Skincure: An Innovative Smart Phone-Based Application To Assist In Melanoma Early Detection And Prevention

摘要

Melanoma spreads through metastasis, and therefore it has been proven to bevery fatal. Statistical evidence has revealed that the majority of deathsresulting from skin cancer are as a result of melanoma. Further investigationshave shown that the survival rates in patients depend on the stage of theinfection; early detection and intervention of melanoma implicates higherchances of cure. Clinical diagnosis and prognosis of melanoma is challengingsince the processes are prone to misdiagnosis and inaccuracies due to doctorssubjectivity. This paper proposes an innovative and fully functionalsmart-phone based application to assist in melanoma early detection andprevention. The application has two major components; the first component is areal-time alert to help users prevent skin burn caused by sunlight; a novelequation to compute the time for skin to burn is thereby introduced. The secondcomponent is an automated image analysis module which contains imageacquisition, hair detection and exclusion, lesion segmentation, featureextraction, and classification. The proposed system exploits PH2 Dermoscopyimage database from Pedro Hispano Hospital for development and testingpurposes. The image database contains a total of 200 dermoscopy images oflesions, including normal, atypical, and melanoma cases. The experimentalresults show that the proposed system is efficient, achieving classification ofthe normal, atypical and melanoma images with accuracy of 96.3%, 95.7% and97.5%, respectively.