A Hybrid Learning Model for Breast Cancer Diagnosis
Soad Khalifa Ali, Yasmina M. A. Abdalnour, Wafa El-Tarhouni, Kenz Amhmed Bozed · 2023
Mammography is a technique for detecting breast cancer early and is able to reduce specific mortality rates. However, it is often difficult to accurately identify breast cancer through mammograms due to the difficulty of interpreting it, Therefore, early identification of breast cancer with mammography is an effective strategy for patients' survival rate. In this article, Based on the principles of deep neural network and machicne learning (ML) models will propose a new technique to precisely detect breast cancer. Firstly, Data collection and preprocessing of Images; second, Feature extraction by Visual Geometry Group (VGG-19); convolution neural network. In designing the proposed breast cancer detection system, Support Vector Machine (SVM), Neural Autoregressive Distribution Estimation (NADE) models, and hybrid of SVM and NADE models of ML are used for classification process. The experimental results showed that the above proposed method is very efficient for Binary classification and provides better accuracy.