Separability Analysis of Color Classes on Dermoscopic Images

Abstract

Dermoscopy is a non-invasive diagnostic technique for the in vivo observation of pigmented skin lesions used in dermatology. There is currently a great interest in the prospects of automatic image analysis methods for dermoscopy, both to provide quantitative information about a lesion, which can be of relevance for the clinician, and as a stand alone early warning tool. The standard approach in automatic dermoscopic image analysis has usually three stages: (i) segmentation, (ii) feature extraction and selection, (iii) lesion classification. This paper evaluates the potential of an alternative approach based on the Menzies method - presence of 1 or more of 6 color classes, indicating that the lesion should be considered a potential melanoma. This method does not require stages (i) and (ii) - lesion segmentation and feature extraction. The Jeffries-Matusita and Transformed Divergence metrics were used to evaluate the color class separability. The preliminary results presented in this paper suggest that a system based on the Menzies method could provide valuable information for automatic dermoscopic image analysis.