Skin Lesion Classification

Contemplating on the importance of early diagnosis of skin cancer, particularly, melanoma, in this work1 we propose a novel machine-learning based framework for skin lesion classification. This hierarchical framework optimizes fused features by selecting the principal components and extricating the redundant and irrelevant data. The simulation results show an accuracy exceeding 97%, and that by utilizing less than 3% features.

Artificial Intelligence is forming the world around us. We associated with things that have been touched by machine learning on a every day premise. From our melody and video suggestions to the shrewd assistants in our phones. But these are the customer applications of AI.

We are planning to construct a solution that leverages a Convolutional Neural Network to assist individuals to classify diverse sorts of skin cancers rapidly and precisely. My main objective was to form a extend that’s effortlessly open and viable. We eventually have chosen on building a web application.