The health and life of women’s are seriously threatened by breast cancer. In female diseases, morbidity breast cancer and mortality breast cancer are ranked as first and second. The breast cancer’s mortality can be reduced by effective lump detection in early stages. The early detection, diagnosis and treatment of breast cancer are enabled by a mammogram-based computer-aided diagnosis (CAD) system. But unsatisfied results are produced by available CAD systems. Feature fusion-based breast CAD method is proposed in this work, which uses deep features of convolutional neural network (CNN). Deep feature of CNN-based mass detection is proposed in the first stage. Clustering is performed by enhanced recurrent extreme learning machine (ERELM) method. Loads are forecasted using recurrent extreme learning machine (RELM) and gray wolf optimizer (GWO) is used to optimize the weights. Deep, morphological, density and texture features are extracted in the next stage. The malignant and benign breast masses are classified by developing fused feature set-based ERELM classifier. High value of efficiency and accuracy is produced by a proposed classification technique.
CITATION STYLE
Agarwal, R., & Sharma, H. (2021). A New Enhanced Recurrent Extreme Learning Machine Based on Feature Fusion with CNN Deep Features for Breast Cancer Detection. In Advances in Intelligent Systems and Computing (Vol. 1158, pp. 461–471). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-981-15-4409-5_42
Mendeley helps you to discover research relevant for your work.