 This paper proposes an integrated computer-aided diagnosis, CAD, system for detecting and categorizing malignant melanomas from dermoscopic images. The system uses a combination of features extracted from the images, including histograms of oriented gradients, HOG, and local binary patterns, LBP. These features are then used to train support vector machines, SVMs, K-nearest neighbors, KNNs, and gentle adaptive boosting, GAB, algorithms to accurately classify the lesions as either melanoma or benign. The system was tested on a publicly available Dermoscopy image dataset and compared against other state-of-the-art approaches. Results showed that the proposed CAD system outperformed all other approaches in terms of accuracy, specificity, and sensitivity. This article was authored by Sami Bekit, Shtwe Al-Saba'i, AML El-Nikor, and others.