机器学习在食管癌领域的研究图景：基于2005~2025年的文献计量与知识图谱分析

Machine Learning in the Diagnosis and Treatment of Esophageal Cancer: A Bibliometric and Visualization Analysis (2005~2025)

摘要: 目的：本研究旨在对过去20年已发表的有关食管癌和机器学习的文献进行分析，揭示该研究领域的研究趋势和协作模式并为该领域的深入研究提供理论参考。方法：本研究以Web of Science数据库作为文献来源，利用SCImago Graphica Beta (版本1.0.51)、VOSviewer软件(版本1.6.20)、CiteSpace工具(版本号6.4.R1)进行可视化分析。结果：我们共纳入386篇文献，近五年的文献出版量显著增加，中国和美国在其中做出了重要贡献。关键词的分析显示，该领域论文聚焦于机器学习在食管癌的疾病诊断、风险预测、治疗规划及决策支持系统等方面的应用。结论：机器学习在食管癌中应用的研究已经是当前相关领域的研究热点，但仍存在一些挑战，包括各国各研究人员间的合作强度不够，还没有用于衡量的统一数据库。未来的研究应侧重于多模态数据，建立标准化模型辅助临床医生对食管癌进行更准确的诊断及治疗规划。

Abstract: Objective: This study analyzed literature from the past two decades on esophageal cancer and machine learning to identify research trends, collaboration patterns, and provide a theoretical reference for future studies. Methods: Publications were retrieved from the Web of Science database. Bibliometric analysis was performed using SCImago Graphica Beta (v1.0.51), VOSviewer (v1.6.20), and CiteSpace (v6.4.R1) to visualize data. Results: A total of 386 articles were included. Publication output increased markedly over the past five years, with China and the US being the leading contributors. Keyword analysis revealed a research focus on the application of machine learning in diagnosing esophageal cancer, risk prediction, treatment planning, and clinical decision support systems. Conclusion: Research on machine learning applications in esophageal cancer is a current hotspot. However, challenges remain, such as limited collaboration among researchers across different countries and the lack of standardized databases for evaluation. Future work should focus on integrating multi-modal data and developing standardized models to assist clinicians in accurate diagnosis and treatment planning.