Preprocessing of Skin Images and Feature Selection for Early Stage of Melanoma Detection using Color Feature Extraction

Abstract

Preprocessing is an essential part to achieve good segmentation since it affects the feature extraction process. Melanoma have various shapes and their extracted features from image are used for early stage detection. Due to the fact that melanoma is one of dangerous diseases, early detection is required to prevent further phase of cancer from developing. In this paper, we propose a new framework to detect cancer on skin images using color feature extraction and feature selection. The default color space of skin images is RGB, then brightness is added to distinguish the normal and darken area on the skin. After that, average filter and histogram equalization are applied as well for attaining a good color intensities which are capable of determining normal skin from suspicious one. Otsu thresholding is utilized afterwards for melanoma segmentation. There are 147 features extracted from segmented images. Those features are reduced using three types of feature selection algorithms: Linear Discriminant Analysis (LDA), Correlation based Feature Selection (CFS), and Relief. All selected features are classified using k-Nearest Neighbor (k-NN). Relief is known to be the best feature selection method among others and the optimal k value is 7 with 10-cross validation with accuracy of 0.835 and 0.845, without and with feature selection respectively. The result indicates that the frameworks is applicable for early skin cancer detection.

DeepSkin: Robust Skin Cancer Classification Using Convolutional Neural Network Algorithm

“…A study discussed the process of preprocessing skin images and selecting features for melanoma diagnosis. The experiment result showed an accuracy of 0.835 and 0.845 [16]. Another paper proposed a regionbased ROI method to classify melanoma with nevus cancer accuracy 97.9% and 97.4% [5].…”

COVID-19 Chest X-Ray Detection Performance Through Variations of Wavelets Basis Function

“…Feature extraction using wavelet transform aims to obtain feature vectors from normal chest X-Ray images and chest X-Ray images diagnosed with COVID-19. The resulting feature vectors are then divided using the k-fold cross-validation method to balance the composition of training data and test data [18]. This feature vector will be used at the classification stage such as in research [19], so that the SVM method can distinguish the two types of images.…”

“…The main objective of this method is to keep the amount of training data and test data balanced such that overfitting does not occur. This method can guarantee that the classification algorithm is applied to each data in the dataset and that all data in the dataset is tested as training and testing data [18]. K-fold cross-validation works by dividing the initial sample size into k subsamples.…”

“…where t is the number of samples that are true when tested, while n is the total number of samples in the dataset [18].…”

Word Cloud of UKSW Lecturer Research Competence Based on Google Scholar Data

“…Machine Learning (ML) has been used in various applications in big data processing such as in the manufacture of artificial intelligence, processing COVID-19 data on the death rate in South Korea [3], diabetes data processing [4], and various other applications. Machine Learning (ML) is translated as machine learning in this article as part of data collection, for example, the study of large amounts of text analysis (big data) [5], pattern recognition, and computational theory in artificial intelligence for the initial process of analyzing skin images and the selection of Melanoma features by staining extraction [6] which builds on the theory, method and application of domains related to big data [7]. Likewise, ML is also often related to data mining where data mining explores text data analysis [8].…”