Deep Mole Skin Cancer Detection Using Convolutional Neural Network(CNN) Model

The creation and assessment of a Convolutional Neural Network (CNN) model for the early diagnosis of skin cancer is the main goal of this research project. With the help of the Skin Cancer MNIST HAM10000 dataset, the model attempts to achieve high classification accuracy for mole images. A customized CNN architecture with convolutional layers, max pooling, batch normalization, and dropout layers is created as part of the study. Photographs of moles can be uploaded and classified via an easy-to-use Flask and Python web interface. The system's performance is assessed using a range of criteria, indicating its potential for use in the real world in the early identification of skin cancer. In addition, the research tackles important issues such user interface design and model optimization, offering a complete solution for practical implementation.