基于人工智能和步态分析的骨科疾病辅助诊疗方法

doi:10.6040/j.issn.1671-7554.0.2024.0799

摘要/Abstract

摘要： 作为非侵入性方法,步态分析在骨科疾病诊疗中发挥着重要作用。通过观察和分析患者的行走方式,步态分析可以揭示步幅变化、步速减慢和关节角度异常等运动功能问题。这些变化都是脊髓型颈椎病、腰椎管狭窄、骨关节炎等疾病的早期症状。传统的步态分析方法通常需要专业人员人工分析步态测量设备的结果,人工智能技术与步态数据相结合的智能分析方法不仅能够实现步态分析的自动化,还显著提高了分析的客观性、一致性以及准确性。将智能分析方法应用于骨科疾病辅助诊疗有助于骨科疾病的高效准确诊断,并能够通过实时监测患者步态变化,为个性化康复治疗方案的制定提供依据。然而,多模态步态数据融合、人工智能模型可解释性以及步态测量设备的便携易用仍需开展进一步的研究工作,是未来的主要研究方向。本文主要探讨步态智能分析在骨科疾病辅助诊疗中的研究进展以及存在的问题,以期推动步态智能分析技术在更多的临床场景中应用。

关键词: 步态分析, 人工智能, 辅助诊疗, 骨科疾病, 步态数据

Abstract: As a non-invasive method, gait analysis plays an important role in the diagnosis and treatment of orthopaedic diseases. By observing and analysing a patients walking style, gait analysis can reveal motor function problems such as changes in stride length, decreased stride speed and abnormal joint angles, all of which are early symptoms of cervical spondylotic myelopathy, lumbar spinal stenosis, osteoarthritis and other diseases. Traditional gait analysis methods typically require professionals to manually interpret the results from gait measurement devices. The integration of artificial intelligence(AI)technology with gait data has led to intelligent analysis methods that not only automate gait analysis but also significantly enhance the objectivity, consistency, and accuracy of the process. The application of intelligent analysis methods in orthopedic disease diagnosis and treatment can facilitate more efficient and accurate diagnoses and provide valuable insights for personalized rehabilitation plans through real-time monitoring of gait changes. However, challenges such as multimodal gait data fusion, the interpretability of AI models, and the portability and ease of use of gait measurement devices remain areas that require further research and development. These issues represent key directions for future studies. This paper primarily explored the research progress and existing challenges of intelligent gait analysis in assisting orthopedic diagnosis and treatment, with the aim of promoting the wider clinical application of gait analysis technology.

Key words: Gait analysis, Artificial intelligence, Assisted diagnosis and treatment, Orthopaedic diseases, Gait data

中图分类号:

R681

季心宇,余思沂,孙圆圆,姬冰. 基于人工智能和步态分析的骨科疾病辅助诊疗方法[J]. 山东大学学报 (医学版), 2026, 64(2): 34-43.

JI Xinyu, YU Siyi, SUN Yuanyuan, JI Bing. Orthopedic disease diagnosis and treatment assistance methods based on artificial intelligence and gait analysis[J]. Journal of Shandong University (Health Sciences), 2026, 64(2): 34-43.

参考文献

[1] 刘芳超, 周谋望, 李涛. 基于人工智能算法的步态分析在疾病临床诊疗中的应用进展[J]. 中国康复医学杂志, 2023, 38(6): 836-840.
[2] Baker R, Esquenazi A, Benedetti MG, et al. Gait analysis: clinical facts[J]. Eur J Phys Rehabil Med, 2016, 52(4): 560-574.
[3] Feng J, Wick J, Bompiani E, et al. Applications of gait analysis in pediatric orthopaedics[J]. Curr Orthop Pract, 2016, 27(4): 455-464.
[4] Broström EW, Esbjörnsson AC, von Heideken J, et al. Gait deviations in individuals with inflammatory joint di-seases and osteoarthritis and the usage of three-dimensional gait analysis[J]. Best Pract Res Clin Rheumatol, 2012, 26(3): 409-422.
[5] Wang YF, Qi YS, Ma BX, et al. Three-dimensional gait analysis of orthopaedic common foot and ankle joint diseases[J]. Front Bioeng Biotechnol, 2024, 12: 1303035. doi:10.3389/fbioe.2024.1303035
[6] Zhang CM, Lu Y. Study on artificial intelligence: the state of the art and future prospects[J]. J Ind Inf Integr, 2021, 23: 100224. doi:10.1016/j.jii.2021.100224
[7] Mahesh B. Machine learning algorithms-a review[J]. Int J Sci Res IJSR, 2020, 9(1): 381-386.
[8] Jordan MI, Mitchell TM. Machine learning: trends, perspectives, and prospects[J]. Science, 2015, 349(6245): 255-260.
[9] Le CY, Bengio Y, Hinton G. Deep learning[J]. Nature, 2015, 521(7553): 436-444.
[10] Ji B, Dai QH, Ji XY, et al. Detection of cervical spondylotic myelopathy based on gait analysis and deterministic learning[J]. Artif Intell Rev, 2023, 56(9): 9157-9173.
[11] Watanabe T, Yoneyama T, Hayashi H, et al. Identification of the causative disease of intermittent claudication through walking motion analysis: feature analysis and differentiation[J]. Sci World J, 2014: 861529. doi:10.1155/2014/861529
[12] Kwon SB, Han HS, Lee MC, et al. Machine learning-based automatic classification of knee osteoarthritis seve-rity using gait data and radiographic images[J]. IEEE Access, 2020, 8: 120597-120603. doi:10.1109/ACCESS.2020.3006335
[13] Zhou ZR, Liang JH, Peng ZZ, et al. Gait patterns as biomarkers: a video-based approach for classifying scoliosis[EB/OL]. 2024: 2407.05726.(2024-07-08)[2024-07-20]. https://arxiv.org/abs/2407.05726v3
[14] Albuquerque P, Verlekar TT, Correia PL, et al. A spatiotemporal deep learning approach for automatic patholog-ical gait classification[J]. Sensors, 2021, 21(18): 6202. doi:10.3390/s21186202
[15] Cotton RJ, McClerklin E, Cimorelli A, et al. Transforming gait: video-based spatiotemporal gait analysis[C] //2022 44th Annual International Conference of the IEEE Engineering in Medicine & Biology Society(EMBC). New York: IEEE, 2022: 115-120. doi:10.1109/EMBC48229.2022.9871036
[16] Tao WJ, Liu T, Zheng RC, et al. Gait analysis using wearable sensors[J]. Sensors, 2012, 12(2): 2255-2283.
[17] Lee SI, Park E, Huang A, et al. Objectively quantifying walking ability in degenerative spinal disorder patients using sensor equipped smart shoes[J]. Med Eng Phys, 2016, 38(5): 442-449.
[18] Sikidar A, Vidyasagar KEC, Gupta M, et al. Classification of mild and severe adolescent idiopathic scoliosis(AIS)from healthy subjects via a supervised learning model based on electromyogram and ground reaction force data during gait[J]. Biocybern Biomed Eng, 2022, 42(3): 870-887.
[19] Visscher RMS, Sansgiri S, Freslier M, et al. Towards validation and standardization of automatic gait event identification algorithms for use in paediatric pathological populations[J]. Gait Posture, 2021, 86: 64-69. doi:10.1016/j.gaitpost.2021.02.031
[20] Ji XY, Zeng W, Dai QH, et al. Machine learning-based detection of cervical spondylotic myelopathy using multiple gait parameters[J]. Biomim Intell Robot, 2023, 3(2): 100103. doi:10.1016/j.birob.2023.100103
[21] Berner K, Cockcroft J, Morris LD, et al. Concurrent validity and within-session reliability of gait kinematics measured using an inertial motion capture system with repeated calibration[J]. J Bodyw Mov Ther, 2020, 24(4): 251-260.
[22] Laroche D, Tolambiya A, Morisset C, et al. A classification study of kinematic gait trajectories in hip osteoarthritis[J]. Comput Biol Med, 2014, 55: 42-48. doi:10.1016/j.compbiomed.2014.09.012
[23] Kwon SB, Ku Y, Han HS, et al. A machine learning-based diagnostic model associated with knee osteoarthritis severity[J]. Sci Rep, 2020, 10(1): 15743. doi:10.1038/s41598-020-72941-4
[24] Kidziński Ł, Yang B, Hicks JL, 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
[25] Paragliola G, Coronato A. Gait anomaly detection of subjects with Parkinsons disease using a deep time series-based approach[J]. IEEE Access, 2018, 6: 73280-73292. doi:10.1109/ACCESS.2018.2882245
[26] Bertaux A, Gueugnon M, Moissenet F, et al. Gait analysis dataset of healthy volunteers and patients before and 6 months after total hip arthroplasty[J]. Sci Data, 2022, 9(1): 399. doi:10.1038/s41597-022-01483-3
[27] Kour N, Gupta S, Arora S. Gait dataset for knee osteoarthritis and Parkinsons disease analysis with severity levels[EB/OL].(2020-01-01)[2024-07-20]. https://data.mendeley.com/datasets/44pfnysy89/1. doi: 10.17632/44pfnysy89.1
[28] Horsak B, Slijepcevic D, Raberger AM, et al. GaiTRec, a large-scale ground reaction force dataset of healthy and impaired gait[J]. Sci Data, 2020, 7(1): 143. doi:10.1038/s41597-020-0481-z
[29] 吕大治,霍洪峰.机器学习在步态识别中的研究综述[C] //中国体育科学学会. 第十三届全国体育科学大会论文摘要集——墙报交流(运动生物力学分会). 石家庄: 河北师范大学体育学院, 2023: 3.
[30] Khera P, Kumar N. Role of machine learning in gait analysis: a review[J]. J Med Eng Technol, 2020, 44(8): 441-467.
[31] Ji B, Dai QH, Ji XY, et al. Exploring gait analysis and deep feature contributions to the screening of cervical spondylotic myelopathy[J]. Appl Intell, 2023, 53(20): 24587-24602.
[32] Khan O, Badhiwala JH, Witiw CD, et al. Machine learning algorithms for prediction of health-related quality-of-life after surgery for mild degenerative cervical myelopathy[J]. Spine J, 2021, 21(10): 1659-1669.
[33] Toyoda H, Terai H, Yamada K, et al. A decision tree analysis to predict clinical outcome of minimally invasive lumbar decompression surgery for lumbar spinal stenosis with and without coexisting spondylolisthesis and scoliosis[J]. Spine J, 2023, 23(7): 973-981.
[34] Hayashi H, Toribatake Y, Murakami H, et al. Gait ana-lysis using a support vector machine for lumbar spinal stenosis[J]. Orthopedics, 2015, 38(11): e959-964.
[35] Zeng W, Ma LM, Yuan CZ, et al. Classification of asymptomatic and osteoarthritic knee gait patterns using gait analysis via deterministic learning[J]. Artif Intell Rev, 2019, 52(1): 449-467.
[36] Choi A, Yun TS, Suh SW, et al. Determination of input variables for the development of a gait asymmetry expert system in patients with idiopathic scoliosis[J]. Int J Precis Eng Manuf, 2013, 14(5): 811-818.
[37] Cho JS, Cho YS, Moon SB, et al. Scoliosis screening through a machine learning based gait analysis test[J]. Int J Precis Eng Manuf, 2018, 19(12): 1861-1872.
[38] Zeng W, Ismail SA, Pappas E. Detecting the presence of anterior cruciate ligament injury based on gait dynamics disparity and neural networks[J]. Artif Intell Rev, 2020, 53(5): 3153-3176.
[39] Ricciardi C, Ponsiglione AM, Scala A, et al. Machine learning and regression analysis to model the length of hospital stay in patients with femur fracture[J]. Bioengineering, 2022, 9(4): 172. doi:10.3390/bioengineering9040172
[40] Kothurkar R, Gad M, Padate A, et al. Prediction of joint moments from kinematics using machine learning in children with congenital talipes equino Varus and typically developing peers[J]. J Orthop, 2024, 57: 83-89. doi:10.1016/j.jor.2024.06.016
[41] Lai DTH, Levinger P, Begg RK, et al. Automatic recognition of gait patterns exhibiting patellofemoral pain syndrome using a support vector machine approach[J]. IEEE Trans Inf Technol Biomed, 2009, 13(5): 810-817.

多维度评价

Viewed

Full text

Abstract

Cited

Shared

Discussed