Artificial Intelligence in Cancer Oncology Through Comprehensive Bibliometric Mapping of Global Trends Impact and Conceptual Structures

Rezzy Eko Caraka,1– 4 Khairunnisa Supardi,5 Soehartati A Gondhowiardjo,5 Vijaya Isnaniawardhani,1 Prana Ugiana Gio,6 Rung Ching Chen,4 Bens Pardamean7,8 1Engineers Profession Program, Graduate School, Universitas Padjadjaran, Bandung, West Java, 45363, Indonesia; 2School of Economics and Business Telkom University, Bandung, 40257, Indonesia; 3Research Center for Data and Information Sciences, Research Organization for Electronics and Informatics, National Research and Innovation Agency (BRIN), Bandung, 40135, Indonesia; 4Department of Information Management, Chaoyang University of Technology, Taichung, 41349, Taiwan; 5Department of Radiation Oncology, Faculty of Medicine Universitas Indonesia - Dr Cipto Mangunkusumo National General Hospital, Jakarta, 10430, Indonesia; 6Department of Mathematics, Universitas Sumatera Utara, Medan, Indonesia; 7Bioinformatics and Data Science Research Center, Bina Nusantara University, Jakarta, 11480, Indonesia; 8Computer Science Department, BINUS Graduate Program – Master of Computer Science Program, Bina Nusantara University, Jakarta, 11480, IndonesiaCorrespondence: Rezzy Eko Caraka, Email [email protected] Rung Ching Chen, Email [email protected]: The integration of artificial intelligence, particularly deep learning, has transformed cancer oncology through more precise diagnostics, personalized therapies, and improved clinical decision-making. This study conducts a bibliometric mapping to capture global research trends, intellectual influence, and conceptual structures in cancer oncology and deep learning. A total of 19,627 peer-reviewed articles published between 2022 and 2025 were analyzed, retrieved from an initial dataset of 34,218 documents. Using Correspondence Analysis (CA) and Multiple Correspondence Analysis (MCA) in RStudio, we evaluated both Author Keywords and Keyword Plus. The results highlight dominant themes, including multimodal learning and radiogenomics, while also uncovering emerging directions such as transformer-based models, federated learning for cross-institutional data, and the ethical dimensions of explainable AI in clinical workflows. MCA on Keyword Plus provided stronger explanatory power (80.01% and 8.76% for the first two dimensions) compared to CA on Author Keywords (5.27% and 3.73%). This mapping reveals critical gaps—such as limited participation from low- and middle-income countries, lack of standardized datasets, and insufficient multidisciplinary collaboration. By identifying these challenges and opportunities, the study provides actionable insights for researchers, policymakers, and clinicians to advance inclusive and ethically responsible AI-driven cancer research.Keywords: cancer oncology, deep learning, artificial intelligence, bibliometric analysis, machine learning in cancer