基于感知-认知解耦的可解释人工智能运动康复评估系统设计与验证

doi:10.3969/j.issn.1006-9771.2026.09.013

《中国康复理论与实践》 ›› 2026, Vol. 32 ›› Issue (9): 1108-1116.doi: 10.3969/j.issn.1006-9771.2026.09.013

• 辅助技术 • 上一篇

基于感知-认知解耦的可解释人工智能运动康复评估系统设计与验证

郭旭霞¹, 虞恒睿¹, 卢宇澄², 杨思创³, 龙雨³, 李佳蔚¹, 陈琳³(), 农飞玉⁴, 龙耀斌⁴

¹ 广西大学体育学院，广西南宁市 530004
² 广西大学土木建筑工程学院，广西南宁市 530004
³ 广西大学机械工程学院，广西南宁市 530004
⁴ 广西医科大学第二附属医院，广西南宁市 530007

收稿日期:2026-03-17 修回日期:2026-07-06 接受日期:2026-07-06 出版日期:2026-09-25 发布日期:2026-09-20
通讯作者: 陈琳, E-mail： gxdxcl@163.com
作者简介:郭旭霞(1983-)，女，汉族，山西长治市人，硕士，副教授，主要研究方向：运动康复。
基金资助:
国家重点研发计划项目(2021YFE0203500);国家自然科学基金项目(52568021);广西自然科学基金面上项目(2022GXNSFAA035511);广西创新驱动发展重大专项(AA17204017);广西中青年教师科研基础能力提升项目(2022KY0011);广西大学"大学生创新创业训练计划"国家级项目(202510593057)

An explainable artificial intelligence-driven motor rehabilitation assessment system based on perception-cognition decoupling: design and validation

Guo Xuxia¹, Yu Hengrui¹, Lu Yucheng², Yang Sichuang³, Long Yu³, Li Jiawei¹, Chen Lin³(), Nong Feiyu⁴, Long Yaobin⁴

¹ School of Physical Education, Guangxi University, Nanning, Guangxi 530004, China
² School of Civil Engineering and Architecture, Guangxi University, Nanning, Guangxi 530004, China
³ School of Mechanical Engineering, Guangxi University, Nanning, Guangxi 530004, China
⁴ The Second Affiliated Hospital of Guangxi Medical University, Nanning, Guangxi 530007, China

Received:2026-03-17 Revised:2026-07-06 Accepted:2026-07-06 Published:2026-09-25 Online:2026-09-20
Contact: Chen Lin, E-mail: gxdxcl@163.com
Supported by:
National Key R & D Program of China(2021YFE0203500);National Natural Science Foundation of China(52568021);Guangxi Natural Science Foundation (General)(2022GXNSFAA035511);Major Special Project of Guangxi Innovation-Driven Development(AA17204017);Guangxi Young and Middle-aged Teachers' Scientific Research Basic Ability Improvement Project(2022KY0011);National-level Project of Guangxi University College Student Innovation and Entrepreneurship Training Program(202510593057)

摘要/Abstract

摘要：

目的针对现有智能步态评估模型存在临床可信度不足、决策过程黑箱化等缺陷，提出一种基于感知-认知解耦架构的X-Gait运动康复评估系统，实现步态识别高精度与临床可解释性协同。

方法在感知层融合YOLOv11-Pose和PoseFormerV2完成2D-3D人体姿态重建，采用长短期记忆网络实现正常、异常、病理性步态的时序分类；引入层级相关性传播算法定位关键判别关节，自动量化躯干前倾角、支撑相占比、步幅对称性等生物力学特征。认知层融合康复医学提示工程和先验知识图谱，调用大语言模型，将量化生物力学指标转化为符合临床逻辑的结构化评估报告。纳入8例受试者开展床旁临床验证，评估系统分类性能和临床评估一致性。

结果 X-Gait系统步态整体分类准确率94.1%；2D姿态检测有效检测率100%，平均检测置信度0.819；3D姿态重建平均单帧逐关节位置误差105.97 mm。临床验证中，可解释人工智能联合大语言模型输出结果与资深康复医师人工评估高度一致。

结论基于感知-认知解耦的X-Gait系统可实现步态高精度识别与临床可解释输出。

关键词: 步态分析, 可解释人工智能, 大语言模型, X-Gait系统, 3D人体姿态估计, 运动康复

Abstract:

Objective To propose an X-Gait motor rehabilitation assessment system based on a perception-cognition decoupling architecture to jointly achieve high-accuracy gait recognition and clinical interpretability for low clinical credibility and opaque decision-making workflows in existing intelligent gait assessment models.

Methods In the perception layer, YOLOv11-Pose and PoseFormerV2 were integrated for 2D-to-3D human pose reconstruction. A long short-term memory network was adopted for temporal classification of normal, abnormal and pathological gaits. The layer-wise relevance propagation algorithm was used to identify discriminative joints and automatically quantify biomechanical indices including trunk forward inclination angle, stance phase percentage and stride symmetry. In the cognition layer, rehabilitation-oriented prompt engineering and prior knowledge graphs were combined to drive large language models, converting quantitative biomechanical measurements into clinically reasonable structured reports. Eight subjects were enrolled for bedside clinical validation to evaluate classification performance and clinical consistency.

Results The overall gait classification accuracy of X-Gait reached 94.1%. The 2D pose detection achieved 100% valid detection rate with an average confidence of 0.819. The mean per-joint position error of 3D pose reconstruction was 105.97 mm. In clinical validation of eight cases, outputs generated by explainable artificial intelligence combined with large language models showed high consistency with manual assessments from experienced rehabilitation physicians.

Conclusion The proposed X-Gait system based on perception-cognition decoupling realizes high-precision gait identification and clinically interpretable outputs.

Key words: gait analysis, explainable artificial intelligence, large language model, X-Gait system, 3D human pose estimation, sports rehabilitation

中图分类号:

R496

郭旭霞, 虞恒睿, 卢宇澄, 杨思创, 龙雨, 李佳蔚, 陈琳, 农飞玉, 龙耀斌. 基于感知-认知解耦的可解释人工智能运动康复评估系统设计与验证[J]. 《中国康复理论与实践》, 2026, 32(9): 1108-1116.

Guo Xuxia, Yu Hengrui, Lu Yucheng, Yang Sichuang, Long Yu, Li Jiawei, Chen Lin, Nong Feiyu, Long Yaobin. An explainable artificial intelligence-driven motor rehabilitation assessment system based on perception-cognition decoupling: design and validation[J]. Chinese Journal of Rehabilitation Theory and Practice, 2026, 32(9): 1108-1116.

图/表 10

图1

图2

表1

表2

图3

表3

图4

表4

表5

表6

参考文献 36

[1]	Beard J R, Officer A M, Cassels A K. The world report on ageing and health[J]. Gerontologist, 2016, 56(S2): S163-S166. doi: 10.1093/geront/gnw037
[2]	Fan C, Yang B, Li X, et al. Temporal-frequency-phase feature classification using 3D-convolutional neural networks for motor imagery and movement[J]. Front Neurosci, 2023, 17: 1250991. doi: 10.3389/fnins.2023.1250991
[3]	Ngiam K Y, Khor I W. Big data and machine learning algorithms for health-care delivery[J]. Lancet Oncol, 2019, 20(5): e262-e273. doi: 10.1016/S1470-2045(19)30149-4
[4]	Kidziński Ł, Yang B, Hicks J L, et al. Deep neural networks enable quantitative movement analysis using single-camera videos[J]. Nat Commun, 2020, 11(1): 4054. doi: 10.1038/s41467-020-17807-z pmid: 32792511
[5]	丁德波, 史耀群. 基于时空图卷积网络与多层次特征融合的快递员3D人体姿态估计[J]. 传感技术学报, 2025, 38(8): 1457-1462.
	Ding D B, Shi Y Q. 3D human pose estimation of couriers based on spatio-temporal graph convolutional networks with multi-level feature fusion[J]. Chin J Sensor Actuator, 2025, 38(8): 1457-1462.
[6]	Cao Z, Hidalgo G, Simon T, et al. OpenPose: realtime multi-person 2D pose estimation using part affinity fields[J]. IEEE Trans Pattern Anal Mach Intell, 2021, 43(1): 172-186. doi: 10.1109/TPAMI.34
[7]	Stenum J, Rossi C, Roemmich R T. Two-dimensional video-based analysis of human gait using pose estimation[J]. PLoS Comput Biol, 2021, 17(4): e1008935. doi: 10.1371/journal.pcbi.1008935
[8]	Uhlrich S D, Falisse A, Kidziński Ł, et al. OpenCap: human movement dynamics from smartphone videos[J]. PLoS Comput Biol, 2023, 19(10): e1011462. doi: 10.1371/journal.pcbi.1011462
[9]	张堃, 张鹏程, 陈孝豪, 等. 基于三维姿态估计的智能康复运动检测系统应用研究[J]. 仪器仪表学报, 2025, 46(6): 181-193.
	Zhang K, Zhang P C, Chen X H, et al. Application research of intelligent rehabilitation exercise detection system based on 3D human pose estimation[J]. Chin J Sci Instrument, 2025, 46(6): 181-193.
[10]	Needham L, Evans M, Wade L, et al. The development and evaluation of a fully automated markerless motion capture workflow[J]. J Biomech, 2022, 144: 111338. doi: 10.1016/j.jbiomech.2022.111338
[11]	程相鑫, 张烁, 杜松骏, 等. 基于IMU信号的人工智能上肢多关节运动状态识别系统构建:卒中后人工智能运动功能评估与检测系统建设前导研究[J]. 中国卒中杂志, 2025, 20(4): 401-409.
	Cheng X X, Zhang S, Du S J, et al. Construction of an artificial intelligence upper limb multi-joint motion state recognition system based on IMU signals: a preliminary study for the development of an artificial intelligence motor function assessment and detection system after stroke[J]. Chin J Stroke, 2025, 20(4): 401-409.
[12]	Singhal K, Azizi S, Tu T, et al. Large language models encode clinical knowledge[J]. Nature, 2023, 620(7972): 172-180. doi: 10.1038/s41586-023-06291-2
[13]	Bedi S, Liu Y, Orr-Ewing L, et al. Testing and evaluation of health care applications of large language models: a systematic review[J]. JAMA, 2025, 333(4): 319-328. doi: 10.1001/jama.2024.21700
[14]	Yin S, Fu C, Zhao S, et al. A survey on multimodal large language models[J]. Natl Sci Rev, 2024, 11(12): nwae403.
[15]	Wang C, Li S, Lin N, et al. Application of large language models in medical training evaluation: using ChatGPT as a standardized patient: multimetric assessment[J]. J Med Internet Res, 2025, 27: e59435.
[16]	Esteva A, Robicquet A, Ramsundar B, et al. A guide to deep learning in healthcare[J]. Nat Med, 2019, 25(1): 24-29. doi: 10.1038/s41591-018-0316-z pmid: 30617335
[17]	Goetz L, Seedat N, Vandersluis R, et al. Generalization: a key challenge for responsible AI in patient-facing clinical applications[J]. NPJ Digit Med, 2024, 7(1): 126. doi: 10.1038/s41746-024-01127-3 pmid: 38773304
[18]	周鑫源, 邱敏, 商江峰, 等. 基于深度学习的甲状腺结节超声图像计算机辅助诊断研究进展[J]. 生物医学工程学杂志, 2025, 42(5): 1069-1075.
	Zhou X Y, Qiu M, Shang J F, et al. Research progress on deep learning-based computer-aided diagnosis of thyroid nodules using ultrasound imaging[J]. J Biomed Eng, 2025, 42(5): 1069-1075.
[19]	宁文卓. 基于深度学习的人体姿态识别方法研究[D]. 大连: 大连交通大学, 2025.
	Ning W Z. Research on human pose recognition method based on deep learning[D]. Dalian: Dalian Jiaotong University, 2025.
[20]	赵兴鹏, 徐航, 常朋发, 等. 基于多域协同训练的半监督雷达人体动作识别方法[J]. [网络首发]. 雷达学报(中英文), 2026. https://link.cnki.net/urlid/10.1030.TN.20260305.2140.002.
	Zhao X P, Xu H, Chang P F, et al. A semisupervised radar-based human action recognition method via multidomain collaborative training[J]. [ahead of print]. J Radars, 2026. https://link.cnki.net/urlid/10.1030.TN.20260305.2140.002.
[21]	石欣, 黄良文, 梁飞, 等. 基于肌电信号的跨被试不变特征运动模式识别[J]. 仪器仪表学报, 2026, 47(2): 244-255.
	Shi X, Huang L W, Liang F, et al. Cross-subject invariant feature-based motion pattern recognition using sEMG[J]. Chin J Sci Instrument, 2026, 47(2): 244-255.
[22]	宋永坤, 游文杰, 岳磊, 等. 结合时空特征的超宽带MIMO雷达图像人体轮廓恢复与动作识别方法[J]. [网络首发]. 雷达学报(中英文), 2026. https://link.cnki.net/urlid/10.1030.tn.20260123.1013.010.
	Song Y K, You W J, Yue L, et al. Human contour restoration and action recognition in ultra-wideband radar imaging based on spatio-temporal features[J]. [ahead of print]. J Radars, 2026. https://link.cnki.net/urlid/10.1030.tn.20260123.1013.010.
[23]	郑屹, 黄向, 秦菲儿, 等. 2D/3D生成式人工智能技术发展及创意产业应用[J]. 中国图象图形学报, 2025, 30(6): 1953-1984.
	Zheng Y, Huang X, Qin F E, et al. Development of 2D/3D generative artificial intelligence technology and its applications in creative industries[J]. J Image Graph, 2025, 30(6): 1953-1984.
[24]	郭卓涛, 张凯, 查国春, 等. 腰椎融合对全髋关节置换后中期疗效影响的匹配对照试验[J]. 中国组织工程研究, 2023, 27(36): 5801-5805.
	Guo Z T, Zhang K, Zha G C, et al. A matched controlled trial of the effect of lumbar fusion on mid-term outcomes after total hip arthroplasty[J]. Chin J Tissue Eng Res, 2023, 27(36): 5801-5805.
[25]	徐大涛, 全文静, 周辉宇, 等. 基于深度神经网络和逐层相关性传播技术探究"高-低"里程跑者步态模式差异[J]. 医用生物力学, 2022, 37(6): 1151-1157.
	Xu D T, Quan W J, Zhou H Y, et al. Exploration of gait pattern differences between high-mileage and low-mileage runners based on deep neural network and layer-wise relevance propagation[J]. J Med Biomech, 2022, 37(6): 1151-1157.
[26]	马腾, 王毓萱, 侯志民, 等. 动力大腿假肢控制技术研究进展及发展趋势[J]. 机器人, 2026, 48(2): 395-413.
	Ma T, Wang Y X, Hou Z M, et al. Research progress and development trend of powered transfemoral prosthesis control technology[J]. Robot, 2026, 48(2): 395-413.
[27]	郑蕾, 李涵, 何健, 等. 基于机器学习的高温防护涂层智能化设计与性能预测[J]. [网络首发]. 北京航空航天大学学报, 2026. https://doi.org/10.13700/j.bh.1001-5965.2025.0911.
	Zheng L, Li H, He J, et al. Intelligent design and performance prediction of high-temperature protective coatings based on machine learning[J]. [ahead of print]. J Beijing Univ Aeronaut Astronaut, 2026. https://doi.org/10.13700/j.bh.1001-5965.2025.0911.
[28]	王福东, 黄诗雅, 赵宏宇. 急性缺血性脑卒中患者首次运动康复治疗时间对运动功能的影响研究[J]. 中国全科医学, 2018, 21(25): 3089-3094. doi: 10.12114/j.issn.1007-9572.2018.25.012
	Wang F D, Huang S Y, Zhao H Y. Effect of timing of initiating exercise rehabilitation on the prognosis of motor function in acute ischemic stroke[J]. Chin Gener Pract, 2018, 21(25): 3089-3094.
[29]	万盛华, 徐兴业, 甘乐, 等. 基于多模态大模型辅助视频动作生成的预训练世界模型[J]. 计算机科学, 2026, 53(1): 51-57. doi: 10.11896/jsjkx.250800033
	Wan S H, Xu X Y, Gan L, et al. Pre-training world models from videos with generated actions by multi-modal large models[J]. Comput Sci, 2026, 53(1): 51-57.
[30]	刘天华, 张维彦, 刘井平, 等. 面向大语言模型的序列知识评测与优化[J]. [网络首发]. 计算机科学, 2026. https://link.cnki.net/urlid/50.1075.TP.20260319.1707.011.
	Liu T H, Zhang W Y, Liu J P, et al. Sequential knowledge evaluation and optimization for large language models[J]. [ahead of print]. Comput Sci, 2026. https://link.cnki.net/urlid/50.1075.TP.20260319.1707.011.
[31]	蔡锦添, 郝耀辉, 陈鑫, 等. 基于大模型的全球航展知识图谱构建与问答系统框架研究[J]. 计算机工程与应用, 2026, 62(9): 108-121. doi: 10.3778/j.issn.1002-8331.2507-0068
	Cai J T, Hao Y H, Chen X, et al. Research on framework of global airshow knowledge graph construction and question-answering system based on large language models[J]. Comput Eng Appl, 2026, 62(9): 108-121. doi: 10.3778/j.issn.1002-8331.2507-0068
[32]	蔡华年, 费思先, 张忆晨, 等. 基于导纳控制的双边康复机器人运动辅助分析[J]. 中国康复理论与实践, 2023, 29(9): 1104-1109. doi: 10.3969/j.issn.1006-9771.2023.09.015
	Cai H N, Fei S X, Zhang Y C, et al. Motion assistance analysis for robot-assisted tele-rehabilitation based on bilateral admittance control[J]. Chin J Rehabil Theory Prac, 2023, 29(9): 1104-1109.
[33]	马超, 邹子炀, 范杰. 基于RoBERTa模型的民机运维数据语料要素识别[J]. 科学技术与工程, 2026, 26(2): 701-707.
	Ma C, Zou Z Y, Fan J. Identification of corpus elements for civil aircraft operation and maintenance data based on roberta model[J]. Sci Technol Eng, 2026, 26(2): 701-707.
[34]	唐文胜, 熊露, 陈明, 等. LLM与多智能体协同的计算机类本科毕业论文智能评级框架[J]. [网络首发]. 计算机应用, 2026. https://link.cnki.net/urlid/51.1307.TP.20251224.1122.002.
	Tang W S, Xiong L, Chen M, et al. Intelligent rating framework for computer undergraduate graduation thesis via LLM and multi-agent collaboration[J]. [ahead of print]. J Comput Appl, 2026. https://link.cnki.net/urlid/51.1307.TP.20251224.1122.002.
[35]	杨浩, 徐旻, 彭荷玲, 等. 大语言模型在智能病历内涵质控中的应用研究[J]. 西安交通大学学报(医学版), 2026, 47(3): 477-485.
	Yang H, Xu M, Peng H L, et al. Application of large language models in intelligent intrinsic quality control of electronic medical records[J]. J Xi'an Jiaotong Univ (Med Sci), 2026, 47(3): 477-485.
[36]	范文仲, 徐李璐邑. 云边协同智能算力网络建设与城市发展创新[J]. 城市问题, 2025(7): 15-24.
	Fan W Z, Xu L L Y. Cloud-edge collaborative artificial intelligent computing power network development and urban innovation[J]. Urban Problem, 2025(7): 15-24.

编号	性别	年龄/岁	原发病	身高/cm	体质量/kg
sub01	男	21	健康	178	75
sub02	男	36	前交叉韧带断裂	172	70
sub03	男	52	右侧股骨骨折	178	70
sub04	女	78	脑出血	160	52
sub05	男	47	脑梗死	173	78
sub06	男	40	脑出血	180	85
sub07	男	52	脑出血	170	76
sub08	男	57	脑梗死	171	74

指标	数值	定性评估
有效检测率	100%	每帧都能成功检测到人体，无漏检，系统鲁棒性强
平均置信度	0.819	> 0.7，可接受，对检测结果有信心
可见关键点数	13.8/17	平均每帧有81%的关节点可见，遮挡较少，输入质量高

指标	总体	正常	异常	病理性
准确率	0.941	-	-	-
精确率	0.948	0.994	0.865	1.000
召回率	0.941	0.886	1.000	0.981
F1分数	0.941	0.937	0.928	0.991
支持数	356	175	128	53

阶段	时间戳	输出
LSTM分类	03:41:02,054	异常，置信度0.974
XAI贡献度分析	03:41:02,054	关键关节：躯干
生物力学特征提取	-	躯干前倾角均值75.4°，标准差3.7°；支撑相占比0.53
语义提示生成	03:41:02,054	模型将该步态分类为异常，置信度97.4%。关键关节为躯干。躯干前倾角度增加70.4°，且支撑相时间缩短至0.53
LLM医学解释生成	03:41:17,762	成功返回结构化自然语言解释

类型	特征描述
单一型	对称步态+单一部位代偿(足弓/躯干)
双维型	对称或轻度不对称+单一平面偏移+支撑延长
复合型	显著不对称+多平面躯干代偿+支撑相严重失衡

基于感知-认知解耦的可解释人工智能运动康复评估系统设计与验证

An explainable artificial intelligence-driven motor rehabilitation assessment system based on perception-cognition decoupling: design and validation

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 10

参考文献 36

相关文章 15

Metrics

本文评价

推荐阅读 0

编号	主要表现	类型	关键量化指标	医生评估要点	是否一致
sub01	显著不对称，骨盆代偿下降	复合型	躯干前倾24.2°；支撑相占比0.963(左长右短)	右臀中肌无力，Trendelenburg征(+)，支撑时间不等	是
sub02	轻度对称异常，足弓塌陷	单一型	躯干前倾15.9°；支撑相占比0.961；膝内扣0.1°	核心稳定性下降，静态足弓指数低，双侧支撑一致	是
sub03	步长不对称伴支撑期延长	双维型	步长不对称；躯干侧弯6.1°；单侧支撑相明显延长	步长不对称，躯干侧弯5°，单侧支撑相明显延长	是
sub04	步长不对称伴躯干代偿倾斜	双维型	躯干前倾13.5°；躯干侧弯5.4°；单侧支撑相明显延长	步长不对称，躯干前倾10°伴轻度侧弯，支撑相延长	是
sub05	步态对称但躯干侧偏	单一型	躯干前倾9.5°；单侧支撑相明显延长	躯干前倾10°，单侧支撑相代偿性延长	是
sub06	对称步态伴明显侧弯	双维型	步长对称，躯干侧弯8.7°；单侧相支撑明显延长	躯干侧弯10°，单侧支撑相明显延长	是
sub07	不对称步态伴轻度侧弯	双维型	步长不对称；躯干前倾5.1°，侧弯6.3°；单侧支撑相明显延长	步长不对称，躯干侧弯5°及单侧支撑相延长	是
sub08	复合不对称与前倾代偿	复合型	步长不对称；躯干前倾13.3°侧弯6.5°；单侧支撑明显延长	步长不对称，躯干前倾10°，侧弯5°及单侧支撑延长	是

[1]	刘华, 马欣, 李静, 王丽媛, 王丽. 基于康复胜任力的运动康复专业硕士研究生培养方案比较[J]. 《中国康复理论与实践》, 2026, 32(8): 984-992.
[2]	刘华, 李静, 王丽媛, 马欣, 王丽. 体医融合背景下基于对分课堂模式的运动康复治疗技术课社协同教学改革效果[J]. 《中国康复理论与实践》, 2026, 32(4): 482-487.
[3]	刘华, 王丽媛, 马欣, 李静, 王丽. 基于OBE-CDIO模式运动康复训练教程课程改革的效果[J]. 《中国康复理论与实践》, 2026, 32(3): 270-275.
[4]	中国康复医学会康复医学教育专业委员会运动康复学组, 钱菁华, 廖远朋, 王琳, 张秋霞, 保文莉, 张海峰. 基于世界卫生组织康复胜任力架构构建运动康复本科专业实习体系[J]. 《中国康复理论与实践》, 2026, 32(2): 241-248.
[5]	陈健, 李哲恒, 王定宣. 基于世界卫生组织康复胜任力架构的中国运动康复专业人才培养现状分析[J]. 《中国康复理论与实践》, 2026, 32(1): 110-116.
[6]	刘芳超, 张元鸣飞, 武美奇, 周谋望, 李涛. 人工智能人体关键点检测技术在步态运动学分析中的效度[J]. 《中国康复理论与实践》, 2025, 31(3): 249-253.
[7]	王微平, 马迎春. 慢性肾脏病患者居家运动康复促进因素与障碍因素的质性研究[J]. 《中国康复理论与实践》, 2025, 31(2): 235-241.
[8]	戴栖, 武庆昌, 管彦舒, 沈海勤, 鲍捷. 基于ICF构建注意缺陷多动障碍儿童运动康复方案[J]. 《中国康复理论与实践》, 2024, 30(8): 903-913.
[9]	荣先凤, 李方晖, 谭嘉俐. 基于世界卫生组织康复胜任力架构建设运动康复专业[J]. 《中国康复理论与实践》, 2024, 30(6): 639-647.
[10]	吴亮, 许秀, 罗亮. 运动康复和适应性身体活动对痉挛性脑性瘫痪儿童青少年心理运动功能、运动功能和动作发展的效益：基于ICF的循证研究[J]. 《中国康复理论与实践》, 2024, 30(2): 148-156.
[11]	杨平, 陈浩源, 高睿馨, 王欣平. 基于低成本惯性测量单元和力敏电阻的足部三维运动数据建模[J]. 《中国康复理论与实践》, 2024, 30(10): 1224-1231.
[12]	陈珺雯, 陈谦, 陈程, 李淑月, 刘玲玲, 吴存书, 龚翔, 鲁俊, 许光旭. 改良八段锦身体活动对脑卒中患者心肺功能、运动功能和日常生活活动能力的效果[J]. 《中国康复理论与实践》, 2024, 30(1): 74-80.
[13]	张园, 杨剑. 基于ICD-11和ICF脊髓损伤患者运动康复干预方案及其健康效益：系统综述的系统综述[J]. 《中国康复理论与实践》, 2023, 29(12): 1377-1385.
[14]	乔虎军, 郝鑫, 刘晓燕, 王国祥. 肌少症老年人运动康复的健康效益：系统综述的系统综述[J]. 《中国康复理论与实践》, 2023, 29(11): 1296-1303.
[15]	王馨婷, 杨剑. 孤独症谱系障碍儿童结构化身体活动方案及其健康效益的系统综述[J]. 《中国康复理论与实践》, 2023, 29(10): 1117-1124.