Published Paper Details:

This paper is about a web application that uses deep learning to detect and classify skin lesions. The application uses a kind of dataset called the HAM10000 dataset, which has a lot of pictures of skin lesions. The system uses something called transfer learning and a new way of combining parts of the system to get better results. The model is made up of two parts: one part is a -trained ResNet50 and the other part is a custom Convolutional Neural Network. These two parts work together to get an understanding of the skin lesions. The system is very good at classifying the skin lesions with a validation accuracy of 91.2% and an F1-score of 91.7%. The system is made to be used by dermatologists to help them diagnose skin lesions. It is a web application. It is easy to use and can be used in real-time. This work shows that deep learning can be very helpful in medical image analysis and it can be used to make a tool to help healthcare practitioners.