Melanoma Classification Using a Novel Deep Convolutional Neural Network with Dermoscopic Images

Abstract

Automatic melanoma detection from dermoscopic skin samples is a very challenging task. However, using a deep learning approach as a machine vision tool can overcome some challenges. This research proposes an automated melanoma classifier based on a deep convolutional neural network (DCNN) to accurately classify malignant vs. benign melanoma. The structure of the DCNN is carefully designed by organizing many layers that are responsible for extracting low to high-level features of the skin images in a unique fashion. Other vital criteria in the design of DCNN are the selection of multiple filters and their sizes, employing proper deep learning layers, choosing the depth of the network, and optimizing hyperparameters. The primary objective is to propose a lightweight and less complex DCNN than other state-of-the-art methods to classify melanoma skin cancer with high efficiency. For this study, dermoscopic images containing different cancer samples were obtained from the International Skin Imaging Collaboration datastores (ISIC 2016, ISIC2017, and ISIC 2020). We evaluated the model based on accuracy, precision, recall, specificity, and F1-score. The proposed DCNN classifier achieved accuracies of 81.41%, 88.23%, and 90.42% on the ISIC 2016, 2017, and 2020 datasets, respectively, demonstrating high performance compared with the other state-of-the-art networks. Therefore, this proposed approach could provide a less complex and advanced framework for automating the melanoma diagnostic process and expediting the identification process to save a life.

Keywords: classification; deep convolutional neural networks; melanoma; skin cancer.

Figure 1

Different types of skin lesions: (a) MEL, (b) BEN, (c) NV, and (d) SK.

Figure 2

Augmented data samples using translation, rotation, and scaling.

Figure 3

The design of the proposed network, LCNet.

Figure 4

Classification accuracy and loss curves of the LCNet with the number of epochs on the validation set (a) MEL vs. BEN lesion classes ISIC 2016, (b) MEL vs. SK and NV lesion classes ISIC 2017, and (c) MEL vs. BEN lesion classes ISIC 2020.

Figure 5

Classification accuracy and loss curves of the LCNet with the number of epochs on the validation set (a) MEL vs. BEN lesion classes ISIC 2016 (b) MEL vs. SK and NV lesion classes ISIC 2017 (c) MEL vs. BEN lesion classes ISIC 2020.