International Research Journal of Education and Technology

A development of abnormal tissue in the breast tissue is called breast cancer. The condition may be associated with serious health concerns. Currently, the most common cancer among women in the world is breast cancer. It is also one of the leading causes of death due to cancer. Its early and accurate diagnosis will alleviate the treatment burden and enhance patient survival. To this end, the paper outlines a new approach for breast cancer diagnosis using histopathological images from the BreakHis dataset of 2,480 benign and 5,429 malignant images. The method has analyzed and interpreted the pictures using the transfer learning model using Convolutional Neural Networks (CNNs), VGG16, and VGG19. Using pre-trained models that VGG19 uses, the suggested strategy helps distinguish between benign and malignant cases with 86% accuracy. Thus, the proposed approach shows major improvements in terms of diagnostic accuracy. This provides a reliable tool to detect breast cancer and eventually, enables timely medical intervention.

KEYWORDS: Breast cancer detection, Histopathological images, BreakHis dataset, Convolutional Neural Network (CNN), VGG16, VGG19, Transfer Learning.