Poster on " from pixels to prognosis - use of AI in breast cancer diagnosis"

Breast cancer remains one of the leading causes of morbidity and mortality among women worldwide, affecting over 2.3 million women annually. Early and accurate diagnosis is critical for effective management and improved survival outcomes. This systematic review explores the evolving role of artificial intelligence (AI) and computer-aided diagnosis (CAD) systems in the detection, screening, and diagnosis of breast cancer. The review synthesizes recent evidence from peer-reviewed literature published within the last five years, focusing on the application of machine learning and deep learning models in radiological and histopathological breast cancer imaging, including mammography and MRI analysis. Data were sourced from PubMed, Google Scholar, arXiv, and the Cochrane Library. The findings demonstrate that AI-based models significantly enhance diagnostic performance, with an average diagnostic accuracy of 88.9%. Conventional machine learning models exhibited higher overall accuracy (average 94.2%), while deep learning models, particularly convolutional neural networks (CNNs), showed greater sensitivity (89.8%). The Random Forest model achieved the highest reported accuracy at 96.49%. Overall, AI systems displayed high sensitivity (91.34%) and specificity (82.38%), highlighting their potential to improve consistency, speed, and reliability in breast cancer diagnosis. Despite promising results, challenges related to clinical validation, data privacy, and real-world implementation persist. This review underscores the growing importance of AI in breast cancer diagnostics while emphasizing the need for standardized validation and ethical integration into clinical practice. Study Design: Systematic review