本文采用文献计量学分析手段，深入探讨肥胖管理与人工智能融合研究的知识结构变迁及其可视化表征，识别关键研究议题、发展脉络与前沿方向，为内分泌学临床应用及跨学科整合提供循证依据。

以Web of Science核心数据库为数据源，运用"Obesity" OR "Weight Loss" AND "Artificial Intelligence" OR "Machine Learning"等检索策略，涵盖2011-2025年期间的Article与Review文献类型。借助CiteSpace工具开展分析，设定年度时间切片，采用Top 50 per slice筛选准则，构建词汇共现关系网、聚类谱系图、时间演进图及作者机构协作图谱。

文献数量呈现指数级增长态势，历经初期积淀、中期快速发展及近期爆发式增长三个阶段。词汇共现网络包含651个节点与2935条连接，网络密度为0.0139，高频术语涵盖"obesity"（1 241次）、"machine learning"（403次）；聚类结果揭示10个研究模块（Q=0.4151，S=0.6977），主要研究维度聚焦AI技术应用（深度学习）、机理阐释（代谢综合征）及治疗干预（减重手术）。时序演进显示研究重心从早期流行病学基础向中期临床应用萌芽，再到近期精准融合发展，如肠道菌群与AI预测模型结合。机构合作网络中Harvard University占据核心地位，中国科研院所展现追赶发展态势。

该研究领域的知识演进体现了AI技术从风险评估向个性化治疗的模式革新，当前热点转向可解释性AI与多模态数据融合。未来研究应加强国际协作、推进临床随机对照研究验证，关注伦理规范与算法偏见问题，推动全球肥胖防治效能提升与精准医疗发展。

This paper employs bibliometric analysis to deeply explore the evolution of the knowledge structure and its visual representation in the integration of obesity management and artificial intelligence (AI), identify key research topics, development trajectories, and frontier directions, and provide evidence-based support for the clinical application of endocrinology and interdisciplinary integration.

Using the Web of Science core database as the data source, the search strategy of "Obesity" OR "Weight Loss" AND "Artificial Intelligence" OR "Machine Learning" was adopted, covering Article and Review types from 2011 to 2025. CiteSpace was utilized for analysis, with annual time slices set and the Top 50 per slice screening criterion applied to construct co-occurrence networks of terms, cluster lineage diagrams, temporal evolution diagrams, and author-institution collaboration maps.

The number of documents showed an exponential growth trend, going through three stages: initial accumulation, rapid development in the middle period, and explosive growth in the recent period. The co-occurrence network of terms included 651 nodes and 2,935 connections, with a network density of 0.0139. High-frequency terms included "obesity" (1 241 times) and "machine learning" (403 times). The clustering results revealed 10 research modules (Q = 0.4151, S = 0.6977), with the main research dimensions focusing on the application of AI technology (deep learning), mechanism explanation (metabolic syndrome), and treatment intervention (weight loss surgery). The temporal evolution showed that the research focus shifted from the early epidemiological foundation to the budding of clinical application in the middle period, and then to the recent precise integration and development, such as the combination of gut microbiota and AI prediction models. In the institutional collaboration network, Harvard University held a core position, while Chinese research institutions demonstrated a catching-up development trend.

The knowledge evolution in this research field reflects the model innovation of AI technology from risk assessment to personalized treatment. Current hotspots have shifted to explainable AI and multimodal data fusion. Future research should strengthen international collaboration, promote clinical randomized controlled trials for validation, pay attention to ethical norms and algorithm bias issues, and enhance the global effectiveness of obesity prevention and treatment and the development of precision medicine.