A reliable and secure analysis of critical data set An early detection of breast cancer using deep learning techniques

Abstract

Breast cancer is one of the most common types of cancer that affects newlinewomen globally. Early detection and accurate diagnosis of the disease are newlinecrucial for effective treatment and management. In recent years, deep learning newlinetechniques have shown great potential in medical image analysis, including newlinebreast cancer detection. The use of cloud-based storage and security is also newlinegaining popularity due to its cost-effectiveness and efficiency in storing and newlineprocessing large amounts of medical data. This thesis aims to develop an newlineefficient framework for breast cancer detection using deep learning strategies newlineand provide a cloud-based storage and security solution. newlineThe proposed framework consists of several stages, including pre newlineprocessing, segmentation, feature extraction, feature selection, and newlineclassification. The pre-processing stage involves the use of Gaussian filtering newlineto reduce noise and enhance the quality of the input images. The segmentation newlinestage uses Cauchy distribution-based techniques to separate the breast region newlinefrom the background and to detect potential tumors. The feature extraction newlinestage uses shearlet transforms to capture the local and global characteristics of newlinethe breast tissues, which can help to differentiate between malignant and newlinebenign tissues. The feature selection stage uses entropy-based principal newlinecomponent analysis to reduce the dimensionality of the feature vectors and to newlineselect the most informative features for classification. newline