Please use this identifier to cite or link to this item: http://hdl.handle.net/1893/29692
Appears in Collections:Computing Science and Mathematics Conference Papers and Proceedings
Author(s): Ali, Abder-Rahman
Li, Jingpeng
Trappenberg, Thomas
Title: Supervised Versus Unsupervised Deep Learning Based Methods for Skin Lesion Segmentation in Dermoscopy Images
Editor(s): Meurs, M-J
Rudzicz, F
Citation: Ali A, Li J & Trappenberg T (2019) Supervised Versus Unsupervised Deep Learning Based Methods for Skin Lesion Segmentation in Dermoscopy Images. In: Meurs M & Rudzicz F (eds.) Advances in Artificial Intelligence. Lecture Notes in Computer Science, 11489. Canadian AI 2019: 32nd Canadian Conference on Artificial Intelligence, Kingston, ON, Canada, 28.05.2019-31.05.2019. Cham, Switzerland: Springer, pp. 373-379. https://doi.org/10.1007/978-3-030-18305-9_32
Issue Date: 2019
Series/Report no.: Lecture Notes in Computer Science, 11489
Conference Name: Canadian AI 2019: 32nd Canadian Conference on Artificial Intelligence
Conference Dates: 2019-05-28 - 2019-05-31
Conference Location: Kingston, ON, Canada
Abstract: Image segmentation is considered a crucial step in automatic dermoscopic image analysis as it affects the accuracy of subsequent steps. The huge progress in deep learning has recently revolutionized the image recognition and computer vision domains. In this paper, we compare a supervised deep learning based approach with an unsupervised deep learning based approach for the task of skin lesion segmentation in dermoscopy images. Results show that, by using the default parameter settings and network configurations proposed in the original approaches, although the unsupervised approach could detect fine structures of skin lesions in some occasions, the supervised approach shows much higher accuracy in terms of Dice coefficient and Jaccard index compared to the unsupervised approach, resulting in 77.7% vs. 40% and 67.2% vs. 30.4%, respectively. With a proposed modification to the unsupervised approach, the Dice and Jaccard values improved to 54.3% and 44%, respectively.
Status: AM - Accepted Manuscript
Rights: This is a post-peer-review, pre-copyedit version of a chapter published in Meurs M & Rudzicz F (eds.) Advances in Artificial Intelligence. Lecture Notes in Computer Science, 11489. Canadian AI 2019: 32nd Canadian Conference on Artificial Intelligence, Kingston, ON, Canada, 28.05.2019-31.05.2019. Cham, Switzerland: Springer, pp. 373-379. The final authenticated version is available online at: https://doi.org/10.1007/978-3-030-18305-9_32

Files in This Item:
File Description SizeFormat 
CanadianAI19_submitted.pdfFulltext - Accepted Version3.51 MBAdobe PDFView/Open



This item is protected by original copyright



Items in the Repository are protected by copyright, with all rights reserved, unless otherwise indicated.

If you believe that any material held in STORRE infringes copyright, please contact library@stir.ac.uk providing details and we will remove the Work from public display in STORRE and investigate your claim.