Toward a quantitative analysis of skin lesion images

Melanoma is the most dangerous form of skin cancer. Skin cancers, and particularly melanomas, can be easily cured if detected early. The regularity of the pigment network structure is a significant sign for early diagnosis of melanoma. We present here a quantitative analysis of the morphology of the pigment network structure. Our method is based on the extraction of the pigment network structure using digital image processing techniques. Once the network is extracted, it is possible to derive statistical indicators relative to its regularity or irregularity. We report experiments on a database of 14 images, showing the effectiveness of our method.