Detection and Analysis of Skin Cancer using different Machine Learning for enhanced accuracy

The main aim of this research work is to analyze the performance of skin cancer with respect to the finding of accuracy level. In this analytic research, 2230 images were employed from Kaggle datasets and ISIC database. The images comprised 350 nevus, 420 melanoma, 150 vascular lesion, 120 actinic keratoses (AK), 190 Squamous cell carcinoma, 380 basal cell carcinoma (BCC), 80 seborrheic keratosis, 450 pigmented benign keratosis, and 90 dermatofibroma. In this research, many machine learning algorithms and deep neural algorithm are considered. Different machine learning algorithms are convolutional neural network (CNN), Recurrent Neural Network (RNN), Support vector machines (SVMs), K-Nearest Neighbors Algorithm (KNN), Residual neural network (ResNet), Long Short Term Memory (LSTM), and Artificial neural network (ANN). In which CNN performs better in comparison with all other algorithm models. The average accuracy obtained for CNN in terms of classification skin cancer image is 91.71 %, 93.74 % for specificity.