Models & Optimisation and Mathematical Analysis Journal
Volume 12, Numéro 1, Pages 22-27
2025-12-31

A Cad-like Tool Using Convolutional Neural Networks To Improve Radiologist Screening Of Breast Cancer

Authors : Kacher Abdelhafidh .

Abstract

Breast cancer remains a significant global health challenge, with early detection being the primary factor in reducing mortality. While mammography is the standard screening tool, the risk of human error in interpretation persists. This paper presents a Computer-Aided Detection (CAD) system based on a Convolutional Neural Network (CNN) to assist radiologists in identifying abnormal mammograms. Using a dataset of 2,000 mammographic images from DDSM dataset, the model was trained over 50 epochs. The experimental results obtained during a practical workshop demonstrate test accuracies of 90%, 83%, 95%, 91%, and 82% across multiple CNN architectures.

Keywords

breast cancer ; mammography ; CNN ; transfer learning ; diagnostic accuracy