AN INTELLIGENT BREAST CANCER DIAGNOSIS USING DEEPSEGMENTATION BASED ALEXNET WITH RANDOM FORESTCLASSIFICATION MODEL
T. Sathya Priya, T. Ramaprabha · Journal of Critical Reviews · 2020
Breast cancer is treated as an important health issue among women globally. When the abnormalities in breast cancer are identified in the earlier stage, the survival rate can be considerably increased. Mammogram is treated as a proficient and widely employed model to detect and screen breast cancer. Deep learning (DL) models can be used by the radiology experts to make a precise diagnosis and helps to attain enhanced predictive outcome. This study proposes a novel deep segmentation based AlexNet with Random Forest (RF) model called DS-ANRF to detect and classify the existence of breast cancer from mammogram images. The presented DS-ANRF model comprises four processes including namely preprocessing, Faster Region based Convolution Neural Network (R-CNN) (Faster R-CNN) with Inception v2 model based segmentation, AlexNet based feature extraction and RF based classification. A comprehensive simulation process takes place and the goodness of the DS-ANRF model is validated using Mini-MIAS dataset. The experimentation outcome ensured the outstanding performance of the DS-ANRF model over the compared methods.