Melanoma Detection Using Deep Learning And Convolutional Neural Network (CNN)

Date
2021
Research Supervisor
Editor
Journal Title
Volume
Issue
Journal Title
Journal ISSN
Volume Title
Abstract
In this report, we present a deep learning model to detect the skin cancer called melanoma. Deep learning has made classification tasks easily achievable with high accuracy. Melanoma is the 5th most common type of cancer in the world and early detection can help save valuable lives. We will be able to recognize if a given skin lesion is melanoma or a benign tumor from the combination of a trained dataset and deep learning algorithm. Our project involves a benchmark dataset as well as some additional data which will be provided by us. In this matter, we introduce a hybrid method for melanoma detection that can be used to examine any suspicious lesion. Our proposed system relies on the prediction of three different methods: A Convolutional Neural Network (CNN) and two classical machine learning classifiers trained with a set of features describing the borders, texture and the color of the skin lesion. The experiments have shown that using the three methods together, gives the highest accuracy level. We will also be adding augmentation to improve the accuracy more.
Description
Keywords
Citation
Department Name
Electrical and Computer Engineering
Publisher
North South University
Printed Thesis
DOI
ISSN
ISBN