Review — Artificial intelligence in breast surgery for older women with breast cancer: A scoping review of the past ten years

Abstract

Despite older women comprising a significant proportion of breast cancer cases, they remain underrepresented in clinical trials and research. This scoping review examines how artificial intelligence (AI) has been applied in breast surgery for older women with breast cancer over the past decade. A systematic search of PubMed, Scopus, Web of Science, and IEEE Xplore databases was conducted for studies published between January 2015 and February 2025. Studies were included if they investigated AI applications in breast surgery and provided age-stratified analysis for women aged 65 and above. The findings were thematically analyzed to identify trends and gaps in the field. Six studies met the inclusion criteria and were categorized into two themes: AI-Driven Diagnostics and Risk Prediction (two studies) and AI-Guided Treatment and Decision Support (four studies). Deep learning (DL)- and radiomics-based models demonstrated superior accuracy in predicting axillary lymph node (ALN) metastasis (area under curve [AUC] 0.906) compared to traditional clinical models. AI-guided treatment decision support tools showed improved survival outcomes, with one study reporting a 12% lower mortality rate for patients following AI recommendations. However, no studies specifically addressed AI applications in intraoperative decision-making for older women. While AI shows promise in improving diagnostic accuracy and treatment planning for older women with breast cancer, significant gaps exist, particularly in intraoperative applications. Future research should focus on prospective validation of AI models, standardization of AI development, and investigation of AI’s potential in real-time surgical decision-making for this vulnerable population.

Ähnliche Arbeiten