A Framework for Skin Cancer Prediction Using CNN Ensembles

Skin cancer is among the deadliest diseases engulfing the world. The early detection of this disease may provide a great relief to the patients and healthcare systems worldwide. Recently, Machine Learning (ML)-based detection techniques have become popular in early skin cancer prediction. This paper comprehensively presents an overview of the latest research in the field of skin cancer prediction using ML techniques, especially the use of Convolutional Neural Networks (CNN) for the said purpose. Various methodologies employed in the past have been discussed along with the outcomes achieved, limitations, and advantages of these predictive models. A detailed review of the latest key references in the area is also discussed. At the end, a way of achieving ensemble using CNN methods for skin care has also been discussed, which we will implement in the next phase of our research.