Recent trends and analysis in artificial intelligence based detection of triple negative breast cancer

Abstract Triple-negative breast cancer (TNBC) is an aggressive and heterogeneous breast cancer subtype for which artificial intelligence (AI) is increasingly being investigated across diagnostic, prognostic, and treatment response applications. This study presents a comparative bibliometric analysis of AI-related TNBC research indexed in Scopus (n=1,037), Web of Science (n=557), and PubMed (n=763) from 2013-2026. Publication trends, citation impact, leading contributors, international collaboration, knowledge structure, and thematic evolution were examined using database-appropriate bibliometric and network analyses. Using complete year data through 2025, publication output showed statistically significant increasing trends across all three databases, with 2020-2025 compound annual growth rates of 50.13% for Scopus, 49.78% for Web of Science, and 42.84% for PubMed. The United States emerged as the dominant international collaboration hub, while collaboration between the United States and China formed the strongest bilateral link. Thematic evolution showed an early emphasis on machine learning, followed by expansion of deep learning and radiomics and, more recently, parallel imaging/computational and clinical/biological themes involving MRI, treatment response prediction, the tumor microenvironment, immunotherapy, and precision medicine. Scopus co-citation and bibliographic coupling analyses further identified established intellectual foundations and contemporary research fronts. Overall, AI-TNBC research is expanding rapidly, while data heterogeneity, external validation, interpretability, and clinical translation remain important challenges.