目的：基于文献计量学梳理惯性测量技术在步态分析领域的研究演进与热点结构，并从护理评估与干预转化角度分析其应用空白。方法：检索 Web of Science 核心合集2005年1月1日至2025年4月5日相关英文文献，纳入1,079篇记录；采用 CiteSpace 6.3.R1、VOSviewer 1.6.20 分析年度发文、国家/地区合作、期刊分布、关键词共现与突现，并在 Python 3.10 中以 PPMI/TF-IDF 表征、SVD降维、UMAP-HDBSCAN聚类开展关键词和摘要语义分析。结果：2006—2024年发文量由1篇增至140篇，年复合增长率为31.6%，2024年达到峰值；最高频关键词为 gait（404次）、gait analysis（268次）、walking（252次）、balance（183次）和 inertial sensors（156次）。关键词与摘要语义聚类的二维轮廓系数分别为0.579和0.642，热点集中于帕金森病/冻结步态、跌倒风险、平衡稳定性、可穿戴传感器、机器学习和康复干预。含 nursing/care 等护理相关词项的记录为142篇，但“护理”尚未形成独立主题簇。结论：惯性测量步态分析已形成医工交叉的成熟热点，但护理主导的连续评估、风险预警和干预闭环仍不足。未来应将步速、步态变异性、稳定性、对称性等参数转化为可执行的护理评估指标，推动精准护理场景中的临床验证与流程整合。

Objective: To map the research evolution and hotspot structure of inertial-measurement-based gait analysis and to examine its translational gap in nursing assessment and intervention. Methods: A total of 1,079 English records published from January 1, 2005 to April 5, 2025 were retrieved from the Web of Science Core Collection. CiteSpace 6.3.R1 and VOSviewer 1.6.20 were used for annual output, collaboration, journal distribution, keyword co-occurrence and burst analyses. Keyword and abstract semantic clusters were further examined in Python 3.10 using PPMI/TF-IDF representation, SVD, UMAP and HDBSCAN. Results: Publications increased from 1 in 2006 to 140 in 2024, with a compound annual growth rate of 31.6%. The most frequent terms were gait, gait analysis, walking, balance and inertial sensors. The two-dimensional silhouette coefficients of keyword and abstract semantic clusters were 0.579 and 0.642, respectively. Major hotspots involved Parkinson disease/freezing of gait, fall risk, balance and stability, wearable sensors, machine learning and rehabilitation. Records containing nursing/care-related terms accounted for 142 publications, but nursing did not form an independent topic cluster. Conclusion: Inertial-measurement-based gait analysis has become a mature medical-engineering research field, while nurse-led continuous assessment, risk warning and intervention feedback loops remain underdeveloped. Translating gait speed, variability, stability and symmetry into actionable nursing indicators should be prioritized in future clinical validation.

目的：基于文献计量学梳理惯性测量技术在步态分析领域的研究演进与热点结构，并从护理评估与干预转化角度分析其应用空白。方法：检索 Web of Science 核心合集2005年1月1日至2025年4月5日相关英文文献，纳入1,079篇记录；采用 CiteSpace 6.3.R1、VOSviewer 1.6.20 分析年度发文、国家/地区合作、期刊分布、关键词共现与突现，并在 Python 3.10 中以 PPMI/TF-IDF 表征、SVD降维、UMAP-HDBSCAN聚类开展关键词和摘要语义分析。结果：2006—2024年发文量由1篇增至140篇，年复合增长率为31.6%，2024年达到峰值；最高频关键词为 gait（404次）、gait analysis（268次）、walking（252次）、balance（183次）和 inertial sensors（156次）。关键词与摘要语义聚类的二维轮廓系数分别为0.579和0.642，热点集中于帕金森病/冻结步态、跌倒风险、平衡稳定性、可穿戴传感器、机器学习和康复干预。含 nursing/care 等护理相关词项的记录为142篇，但“护理”尚未形成独立主题簇。结论：惯性测量步态分析已形成医工交叉的成熟热点，但护理主导的连续评估、风险预警和干预闭环仍不足。未来应将步速、步态变异性、稳定性、对称性等参数转化为可执行的护理评估指标，推动精准护理场景中的临床验证与流程整合。

Objective: To map the research evolution and hotspot structure of inertial-measurement-based gait analysis and to examine its translational gap in nursing assessment and intervention. Methods: A total of 1,079 English records published from January 1, 2005 to April 5, 2025 were retrieved from the Web of Science Core Collection. CiteSpace 6.3.R1 and VOSviewer 1.6.20 were used for annual output, collaboration, journal distribution, keyword co-occurrence and burst analyses. Keyword and abstract semantic clusters were further examined in Python 3.10 using PPMI/TF-IDF representation, SVD, UMAP and HDBSCAN. Results: Publications increased from 1 in 2006 to 140 in 2024, with a compound annual growth rate of 31.6%. The most frequent terms were gait, gait analysis, walking, balance and inertial sensors. The two-dimensional silhouette coefficients of keyword and abstract semantic clusters were 0.579 and 0.642, respectively. Major hotspots involved Parkinson disease/freezing of gait, fall risk, balance and stability, wearable sensors, machine learning and rehabilitation. Records containing nursing/care-related terms accounted for 142 publications, but nursing did not form an independent topic cluster. Conclusion: Inertial-measurement-based gait analysis has become a mature medical-engineering research field, while nurse-led continuous assessment, risk warning and intervention feedback loops remain underdeveloped. Translating gait speed, variability, stability and symmetry into actionable nursing indicators should be prioritized in future clinical validation.

运动捕捉技术已经广泛应用于步态分析、运动康复、动作对比、技战术分析、生物运动力学分析、损伤防护、运动装备设计研发等领域,实现了人机交互的全新体验。而步行是人类运动中最基础的动作,在生物力学研究上对异常步态进行分析能够有效改善患者治疗和康复的效果。本文总结了运动捕捉技术在各种异常步态分析中的应用,并对其优缺点进行了总结与展望。检索PubMed、Embase、EBSCO、Web of Science、中国知网,收集2018年1月至2023年12月公开发表的有关有标记运动捕捉及无标记运动捕捉在异常步态的相关研究,进行系统综述。最终纳入了22篇英文文献,这些文献主要集中在神经科学、生物医学工程和临床医学等领域,特别是在步态分析、运动捕捉技术、神经病理学和康复医学等方面的应用。这些研究为我们理解和改善各种神经系统疾病,如帕金森病、多发性硬化症和脑卒中以及骨关节炎等疾病的步态提供了宝贵的见解。利用运动捕捉来进行异常步态分析能有效地给患者提供准确的康复治疗,将结果用于临床诊断、康复规划或研究目的。异常步态分析在评估肌肉骨骼状况、神经系统疾病或干预措施的有效性方面具有重要价值。

Motion capture technology has been widely used in the fields of gait analysis,sports rehabilitation,action comparison,technical and tactical analysis,biomotor mechanics analysis,injury protection,sports equipment design and development,etc.,which realizes a brand-new experience of human-computer interaction．While walking is the most basic action in human movement,the analysis of abnormal gait on biomechanical research can effectively improve the effect of patient treatment and rehabilitation．This paper summarizes the application of motion capture technology in the analysis of various abnormal gaits,and summarizes and prospects its advantages and disadvantages．PubMed,Embase,EBSCO,Web of Science and China Knowledge Network were searched to collect related studies,openly published from January 2018 to December 2023,about labeled motion capture and unlabeled motion capture in abnormal gait for systematic review．Twenty-two English-language papers were finally included,which focused on the fields of neuroscience,biomedical engineering and clinical medicine,especially in the applications of gait analysis,motion capture technology,neuropathology and rehabilitation medicine．These studies provide valuable insights into understanding and improving gait in various neurological disorders such as Parkinson’s disease,multiple sclerosis and stroke and osteoarthritis．The use of motion capture for abnormal gait analysis can be effective in providing accurate rehabilitation to patients,using the results for clinical diagnosis,rehabilitation planning,or research purposes．Abnormal gait analysis is valuable in assessing musculoskeletal conditions,neurological disorders,or the effectiveness of interventions．