A Bibliometric Analysis of CT Image Segmentation for Liver Cancer Screening Using Artificial Intelligence (1998--2026): A Review

Liver cancer is a global challenge, with increasing mortality rates. This article presents a detailed bibliometric analysis of the integration of artificial intelligence (AI) methods to segment CT images and, subsequently, screen for liver cancer. We analyzed 2029 articles indexed in the recognized Scopus database, which covers the period between 1998 and 2026 (data collected on July 25, 2026), using a bibliometric analysis and visualization software called VOSviewer, in combination with complementary network analysis methods to explore co-occurrence networks, authors' scientific performance, and influence on citations. Major findings include the identification of researchers, research organizations, and countries contributing to the development of AI-based liver cancer screening techniques, with China, India, and the United States being the main contributors. Similarly, our study highlights the most influential publications and journals in this field, and also the accelerated increase in research output after 2019, mainly supported by deep learning algorithms. Five thematic clusters emerged from keyword co-occurrence analysis: clinical and radiomic evaluation, deep learning-based CT segmentation, quality imaging and contrast-enhanced imaging, neural networks, and AI applications in MRI-based oncology imaging. The analysis also highlights gaps in the existing and burgeoning literature for future investigations, most notably explainable AI, integration of multimodal imaging data, and testing in large-scale clinical settings.