Melanoma is a form of skin canker that requires that requires early and accurate diagnosis for successful treatment. Traditional diagnosis procedure includes visual inspections for skin lesions by dermatologists with the support of dermoscopic images. However , these methods are inherently subjective and can the diagnosis may vary from person to person. The paper aims to investigate the application of Deep Neural Networks(DNN) for Melanoma detection from dermoscopic images. The outcomes of the DNN model and compared with baseline models like SVM, Random Forest and Logistic Regression in terms of accuracy, sensitivity and specificity. The results shows that the DNN model outperforms the baseline models highlighting the superior performance of the proposed approach. The paper concludes by highlighting the directions for future work in this domain.