Paper Contents
Abstract
Skin cancer is one of the most prevalent types of cancer worldwide. Early and accurate detection is crucial for effective treatment and improved patient outcomes. Classification techniques have shown great potential in assisting dermatologists in the screening of images. This paper presents a two-class classification approach for skin cancer image detection using Cubic Support Vector Machine (SVM). SVMs are powerful supervised learning models that have been widely employed in various classification tasks, including medical image analysis. The proposed approach utilizes a set of carefully engineered features extracted from skin cancer images from MedNode dataset to capture relevant diagnostic information for distinguishing between Melanoma and naevus skin cancers. The texture features like Difference theoretic texture features (DTTF), First Order Statistics (FOS), Fractal Texture (FT), Grey Level Difference Statistics (GLDS), Statistical Feature Matrix (SFM), Local Binary Patterns (LBP), Segmentation Based Fractal Texture Analysis (SFTA) and Tamura features are calculated to creature a feature set for classification using cubic SVM. The evaluation of the proposed methodology is performed using accuracy, sensitivity, specificity, F1 Score and precision. The results obtained for performance parameters outperformed the state of the art methods.
Copyright
Copyright © 2024 Suryanshu. This is an open access article distributed under the Creative Commons Attribution License.
