An Overview of Automated Breast Cancer Diagnosis Based on Mammographic Images Using Deep Learning Methods

Breast cancer remains one of the leading causes of mortality among women worldwide, making early and reliable diagnosis a critical healthcare priority. With the increasing availability of mammographic imaging and advances in artificial intelligence, deep learning has emerged as a transformative approach for computer-aided diagnosis. This paper presents a comprehensive review of automated breast cancer detection methods using mammographic images, focusing specifically on deep learning-based techniques. Unlike original research studies, this review compiles and synthesizes findings from recent works (2018-2025), highlighting commonly used datasets, preprocessing strategies, model architectures, and evaluation metrics. Comparative insights are provided to understand why certain models outperform others under specific conditions, along with a discussion of research gaps such as dataset bias, interpretability, federated learning, and domain adaptation. The novelty of this work lies in providing an organized overview and critical synthesis of existing studies rather than proposing a new model or experiment. This survey aims to guide future research toward addressing limitations in current methodologies and fostering clinically reliable breast cancer diagnostic systems.