Melanoma and Nevus Skin Lesion Classification Using Handcraft and Deep Learning Feature Fusion via Mutual Information Measures

Abstract

In this paper, a new Computer-Aided Detection (CAD) system for the detection and classification of dangerous skin lesions (melanoma type) is presented, through a fusion of handcraft features related to the medical algorithm ABCD rule (Asymmetry Borders-Colors-Dermatoscopic Structures) and deep learning features employing Mutual Information (MI) measurements. The steps of a CAD system can be summarized as preprocessing, feature extraction, feature fusion, and classification. During the preprocessing step, a lesion image is enhanced, filtered, and segmented, with the aim to obtain the Region of Interest (ROI); in the next step, the feature extraction is performed. Handcraft features such as shape, color, and texture are used as the representation of the ABCD rule, and deep learning features are extracted using a Convolutional Neural Network (CNN) architecture, which is pre-trained on Imagenet (an ILSVRC Imagenet task). MI measurement is used as a fusion rule, gathering the most important information from both types of features. Finally, at the Classification step, several methods are employed such as Linear Regression (LR), Support Vector Machines (SVMs), and Relevant Vector Machines (RVMs). The designed framework was tested using the ISIC 2018 public dataset. The proposed framework appears to demonstrate an improved performance in comparison with other state-of-the-art methods in terms of the accuracy, specificity, and sensibility obtained in the training and test stages. Additionally, we propose and justify a novel procedure that should be used in adjusting the evaluation metrics for imbalanced datasets that are common for different kinds of skin lesions.

Skin Lesion Classification Based on Deep Convolutional Neural Networks Architectures

“…In skin lesion images, captured images typically carry noises encompassing uneven illumination, skin surface light reflection, and hair. Such noises may affect the performance of segmentation undesirably and thus, need to be resolved [6,35]. One of the techniques to resolve noises in preprocessing is the usage of filters.…”

“…Despite this, caution is exercised due to limitations in dermoscopic images that need to be addressed. Since the emergence of image processing techniques, improvements have been pursued to improve Computer-Aided Detection (CAD) systems and approaches in Pigmented Skin Lesion (PSL) segmentation and classification, leading to easing patients' early diagnoses with less invasive or traumatizing medical procedures [6,7]. Stateof-the-art advances in machine learning, particularly deep neural networks, have made great strides in various areas.…”

Malignant Melanoma Classification Using Deep Learning: Datasets, Performance Measurements, Challenges and Opportunities

“…Almaraz et al [93] proposed a method which fuses deep learning features with handcrafted features via mutual information measures to extract the important information from both type of features. The proposed method uses several methods such as linear regression, support vector machine and relevant vector machines for classification.…”

FCN-Based DenseNet Framework for Automated Detection and Classification of Skin Lesions in Dermoscopy Images

“…The system achieved an overall accuracy of 86.21% when evaluated on skin lesion image datasets. Another CNN model pre-trained on Imagenet was utilized by Almaraz-Damian et al [53] for the extraction and segmentation of both handcraft and deep learning features. The system achieved similar results of 87% accuracy with the models developed by El-Khatib et al…”

“…Preprocessing steps such as lesion image enhancement, filtering, and segmentation were utilized on lesion images to acquire the Region-of-Interest (ROI) by Jose-Agustin et al [53]. Both handcraft features and deep learning features were extracted.…”