Automatic detection of skin cancer in dermoscopy images: A medically oriented approach

Melanoma (skin cancer) is considered one of the most concerning forms of cancer due to its great potential to metastasize. Furthermore, statistical data show that the incidence rates of melanoma have been steadily growing in the past decades, leading to an increase in the number of deaths related with it. However, if melanoma is diagnosed on an early stage, the life span of the patient can be significantly extended. Thus, it is necessary to develop reliable tools that help dermatologists make their diagnosis and the consequent follow up process. This talk is going to describe how machine learning can be used in this task, from a more common pattern recognition approach using traditional discriminant features (such as color histograms), to a more medical oriented approach where we try to incorporate the medical knowledge in our system and search for localized relevant structures.

This last approach uses the Bag of Visual Words model. Both strategies achieve promising results and show and show that machine learning is a useful tool.