Automated Breast Cancer Classification from Medical Imaging using MobileNet

Abstract

Breast cancer remains one of the leading health threats for women worldwide, where early detection plays a critical role in improving patient prognosis and survival rates. Traditional diagnostic approaches, which are heavily based on the manual interpretation of medical images, often suffer from limitations in precision, efficiency, and scalability. In this study, we propose an automated diagnostic framework that utilizes deep learning techniques to detect and classify breast cancer from medical imaging data. Specifically, we focus on MobileNet, a lightweight convolutional neural network architecture, and evaluate its performance compared to classical CNNs and Vision Transformers (ViT). We showed that MobileNetV3, due to its efficiency and compact architecture, offers a compelling solution for low-resource clinical environments without compromising accuracy. Following rigorous preprocessing and optimized data loading strategies, all models were trained and validated on the MIAS dataset using standard evaluation metrics, including precision, sensitivity, and the F1 score. Our findings demonstrate that MobileNet offers a balanced trade-off between computational efficiency and classification performance, making it a suitable candidate for deployment in real-time clinical settings. The results underscore the potential of integrating deep learning-assisted diagnostic systems to enhance early detection and support clinical decision-making in breast cancer management.