Skin cancer stands as the most prevalent cancer type in the United States, with an annual diagnosis of over 5 million cases. The timely identification and treatment of skin cancer play a pivotal role in enhancing patient prognosis. Machine learning has emerged as a promising avenue for aiding in the early detection of skin cancer, and the Radial Basis Function (RBF) approach has gained popularity as a technique in this regard. RBF networks, a subtype of artificial neural networks, utilize radial basis functions as activation functions. These functions, represented by bell-shaped curves, yield output values based on the distance between the input and the function's center. RBF networks have demonstrated effectiveness in classifying intricate data, making them well-suited for the detection of skin cancer. Among skin cancers, melanoma, originating from melanocytes—the pigment-producing cells—is the most perilous form and has been increasingly identified as a leading cause of death. Melanoma presents itself with regions appearing black or brown due to the melanin pigment. However, some melanomas do not produce melanin, manifesting in pink, tan, or white colors. Therefore, an efficient melanoma detection technique becomes imperative. RBFN, falling under the category of Artificial Neural Networks (ANN), has found utility in various classification problems in science and engineering. The Back Propagation (BP) algorithm, widely used in ANN, suffers from drawbacks such as slow error rate convergence and susceptibility to getting stuck at local minima. To address these issues, a recent MATLAB tool has been employed for implementing the proposed system, designed using High-Level Synthesis (HLS) design methodology.