Artificial Intelligence as an Assistive Tool in Breast Cancer Screening and Diagnosis in the Digital Era.

OBJECTIVE: Breast cancer (BC) is a worldwide health task, and its increasing frequency and the associated burden on healthcare professionals are also increasing. Hence, integrating machine learning (ML) and artificial intelligence (AI) into screening and diagnosis is irrefutable in the digital age. A comprehensive systematic review was conducted aimed at assessing the application of AI and ML techniques to enhance BC screening, diagnosis, classification, and tumor marker scoring. METHODS: An electronic literature search in PubMed and google scholar by following systematic review guidelines was conducted using keywords AI, cytology, histology, tumor marker expression, deep learning (DL) and ML and carcinoma breast. Articles exclusively pertaining to BC pathology were included in the study. RESULTS: This study reviews the AI tools incorporating advanced DL models, such as artificial neural networks (ANNs), support vector machines (SVMs), convolutional neural networks (CNNs), and faster R-CNN, which have demonstrated high accuracy in analyzing medical images from various modalities, including cytology and histopathology. AI applications also extend to combined imaging-pathology diagnostics and tumor biomarkers, showing promising outcomes for consistency, rapidity, and cost-effectiveness. Various Food and Drug Administration (FDA) and internationally approved AI software products for BC detection illustrate the translation potential of these technologies. CONCLUSIONS: The use of AI in the analysis of pathological samples in diagnosis, research, precision medicine, and patient care is increasing, and such assistance has great potential to reduce the workload of pathologists.