Title: Wavelet-based energy features for diagnosis of melanoma from dermoscopic images

Authors: K. Narasimhan; V. Elamaran

Addresses: School of Electrical and Electronics Engineering, SASTRA University, Thanjavur 613401, Tamil Nadu, India ' School of Electrical and Electronics Engineering, SASTRA University, Thanjavur 613401, Tamil Nadu, India

Abstract: Melanoma causes majority of deaths related to skin cancer if not detected and treated at an early stage. It is considered as one of the dangerous types of cancer, since it quickly spreads to other parts. A novel approach has been proposed in this paper to detect Melanoma from dermoscopic images. Pre-processing is done to remove hair and noise in the image. Initial segmentation is carried out with watershed transform. This is followed by Maximal Similarity Region Merging process. After pre-processing and segmentation, wavelet-based energy features are extracted using daubechies (DB3), and reverse biorthogonal (RBIO3.3, RBIO3.5, and RBIO3.7) wavelet filters. 12 features are extracted using four wavelet filters. Using the Gain Ratio feature selection method, the most discriminative six features are selected for the classification purpose. Then classification has been done by using K-nearest neighbour, support vector machine, random forest, and Naïve Bayes classifier. The highest sensitivity of 97.5% is achieved in the case of support vector machine.

Keywords: wavelet energy features; watershed transform; maximal similarity region merging; classification rate; specificity; melanoma diagnosis; dermoscopic images; medical imaging; skin cancer; image segmentation; pre-processing; wavelet filters; feature extraction; K-nearest neighbour; kNN; support vector machines; SVM; random forest; naive Bayes classifier.

DOI: 10.1504/IJBET.2016.075427

International Journal of Biomedical Engineering and Technology, 2016 Vol.20 No.3, pp.243 - 252

Received: 01 Apr 2015
Accepted: 06 Sep 2015

Published online: 22 Mar 2016 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article