Global trends and hotspots of the sepsis machine learning prediction model: a bibliometric analysis

Sepsis is a systemic syndrome caused by infection-induced organ dysfunction, and machine learning prediction models have shown important value in the early warning of sepsis. However, the overall research trends, main research topics, and hotspot directions in this field are still not clear. To explore the current development of machine learning prediction models for sepsis, this study conducted a bibliometric analysis of 857 publications and discussed current research hotspots and future development directions. The data were obtained from the Web of Science Core Collection (WoSCC), and only English-language publications were included. Microsoft Excel was used to collect and analyze bibliometric indicators, including annual publication output and citation frequency. In addition, VOSviewer(v1.6.20), CiteSpace(v6.3.R1), and the online bibliometric platform ( https://bibliometric.com/ ) were used to visualize research trends and collaboration networks. From 1990 to 2024, a total of 857 publications related to machine learning prediction models for sepsis were indexed in the Web of Science Core Collection (WoSCC). Among them, the most frequently cited article was “Big data in health care: using analytics to identify and manage high-risk and high-cost patients”. “Sepsis”, “mortality”, and “procalcitonin” were among the most frequently occurring keywords in this field, indicating sustained research interest in sepsis outcomes, risk assessment, and biomarker-related studies. However, the prominence of biomarker-related keywords should be interpreted cautiously, as they may reflect broader sepsis research rather than exclusively machine learning–based prediction studies. This study clarified the publication trends, major contributors, and research hotspots in the field of machine learning prediction for sepsis, providing valuable references for future research and clinical translation, and further promoting the more effective application of machine learning technologies in the early diagnosis and precision treatment of sepsis.