Towards an automatic diagnosis system for skin lesions: Estimation of blue-whitish veil and regression structures

This paper deals with ELM image processing for automatic analysis of pigmented skin lesions which represents one of the greatest challenges of dermatologic practice today. The "ELM 7 point checklist" defines a set of seven features, based on colour and texture parameters, which describe the malignancy of a lesion. It has been revealed as faster and with the same accuracy than the traditional ABCD criteria in the diagnosis of melanoma. A preliminary approach to the automated diagnosis of melanocytic skin lesions, based on ELM 7 point checklist is proposed. In particular, the image processing algorithms and classification techniques involved in the automatic detection of the occurrence of two criteria (Blue-whitish Veil and Regression structures) are introduced and the experimental results are reported.