Akademik Veri Yönetim Sistemi
NWSA Academic Journals, cilt.20, ss.129-147, 2025 (TRDizin)
Breast cancer is the commonly diagnosed cancer in women all over the world, and its prevalence is constantly increasing despite significant advancements in the area of early diagnosis and individual treatment approaches. Nevertheless, present-day workflows in diagnostic interventions are struggling with problems such as overdiagnosis in populations with low risks, growing workloads among radiologists and pathologists, and inconsistencies in the interpretation of the findings of the imaging and pathological studies. In that regard, artificial intelligence (AI) has proven to be an effective solution to these drawbacks by enhancing image analysis, automating the working processes that consume a lot of labor, and facilitating clinical decision-making. This paper provides a narrative review of the recent AI implementation in breast cancer screening and diagnosis, including malignancy detection and classification, tumor segmentation, prediction of molecular subtype, and recurrence or metastatic risk. The data sources are analyzed both in imaging and non-imaging, which are mammography, ultrasound, magnetic resonance imaging (MRI), histopathology, clinical variables, and multi modal data integration. Also, the reviewed articles identify explainable artificial intelligence (XAI) methods, including SHAP, Grad-CAM, and LIME, as central to improving the transparency, interpretability, and confidence clinicians have in AI-assisted systems. On the whole, the current evidence indicates that AI-based tools have the potential to increase the level of diagnostic accuracy, minimize inter-observer variability, and provide a personalized risk evaluation and treatment planning. However, there are still multiple obstacles to widespread clinical implementation such as heterogeneity of datasets, a lack of external and prospective validation, interpretability issues, and constraints based on real-world application. Future studies must, therefore, focus on the creation of more and better-quality data, standard assessment guidelines, solid explainability models, and future clinical trials to allow the safe, productive, and fair integration of AI into regular breast cancer care.