BCDNet: An Enhanced Convolutional Neural Network in Breast Cancer Detection Using Mammogram Images
Bandla Raghuramaiah, Suresh Chittineni
Abstract
Breast cancer is a leading cause of death among women worldwide. The emergence of Artificial Intelligence (AI) has led to significant progress in breast cancer detection research. Early detection of breast cancer is crucial for making informed decisions about treatment and eradicating the disease. Deep Learning (DL) techniques, commonly used in computer vision, have been applied to various domains, including healthcare. The Convolutional Neural Network (CNN) is a widely used model for medical image processing, but its performance may not be optimal for a specific imaging modality without empirical study. This paper introduces an enhanced CNN model called Breast Cancer Detection Network (BCDNet), designed to be more efficient with breast mammogram images. We also propose an algorithm called Learning-Based Cancer Screening (LBCS) that leverages the BCDNet model. An empirical study using the CBID-DDSM benchmark dataset demonstrates that BCDNet outperforms many existing deep learning models, achieving the highest accuracy of 97.68%. This proposed model can be utilized for breast cancer screening in healthcare units as part of a Clinical Decision Support System (CDSS).