An intelligent decision support system for skin cancer detection from dermoscopic images

Tan, Teck Yan, Zhang, Li and Jiang, Ming (2016) An intelligent decision support system for skin cancer detection from dermoscopic images. In: 2016 12th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD). IEEE, Piscataway, pp. 2194-2199. ISBN 978-1-5090-4094-0

Full text not available from this repository.

Abstract

It is challenging to develop an intelligent agent-based or robotic system to conduct long-term automatic health monitoring and robust efficient disease diagnosis as autonomous e-Carers in real-world applications. In this research, we aim to deal with such challenges by presenting an intelligent decision support system for skin lesion recognition as the initial step, which could be embedded into an intelligent service robot for health monitoring in home environments to promote early diagnosis. The system is developed to identify benign and malignant skin lesions using multiple steps, including pre-processing such as noise removal, segmentation, feature extraction from lesion regions, feature selection and classification. After extracting thousands of raw shape, colour and texture features from the lesion areas, a Genetic Algorithm (GA) is used to identify the most discriminating significant feature subsets for healthy and cancerous cases. A Support Vector Machine classifier has been employed to perform benign and malignant lesion recognition. Evaluated with 1300 images from the Dermofit dermoscopy image database, the empirical results indicate that our approach achieves superior performance in comparison to other related research reported in the literature.

Item Type:	Book Section
Uncontrolled Keywords:	dermoscopy, Image processing, classification, feature selection, support vector machine
Subjects:	G400 Computer Science
Department:	Faculties > Engineering and Environment > Computer and Information Sciences
Depositing User:	Becky Skoyles
Date Deposited:	09 Dec 2016 14:38
Last Modified:	12 Oct 2019 12:16
URI:	http://nrl.northumbria.ac.uk/id/eprint/28849

Downloads