ARTIFICIAL INTELLIGENCE IN BREAST CANCER SCREENING: A REVIEW

Abstract

Abstract Breast cancer is the most frequently diagnosed cancer among women worldwide, and early detection through screening remains essential for reducing mortality. Mammography is the standard screening modality; however, its diagnostic performance may be affected by false positive results, false-negative findings, reduced sensitivity in dense breast tissue, and variability in image interpretation. Recent advances in artificial intelligence (AI), particularly machine learning and deep learning, have enhanced breast cancer screening by improving lesion detection, image classification, risk prediction, and workflow efficiency. AI has been successfully applied to mammography, digital breast tomosynthesis, ultrasound, and magnetic resonance imaging, demonstrating promising improvements in diagnostic accuracy and screening performance. Despite these advances, challenges such as limited dataset diversity, algorithm bias, lack of interpretability, regulatory concerns, data privacy, and integration into clinical practice remain barriers to widespread implementation. Ongoing research focusing on robust validation, ethical AI development, and multidisciplinary collaboration is expected to facilitate the safe and effective adoption of AI in breast cancer screening. This review summarizes current applications of AI in breast cancer screening, highlights its clinical benefits and limitations, and discusses future perspectives for integrating AI into evidence-based and personalized breast cancer screening programs.