Automated Classification Breast Cancer System From Histopathologic Scans Of Lymph Node Sections Using Deep Learning

Abstract

Breast cancer is a prevalent and potentially life-threatening disease that affects a significant number of people globally. Timely detection and accurate diagnosis are crucial in improving treatment outcomes and reducing mortality rates associated with this condition. Histopathology, a well-established diagnostic method, has been widely used for detection and determining breast cancer stage. However, the traditional manual process of histopathological analysis is labor-intensive and prone to human errors. To address these challenges, this study proposes a novel approach utilizing convolutional neural network (CNN) methods to classify breast cancer from histopathological images of lymph node sections. By leveraging the power of transfer learning, a pre-trained DenseNet architecture is employed and fine-tuned to achieve the best possible results. The developed model is then deployed in a web-based application, enabling pathologists to access and utilize it conveniently, benefiting from its simplicity and rapid performance.