Abstract
In recent years, autonomous rehabilitation training systems with no or reduced supervision have attracted much research attentions. It is essential to automatically record and recognize motions so as to realize autonomous rehabilitation training. In this paper, we propose a motion recognition model of upper-limb rehabilitation exercise (MRURE) for post-strokes. Wearable inertial measurement unit (IMU) is used to collect motion data including tri-axial acceleration and angular velocity. A Mahalanobis distance-based dynamic time warping (MDDTW) technique and an improved metric learning algorithm Nearest Neighbors Constraint (NNC) are built to measure the similarity between motions. Then, the kNN classifier is adopted to classify motions based on Mahalanobis distance. In order to verify and validate the proposed motion recognition model, we conduct extensive experiments by motion samples collected from healthy participants. The overall recognition accuracy for five motions achieves 99.34%, which demonstrates that MRURE is effective for motion recognition of upper-limb exercises. The performance of MRURE is also compared with kNN based on Euclidean and Manhattan distance and the state-of-the-art motion recognition methods. Experiment results show that the accuracy of kNN based on Mahalanobis distance is better than that of Euclidean and Manhattan distance, and MRURE outperforms methods based on feature extraction.
Similar content being viewed by others
References
An S, Gao X, Jiang M, Sun X (2018) Multivariate financial time series in the light of complex networks analysis. Phys A 503:1241–1255
Baptista RD, Bo APL, Hayashibe M (2017) Automatic human movement assessment with switching linear dynamic system: motion segmentation and motor performance. IEEE Trans Neural Syst Rehab Eng 25(6):628–640
Chen X, Gong L, Wei L, Yeh SC, Xu L, Zheng L, Zou Z (2021) A wearable hand rehabilitation system with soft gloves. IEEE Trans Indus Inf 17(2):943–952
Crema A, Maleševi´c N, Furfaro I, Raschellà F, Pedrocchi A, Micera S (2018) A wearable multi-site system for NMES-based hand function restoration. IEEE Trans Neural Syst Rehab Eng 26(2):428–440
Cui C, Bian G, Hou Z, Zhao J, Su G, Zhou H, Peng L, Wang W (2018) Simultaneous recognition and assessment of post-stroke hemiparetic gait by fusing kinematic, kinetic, and electrophysiological data. IEEE Trans Neural Sys Rehab Eng 26(4):856–864
Donnan GA, Fisher M, Macleod M, Davis SM (2008) Stroke Lancet 371(9624):1612–1623
D’Urso P, Cappelli C, Lallo D, Massari R (2013) Clustering of financial time series. Phys A 392(9):2114–2129
Galeano P, Joseph E, Lillo RE (2015) The Mahalanobis distance for functional data with applications to classification. Technometrics 57(2):281–191
Goceri E (2020) CapsNet topology to classify tumours from brain images and comparative evaluation. IET Image Process 14(5):882–889
Goceri E (2019) Diagnosis of Alzheimer’s disease with Sobolev gradient based optimization and 3D convolutional neural network. Int J Numer Method Biomed Eng 35(7):1–16
Górecki T, Łuczak M (2015) Multivariate time series classification with parametric derivative dynamic time warping. Expert Syst Appl 42(5):2305–2312
Hsu YL, Yang SC, Chang HC, Lai HC (2018) Human daily and sport activity recognition using a wearable inertial sensor networks. IEEE Access 6:31715–31728
Jones SS, Evans RS, Allen TL, Thomas A, Haug PJ, Welch SJ, Snow GL (2009) A multivariate time series approach to modeling and forecasting demand in the emergency department. J Biomed Inf 42(1):123–139
Jonnson W, Omuma O, Owolabi M, Sachdev S (2016) Stroke: a global response is needed. Bull World Health Organization 94(9):634-634A
Jung HT, Park J, Jeong J, Ryu T, Kim Y, Lee SI (2018) A wearable monitoring system for at-home stroke rehabilitation exercises: a preliminary study. In: Proc IEEE Int conf on Biomed & Heal Inform, Las Vegas, Nevada, USA, pp 4–7
Kaur PD, Sharma P (2020) IC-SMART: IoTCloud enabled seamless monitoring for Alzheimer diagnosis and rehabilitation system. J Ambi Intell Human Comp 11(8):3387–3403
Kim J-H, Hong G, Kim B-G, Dogra D-P (2018) deepGesture: Deep learning-based gesture recognition scheme using motion sensors. Displays 55:38–45
Lewandowski M, Makris D, Velastin SA, Nebel JC (2014) Structural laplacian eigenmaps for modeling sets of multivariate sequences. IEEE Trans Cyber 44(6):936–949
Liao Y, Vakanski A, Xian M, Paul D, Baker R (2020) A review of computational approaches for evaluation of rehabilitation exercises. Comput Biol Med 119:103687
Lim HS, Whang KY, Moon YS (2008) Similar sequence matching supporting variable-length and variable-tolerance continuous queries on time-series data stream. Inf Sci 178(6):1461–1478
Luay A, Tamam A, Mohammad A-Z, Mahmoud A-A, Yaser J, Hongtao L (2021) Enhancing human activity recognition using deep learning and time series augmented data. J Amb Intell Human Comp 1:1–16
Mei J, Liu M, Wang Y, Gao H (2016) Learning a Mahalanobis distance-based dynamic time warping measure for multivariate time series classification. IEEE Trans Cyb 46(6):1363–1374
Meng Y, Yu Z, Lu N, Shi D (2021) Time series classification for locating forced oscillation sources. IEEE Trans Smart Gird 12(2):1712–1721
Mensah GA, Norrving B, Feigin VL (2015) The global burden of stroke. Neuroepidemiology 45(3):143–145
Muro-de-la-Herran A, Garcia-Zapirain B, Mendez-Zorrilla A (2014) Gait analysis methods: an overview of wearable and non-wearable systems, highlighting clinical applications. Sensors 14:3362–3394
Oregi I, Del Ser J, Perez A, Lozano JA (2017) Nature-inspired approaches for distance metric learning in multivariate time series classification. In: Proc. IEEE CEC, San Sebastian, Spain 1992–1998
Patel S, Park H, Bonato P, Chan L, Rodgers M (2012) A review of wearable sensors and systems with application in rehabilitation. J Neuro Rehab 9:21
Phinyomark A, Phukpattaranont P, Limsakul C (2012) Feature reduction and selection for EMG signal classification. Expert Syst Appl 39:7420–7431
Preece SJ, Goulermas JY, Kenney LPJ, Howard D, Meijer K, Crompton R (2009) Activity identification using body-mounted sensors—a review of classification techniques. Physiol Meas 30:R1–R33
Salguero AG, Medina J, Delatorre P, Espinilla M (2019) Methodology for improving classification accuracy using ontologies: application in the recognition of activities of daily living. J Ambi Intell Human Comp 10(6):2125–2142
Seto S, Zhang W, Zhou Y (2015) Multivariate time series classification using dynamic time warping template selection for human activity recognition. In: Proc IEEE Sym series Comp Intel, Cape Town, South Africa, pp 1399–1406
Stephen JP, John YG, Laurence PJK, David H (2009) A comparison of feature extraction methods for the classification of dynamic activities from accelerometer data. IEEE Trans Biomed Eng 56(3):871–879
Switonski A, Krzeszowski T, Josinski H, Kwolek B, Wojciechowski K (2018) Gait recognition on the basis of markerless motion tracking and DTW transform. IET Biometrics 7(5):415–422
Villeneuve E, Hawin W, Holderbaum W, Janko B, Sherratt RS (2017) Reconstruction of angular kinematics from wrist-worn inertial sensor data for smart home healthcare. IEEE Access 5:2351–2363
Wang Q, Markopoulos P, Yu B, Chen W, Timmermans A (2017) Interactive wearable systems for upper body rehabilitation: a systematic review. J Neuro Rehabil 14:20
Whelan DF, O’Reilly MA, Ward TE, Delahunt E, Caulfield B (2017) Technology in rehabilitation: evaluating the single leg squat exercise with wearable inertial measurement units. Methods Info Med 56(2):88–94
Wu J, Yao L, Liu B (2018) An overview on feature-based classification algorithm for multivariate time series. Proc ICCCBDA, Chengdu, China 32–38
Ye Y, Jiang J, Ge B, Dou Y, Yang K (2019) Similarity measures for time series data classification using grid representation and matrix distance. Know Info Sys 60:1105–1134
Yu L, Xiong D, Guo L, Wang J (2016) A remote quantitative Fugl-Meyer assessment framework for stroke patients based on wearable sensor networks. Comp Meth Prog Biomed 128:100–110
Zhang Z, Liparulo L, Panella M, Gu XD, Fang Q (2016) A fuzzy kernel motion classifier for autonomous stroke rehabilitation. IEEE J Biomed Health Info 20(3):893–901
Zhao W, Reinthal MA, Espy DD, Luo X (2017) Rule-based human motion tracking for rehabilitation exercises: realtime assessment, feedback, and guidance. IEEE Access 5:21382–21394
Zhou B, Wang H, Hu F, Feng N, Xi H, Zhang Z, Tang H (2020) Accurate recognition of lower limb ambulation mode based on surface electromyography and motion data using machine learning. Comp Meth Prog Biomed 193:105486
Funding
This study was supported in part by the National Key Research and Development Program of China (grant 2019YFC1710300), Yibin Science and Technology Plan Project (grant 2022ZXD10), Open Project of Key Research Office for the Development of Health Preservation Industry of Traditional Chinese Medicine of National Administration of Traditional Chinese Medicine (grant GZ2022009), and the Sichuan Science and Technology Program (grant 2021YJ0184, 2019YFS0019 and 2020YFS0283).
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The authors declare that they have no conflict of interest.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Li, Q., Liu, Y., Zhu, J. et al. A motion recognition model for upper-limb rehabilitation exercises. J Ambient Intell Human Comput 14, 16795–16805 (2023). https://doi.org/10.1007/s12652-023-04688-5
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s12652-023-04688-5