st international conference on Medical image computing and computer assisted interventions 2018 (MICCAI 2018), 16-20 September 2018, Granada, Spain. Lecture notes in computer science, 11071. Cham: Springer [online], pages 21-29. Available from: https://doi.org/10.1007/978-3-030-00934-2_3

• Conference Name: 1st international conference on Medical image computing and computer assisted interventions 2018 (MICCAI 2018)
• Acceptance Date: Mar 2, 2018
• Online Publication Date: Sep 26, 2018
• Publication Date: Dec 31, 2018
• Deposit Date: Dec 4, 2021
• Publicly Available Date: Jan 18, 2022
• Publisher: Springer
• Pages: 21-29
• Series Title: Lecture notes in computer science (LNCS)
• Series Number: 11071
• Series ISSN: 0302-9743
• Book Title: Medical image computing and computer assisted intervention (MICCAI 2018): proceedings of 21st international conference on Medical image computing and computer assisted interventions 2018 (MICCAI 2018), 16-20 September 2018, Granada, Spain
• DOI: https://doi.org/10.1007/978-3-030-00934-2_3
• Keywords: Skin lesion segmentation melanoma; Deep learning; Dilated residual networks; Pyramid pooling
• Public URL: https://rgu-repository.worktribe.com/output/1542109