Advanced Image Processing Techniques for Breast Cancer Detection Using Multi-Feature Extraction Methods

Abstract

Breast cancer remains a significant global health challenge, necessitating the development of advanced image processing techniques to enhance early detection and improve patient outcomes. In this study, we propose a novel approach utilizing multi-feature extraction methods for breast cancer detection, aiming to leverage diverse imaging characteristics to enhance accuracy and reliability. Our method incorporates a comprehensive set of features extracted from mammographic images, including texture, shape, intensity, and spatial information. By integrating these diverse features, our approach aims to capture the complex and subtle patterns indicative of breast cancer, thus enabling more accurate detection compared to traditional methods. To extract texture features, we employ advanced techniques such as gray-level co-occurrence matrices (GLCM) and local binary patterns (LBP), which enable the characterization of texture variations within mammographic images. Additionally, shape features are extracted using techniques such as contour analysis and geometric descriptors, providing valuable information about the morphological characteristics of lesions. Furthermore, intensity-based features are extracted to capture variations in pixel intensity distribution, while spatial features are computed to analyze the spatial arrangement of image structures. By combining these different types of features, our approach aims to provide a more comprehensive representation of the underlying tissue properties, facilitating more accurate discrimination between benign and malignant lesions. We evaluate the performance of our proposed method using a dataset comprising mammographic images from diverse patient populations. Experimental results demonstrate that our approach achieves superior performance compared to existing techniques, with high sensitivity and specificity in detecting breast cancer lesions.