Melanoma Detection Project

‘Melanoma Detection’ is a Fraunhofer AICOS’ internal study which intends to create a prototype for a patient-oriented system of skin lesion analysis using a smartphone.

As skin cancers constitute nowadays the most common malignancies in the Caucasian population, with incidences that are rising faster than any other type of cancer, the early detection of skin cancers is of vital importance – and this is the driving force of the ‘Melanoma Detection’ project.

The idea of analyzing skin lesions using a smartphone was presented by Luis Rosado in his participation in the Fraunhofer Portugal Challenge 2010, which led to Fraunhofer AICOS inviting him to further develop the concept as an AICOS’ researcher.

The work aims to implement a self-monitoring system that collects, processes and storages information of skin lesions through its automatic classification. This classification is based on the ABCD rule, which consists in 4 visual features highly relevant for skin cancer detection (asymmetry, border, color and differential structures).

Moreover, the classification of the skin lesions is made through a supervised method, based on a dataset of images previously classified by specialists of Instituto Português de Oncologia do Porto.

Recently, an R&D joint project proposal was submitted to FCT based on the work “SMARTSKINS: A Novel Framework for Supervised Mobile Assessment and Risk Triage of Skin lesions via Non-invasive Screening”.

This project aims to design and effectively develop and implement a mobile-based risk triage framework with the active involvement of a community of dermatology specialists on the supervision and validation of the framework.

The project arises from cooperation between researchers from different fields of Engineering and Mathematics from INEGI-LAETA and Fraunhofer Portugal AICOS, in joint collaboration with medical specialists from the Dermatology Department of the Portuguese Institute of Oncology of Porto.


Source: 20.06.2012, Fraunhofer Portugal