Active Contours Based Segmentation and Lesion Periphery Analysis for Characterization of Skin Lesions in Dermoscopy Images

• Type: Article in International Scientific Journal
• Date: 2019

Authors

Riaz, F (Author) Other; Naeem, S (Author) Other; Nawaz, R (Author) Other; Coimbra, M (Author) FCUP

Journal

Title: IEEE Journal of Biomedical and Health Informatics

Vol. 23

Pages: 489-500

ISSN: 2168-2194

Publisher: IEEE

Indexing

ISI Web of Knowledge - 33 Citations

Other information

Authenticus ID: P-00Q-9FW

DOI: 10.1109/jbhi.2018.2832455

Abstract (EN): This paper proposes a computer assisted diagnostic system for the detection of melanoma in dermoscopy images. Clinical findings have concluded that in case of melanoma, the lesion borders exhibit differential structures such as pigment networks and streaks as opposed to normal skin spots, which have smoother borders. We aim at validating these findings by performing segmentation of the skin lesions followed by an extraction of the peripheral region of the lesion that is subjected to feature extraction and classification for detecting melanoma. For segmentation, we propose a novel active contours based method that takes an initial lesion contour followed by the usage of Kullback-Leibler divergence between the lesion and skin to fit a curve to the lesion boundaries. After segmentation of the lesion, its periphery is extracted to detect melanoma using image features that are based on local binary patterns. For validation of our algorithms, we have used the publicly available PH2 and ISIC dermoscopy datasets. An extensive experimental analysis reveals two important findings: 1) the proposed segmentation method mimics the ground truth data; and 2) the most significant melanoma characteristics in the lesion actually lie on the lesion periphery.

Language: English

Type (Professor's evaluation): Scientific

No. of pages: 12