Melanoma is a critical variety of skin cancer that, if not caught in time, can be fatal. With the increasing availability of digital imaging technology, Algorithms for machine learning and deep learning have been created to help with melanoma early detection. There is a need for an automated system to evaluate a patient’s risk of melanoma using photos of their skin lesions recorded using a typical digital image due to the costs for dermatologists to screen every patient. Machine learning methods like feedforward backpropagation neural network (FFBNN), support vector machines (SVM), random forests, and k-nearest neighbors (KNN) have successfully able to detect melanoma. Even better performance has been achieved with deep learning algorithms such as CNN, U-Net, and Mask R-CNN. Such algorithms can speed up the accuracy of melanoma diagnosis and may be even earlier identification leading to better treatment outcomes.