Melanoma detection using a mobile phone app

Mobile phones have had their processing power greatly increased since their invention a few decades ago. As a direct result of Moore's Law, this improvement has made available several applications that were impossible before. The aim of this project is to develop a mobile phone app, integrated with its camera coupled to an amplifying lens, to help distinguish melanoma. The proposed device has the capability of processing skin mole images and suggesting, using a score system, if it is a case of melanoma or not. This score system is based on the ABCDE signs of melanoma, and takes into account the area, the perimeter and the colors present in the nevus. It was calibrated and tested using images from the PH2 Dermoscopic Image Database from Pedro Hispano Hospital. The results show that the system created can be useful, with an accuracy of up to 100% for malign cases and 80% for benign cases (including common and atypical moles), when used in the test group.