This document discusses using pattern recognition and artificial neural networks (ANNs) to detect malignancy, specifically melanoma skin cancer. It describes preprocessing dermoscopy images to remove noise, then implementing an ANN with 12 neurons in each layer to classify images as cancerous or non-cancerous based on 12 selected features. After training the ANN, it is tested on new data for decision making. The method provides efficient classification compared to alternative gradient descent approaches that may result in incorrect predictions. Publicly available skin cancer data is used to train and validate the ANN model.