Classification of Skin Cancer With Using Color-ILQP and MEETG

Abstract

Skin cancer is one of the most common forms of cancer in the world that has grown dramatically over the past decades. Malignant melanoma is the deadliest type of skin cancer. Melanocytic nevi are benign whereas melanoma is malignant. Most skin cancers are treatable in the early stages. So, rapid diagnosis and the importance of early stage can be very important to cure it and increasing day by day. Today, artificial intelligence can represent an important role in medical image diagnosis. The aim of this paper is to an auto-diagnosis system can be deployed to help dermatologists in identifying melanoma that may facilitate early detection of melanoma, and hence substantially reduce the mortality chance of this dangerous malignancy. We used image processing tools to diagnose melanoma skin cancer. In this paper, the advantage of improved local quinary pattern (ILQP) is used as texture feature extraction method and used mixture of ELM-based experts with a trainable gating network (MEETG) for skin cancer classification. Our proposed method achieved the classification accuracy on f and d datasets, 97.05% and 86.61% respectively.