Clinical applications of artificial intelligence-driven nitric oxide: a bibliometric and scientific mapping analysis

Liu, Zhen1,#; Abulaiti, Ainikaer1,#; Zhao, Yan1,#; Zang, Junting2; Li, Guohua1,*; Shu, Li1,*; Wahafu, Paerhati1,*; Yakufu, Maihemuti1,*,#

Author Information

1Orthopedic Research Center, Sixth Affiliated Hospital of Xinjiang Medical University, Urumqi, Xinjiang Uygur Autonomous Region, China

2Department of Orthopedic Surgery, Orthopedic Center, The First Hospital of Jilin University, Changchun, Jilin Province, China

*Correspondence to: Maihemuti Yakufu, PhD, mhmtykf@xjmu.edu.cn; Guohua Li, MM, lighxjmu@yeah.net; Li Shu, MM, shuliyundongyixue@yeah.net; Paerhati Wahafu, MM, Parhat727@163.com

#All authors contributed equally to this work.

Funding:This study was financially supported by the Youth Medical Science and Technology Talent Special Research Project (No. WJWY-202331); Health Care and Medical Research Special Project of Xinjiang Uygur Autonomous Region (Nos. BL202460, BL202573 and BL202462); National Natural Science Foundation of China (No. 82460420); Research and Innovation Team Project of Xinjiang Medical University (No. XYD2024C12); Tianshan talent Training Program (Nos. TSYC202301B039, TSYC202301B077 and TSYC202301B106); Key Laboratory of High Incidence Disease Research in Xinjiang (Xinjiang Medical University), Ministry of Education (No. 2023B04), Natural Science Foundation of Xinjiang Uygur Autonomous Region (No. 2022D01C821); Research and Innovation Team Project, Sixth Affiliated Hospital of Xinjiang Medical University (No. LFYKYZXJJ2024001); and Scientific and Technological Development Program of Jilin Province, China (No. 20200201394JC).

Abstract

Nitric oxide, a pivotal endogenous signaling molecule, plays crucial roles in cardiovascular regulation, immune response, and neuromodulation. The rapid advancement of artificial intelligence technologies offers novel approaches to optimize real-time nitric oxide monitoring, dosing regimens, and toxicity prediction. Current interdisciplinary research on the artificial intelligence-driven nitric oxide intersection remains fragmented, with a lack of systematic investigations into knowledge architecture, technological evolution, and translational barriers. This study addressed this critical gap by presenting the knowledge graph-based analysis of artificial intelligence-driven nitric oxide system. A total of 384 relevant articles (2005–2024) were retrieved in the Web of Science Core Collection and analyzed using CiteSpace, VOSviewer, and Bibliometrix R package. Annual publications demonstrated a biphasic growth, accelerating after 2017 in tandem with breakthroughs in artificial intelligence architectures. Although China and the United States were dominated in this field, international collaborations exhibited a core-periphery structure. Research themes predominantly focused on cardiovascular and respiratory diseases, with underdeveloped applications in neuroimmunology and infectious diseases. Highly cited literature that emphasized photodynamic therapy and disease risk assessment revealed insufficient integration between artificial intelligence algorithms and fundamental nitric oxide mechanisms. Keyword evolution analysis identified a paradigm shift from traditional mechanisms (e.g., “blood pressure,” “inflammation”) to technology-driven approaches (e.g., “machine learning, ” “deep learning”). Clinical translation has faced challenges, including data heterogeneity, algorithm interpretability, and deficiencies in multicenter validation. This pioneering study systematically delineates the knowledge framework and translational bottlenecks in artificial intelligence-driven nitric oxide convergence. Future research should prioritize artificial intelligence modeling of nitric oxide dynamic metabolism, the development of explainable algorithms, and prospective clinical trials to bridge the laboratory-to-clinic gap.

摘要

一氧化氮作为一种关键的内源性信号分子，在心血管调节、免疫应答和神经调控中起着至关重要的作用。人工智能技术的快速发展为优化一氧化氮的实时监测、给药方案和毒性预测提供了新方法。当前，关于人工智能驱动的一氧化氮交叉领域的跨学科研究仍然零散，缺乏对知识架构、技术演进和转化障碍的系统性探究。本研究通过呈现基于知识图谱的人工智能驱动一氧化氮系统分析，填补了这一关键空白。研究从Web of Science核心合集检索了2005年至2024年间发表的384篇相关文献，并运用CiteSpace、VOSviewer和Bibliometrix R包进行分析。年发文量呈现双阶段增长，自2017年起随着人工智能架构的突破而加速。尽管中国和美国在该领域占据主导地位，但国际合作呈现核心-边缘结构。研究主题主要集中在心血管和呼吸系统疾病，在神经免疫学和感染性疾病中的应用尚不充分。高被引文献侧重于光动力疗法和疾病风险评估，揭示出人工智能算法与一氧化氮基础机制之间融合不足。关键词演进分析识别出从传统机制（如"血压"、"炎症"）向技术驱动方法（如"机器学习"、"深度学习"）的范式转变。临床转化面临数据异质性、算法可解释性以及多中心验证不足等挑战。这项开创性研究系统描绘了人工智能驱动的一氧化氮融合领域的知识框架与转化瓶颈。未来研究应优先关注一氧化氮动态代谢的人工智能建模、可解释性算法的开发以及前瞻性临床试验，以弥合从实验室到临床的鸿沟。