BCNet: A Novel Deep Learning Model for Enhanced Breast Cancer Classification Using Histopathological Images
Mikiyas Amare Getu, Chao Lu, Yumeng Liu, Anam Mehmood, Zoya Iqbal, Xianbin Zhang · IntechOpen eBooks · 2025
Breast cancer is the most commonly diagnosed cancer among women and a leading cause of cancer-related deaths globally, necessitating accurate and timely diagnosis for effective treatment. Histopathological examination of breast tissue samples is the gold standard for diagnosing breast cancer, but this process is subjective, time-consuming, and reliant on the level of the pathologist’s expertise. This study introduces a new deep learning model, Breast Cancer Network (BCNet), specifically designed to detect and classify breast cancer. BCNet, a 22-layer convolutional neural network (CNN), aims to enhance diagnostic accuracy by capturing high-level discriminative features tailored to breast tissue images. The BCNet model was evaluated against established CNN models, demonstrating superior performance, achieving an accuracy of up to 99.8% for binary classification and 99.6% for multi-class classification at different magnifications. These results highlight BCNet’s robustness and potential to reduce diagnostic errors and assist pathologists. Future research should explore the generalizability of BCNet across larger datasets and its integration into clinical workflows to provide real-time, AI-assisted diagnostic support.