Abstract
Melanoma, the most dreadful cancer of the skin with a high mortality rate is initially diagnosed visually by a clinical screening, dermoscopic analysis, a biopsy, and histopathological examination. It becomes dangerous with delays in diagnosis and early treatment. Recent developments in image processing techniques help in detecting melanoma in an efficient way as it is a difficult job due to fine-grained variability in the lesion. This paper looks into a new classification procedure for analyzing lesion irregularities using Particle Swarm Optimized Artificial Neural Network. In this research paper, the color features from the lesion are extracted and classification is done using the PSO-ANN classifier. Receiver Operating Characteristics obtained from marking false positive and true positive rates have a vital role in analyzing the diagnostic potential of the computer-aided diagnosis system. Classification techniques applied to the ISIC database indicate 0.96853 as the area under the curve with a specificity of 90.0%, a sensitivity of 94.07%, and an accuracy of 93.04%.
)