AI Pancreatic Cancer Research Booms Worldwide

Research on artificial intelligence in pancreatic cancer has taken off since 2019. A new study looked at 583 papers published between 1998 and 2026. The data comes from the Web of Science Core Collection. Only English articles and reviews were included. The analysis used tools like Biblioshiny, VOSviewer, and CiteSpace.

China leads in sheer number of publications with 222 papers. The United States follows with 150 but holds the highest centrality score at 0.39. Shanghai Jiao Tong University produced the most papers at 20. Harvard University tops institutional centrality at 0.22. The work spans 59 countries and over 1,200 institutions.

Frontiers in Oncology published the most articles at 32. Early research focused on neural networks and texture analysis. Now the hot topics are radiomics, deep learning, early detection, and risk prediction. Liquid biopsy, tumor biology, treatment response, and precision oncology are also big themes.

Experts say moving to clinical use needs more work. Prospective multicenter validation is a must. Datasets must represent diverse patients. Reporting has to be transparent. Models need calibration and fairness checks. Workflow integration and proof of better patient outcomes are essential.

The field is growing fast and becoming more multidisciplinary. Yet productivity and collaboration are still uneven across regions. Critical thinking asks whether current trends will close gaps or widen them.