{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,2,21]],"date-time":"2026-02-21T18:35:27Z","timestamp":1771698927960,"version":"3.50.1"},"reference-count":65,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2023,7,14]],"date-time":"2023-07-14T00:00:00Z","timestamp":1689292800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Neurorobot."],"abstract":"<jats:sec><jats:title>Introduction<\/jats:title><jats:p>Stroke survivors often compensate for the loss of motor function in their distal joints by altered use of more proximal joints and body segments. Since this can be detrimental to the rehabilitation process in the long-term, it is imperative that such movements are indicated to the patients and their caregiver. This is a difficult task since compensation strategies are varied and multi-faceted. Recent works that have focused on supervised machine learning methods for compensation detection often require a large training dataset of motions with compensation location annotations for each time-step of the recorded motion. In contrast, this study proposed a novel approach that learned a linear classifier from energy-based features to discriminate between healthy and compensatory movements and identify the compensating joints without the need for dense and explicit annotations.<\/jats:p><\/jats:sec><jats:sec><jats:title>Methods<\/jats:title><jats:p>Six healthy physiotherapists performed five different tasks using healthy movements and acted compensations. The resulting motion capture data was transformed into joint kinematic and dynamic trajectories. Inspired by works in bio-mechanics, energy-based features were extracted from this dataset. Support vector machine (SVM) and logistic regression (LR) algorithms were then applied for detection of compensatory movements. For compensating joint identification, an additional condition enforcing the independence of the feature calculation for each observable degree of freedom was imposed.<\/jats:p><\/jats:sec><jats:sec><jats:title>Results<\/jats:title><jats:p>Using leave-one-out cross validation, low values of mean brier score (&amp;lt;0.15), mis-classification rate (&amp;lt;0.2) and false discovery rate (&amp;lt;0.2) were obtained for both SVM and LR classifiers. These methods were found to outperform deep learning classifiers that did not use energy-based features. Additionally, online classification performance by our methods were also shown to outperform deep learning baselines. Furthermore, qualitative results obtained from the compensation joint identification experiment indicated that the method could successfully identify compensating joints.<\/jats:p><\/jats:sec><jats:sec><jats:title>Discussion<\/jats:title><jats:p>Results from this study indicated that including prior bio-mechanical information in the form of energy based features can improve classification performance even when linear classifiers are used, both for offline and online classification. Furthermore, evaluation compensation joint identification algorithm indicated that it could potentially provide a straightforward and interpretable way of identifying compensating joints, as well as the degree of compensation being performed.<\/jats:p><\/jats:sec>","DOI":"10.3389\/fnbot.2023.1155826","type":"journal-article","created":{"date-parts":[[2023,7,15]],"date-time":"2023-07-15T08:47:34Z","timestamp":1689410854000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":4,"title":["Online detection of compensatory strategies in human movement with supervised classification: a pilot study"],"prefix":"10.3389","volume":"17","author":[{"given":"Neha","family":"Das","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Satoshi","family":"Endo","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Sabrina","family":"Patel","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Carmen","family":"Krewer","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Sandra","family":"Hirche","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1965","published-online":{"date-parts":[[2023,7,14]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"102594","DOI":"10.1016\/j.msksp.2022.102594","article-title":"Physiotherapists could detect changes of 12 degrees or more in single-plane movement when observing forward bending, squat or hand-over-head: A cross-sectional experiment","volume":"61","author":"Abbott","year":"2022","journal-title":"Musculoskeletal Sci. Pract"},{"key":"B2","doi-asserted-by":"publisher","first-page":"216","DOI":"10.1016\/j.patrec.2021.02.013","article-title":"Action recognition using kinematics posture feature on 3d skeleton joint locations","volume":"145","author":"Ahad","year":"2021","journal-title":"Patt. Recogn. Lett"},{"key":"B3","doi-asserted-by":"publisher","first-page":"3","DOI":"10.1097\/NPT.0000000000000295","article-title":"Upper limb robotic rehabilitation after stroke: a multicenter, randomized clinical trial","volume":"44","author":"Aprile","year":"2020","journal-title":"J. Neurol. Phys. Ther"},{"key":"B4","doi-asserted-by":"publisher","first-page":"146","DOI":"10.18517\/ijaseit.7.1.1790","article-title":"Accelerator-based human activity recognition using voting technique with nbtree and mlp classifiers","volume":"7","author":"Azmi","year":"2017","journal-title":"Int. J. Adv. Sci. Eng. Inf. Technol"},{"key":"B5","first-page":"185","article-title":"\u201cSegmenting motion capture data into distinct behaviors,\u201d","volume-title":"Proceedings of Graphics Interface","author":"Barbi\u010d","year":"2004"},{"key":"B6","doi-asserted-by":"publisher","first-page":"58","DOI":"10.1310\/BQM5-6YGB-MVJ5-WVCR","article-title":"The role of task-specific training in rehabilitation therapies","volume":"12","author":"Bayona","year":"2005","journal-title":"Topics Stroke Rehabil"},{"key":"B7","first-page":"117","article-title":"\u201cMotion-motif graphs,\u201d","author":"Beaudoin","year":"2008","journal-title":"Proceedings of the 2008 ACM SIGGRAPH\/Eurographics Symposium on Computer Animation"},{"key":"B8","volume-title":"Pattern Recognition and Machine Learning, vol 4","author":"Bishop","year":"2006"},{"key":"B9","doi-asserted-by":"publisher","first-page":"59","DOI":"10.1016\/0022-510X(95)00003-K","article-title":"Repetitive training of isolated movements improves the outcome of motor rehabilitation of the centrally paretic hand","volume":"130","author":"B\u00fctefisch","year":"1995","journal-title":"J. Neurol. Sci"},{"key":"B10","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1186\/s12984-019-0609-6","article-title":"Detecting compensatory movements of stroke survivors using pressure distribution data and machine learning algorithms","volume":"16","author":"Cai","year":"2019","journal-title":"J. Neuroeng. Rehabilit"},{"key":"B11","doi-asserted-by":"publisher","first-page":"2585","DOI":"10.1016\/S0031-3203(03)00136-5","article-title":"Efficient leave-one-out cross-validation of kernel fisher discriminant classifiers","volume":"36","author":"Cawley","year":"2003","journal-title":"Patt. Recogn"},{"key":"B12","doi-asserted-by":"publisher","first-page":"633","DOI":"10.1016\/j.neucom.2013.04.003","article-title":"Joint segmentation of multivariate time series with hidden process regression for human activity recognition","volume":"120","author":"Chamroukhi","year":"2013","journal-title":"Neurocomputing"},{"key":"B13","doi-asserted-by":"publisher","first-page":"398","DOI":"10.1177\/1545968306298414","article-title":"Improvement of arm movement patterns and endpoint control depends on type of feedback during practice in stroke survivors","volume":"21","author":"Cirstea","year":"2007","journal-title":"Neurorehabilit. Neur. Rep"},{"key":"B14","doi-asserted-by":"publisher","first-page":"940","DOI":"10.1093\/brain\/123.5.940","article-title":"Compensatory strategies for reaching in stroke","volume":"123","author":"Cirstea","year":"2000","journal-title":"Brain"},{"key":"B15","doi-asserted-by":"publisher","first-page":"833","DOI":"10.1007\/s11517-016-1555-0","article-title":"Novel kinematic indices for quantifying upper limb ability and dexterity after cervical spinal cord injury","volume":"55","author":"de Los Reyes-Guzm\u00e1n","year":"2017","journal-title":"Med. Biol. Eng. Comput"},{"key":"B16","doi-asserted-by":"publisher","first-page":"1940","DOI":"10.1109\/TBME.2007.901024","article-title":"Opensim: open-source software to create and analyze dynamic simulations of movement","volume":"54","author":"Delp","year":"2007","journal-title":"IEEE Trans. Biomed. Eng"},{"key":"B17","doi-asserted-by":"publisher","first-page":"771","DOI":"10.2466\/pms.1997.85.3.771","article-title":"Repetitive practice of a single joint movement for enhancing elbow function in hemiparetic patients","volume":"85","author":"Dickstein","year":"1997","journal-title":"Percep. Motor Skills"},{"key":"B18","doi-asserted-by":"publisher","first-page":"375","DOI":"10.1145\/3154862.3154925","article-title":"\u201cThe toronto rehab stroke pose dataset to detect compensation during stroke rehabilitation therapy,\u201d","author":"Dolatabadi","year":"2017","journal-title":"Proceedings of the 11th EAI International Conference on Pervasive Computing Technologies for Healthcare"},{"key":"B19","doi-asserted-by":"publisher","first-page":"39","DOI":"10.1023\/A:1013254724861","article-title":"Automated derivation of primitives for movement classification","volume":"12","author":"Fod","year":"2002","journal-title":"Auton. Rob"},{"key":"B20","first-page":"389","article-title":"Is health-related-quality of life of stroke patients influenced by neurological impairments at one year after stroke?","volume":"46","author":"Franceschini","year":"2010","journal-title":"Eur. J. Phys. Rehabilit. Med"},{"key":"B21","doi-asserted-by":"publisher","first-page":"918804","DOI":"10.3389\/fnhum.2022.918804","article-title":"Automating provision of feedback to stroke patients with and without information on compensatory movements: A pilot study","volume":"16","author":"Fruchter","year":"2022","journal-title":"Front. Human Neurosci"},{"key":"B22","doi-asserted-by":"publisher","first-page":"7001350010p1","DOI":"10.5014\/ajot.2016.015487","article-title":"Range of motion requirements for upper-limb activities of daily living","volume":"70","author":"Gates","year":"2016","journal-title":"Am. J. Occup. Ther"},{"key":"B23","doi-asserted-by":"publisher","first-page":"93","DOI":"10.1134\/S0081543810010098","article-title":"A biomechanical inactivation principle","volume":"268","author":"Gauthier","year":"2010","journal-title":"Proc. Steklov Inst. Mathem"},{"key":"B24","first-page":"1321","article-title":"\u201cOn calibration of modern neural networks,\u201d","volume-title":"International Conference on Machine Learning","author":"Guo","year":"2017"},{"key":"B25","doi-asserted-by":"publisher","first-page":"1028","DOI":"10.1016\/j.tics.2020.09.004","article-title":"Embracing change: Continual learning in deep neural networks","volume":"24","author":"Hadsell","year":"2020","journal-title":"Trends Cogn. Sci"},{"key":"B26","doi-asserted-by":"publisher","first-page":"263","DOI":"10.1016\/j.math.2006.05.005","article-title":"Accuracy and reliability of observational motion analysis in identifying shoulder symptoms","volume":"12","author":"Hickey","year":"2007","journal-title":"Manual Ther"},{"key":"B27","doi-asserted-by":"publisher","first-page":"829","DOI":"10.1007\/s10439-005-3320-7","article-title":"A model of the upper extremity for simulating musculoskeletal surgery and analyzing neuromuscular control","volume":"33","author":"Holzbaur","year":"2005","journal-title":"Ann. Biomed. Eng"},{"key":"B28","doi-asserted-by":"publisher","first-page":"2182","DOI":"10.1523\/JNEUROSCI.4003-11.2012","article-title":"Reduction of metabolic cost during motor learning of arm reaching dynamics","volume":"32","author":"Huang","year":"2012","journal-title":"J. Neurosci"},{"key":"B29","doi-asserted-by":"publisher","first-page":"1234","DOI":"10.1109\/TETC.2020.2988945","article-title":"A machine-learning model for automatic detection of movement compensations in stroke patients","volume":"9","author":"Kashi","year":"2020","journal-title":"IEEE Trans. Emerg. Topics Comput"},{"key":"B30","doi-asserted-by":"crossref","first-page":"465","DOI":"10.1109\/ICRoM54204.2021.9663458","article-title":"\u201cCompensatory movement detection in upper limb rehabilitation with deep learning methods,\u201d","volume-title":"2021 9th RSI International Conference on Robotics and Mechatronics (ICRoM)","author":"Khoramdel","year":"2021"},{"key":"B31","doi-asserted-by":"publisher","first-page":"330","DOI":"10.1177\/0278364911426178","article-title":"Incremental learning of full body motion primitives and their sequencing through human motion observation","volume":"31","author":"Kuli\u0107","year":"2012","journal-title":"Int. J. Robot. Res"},{"key":"B32","doi-asserted-by":"publisher","first-page":"2181","DOI":"10.1161\/01.STR.0000087172.16305.CD","article-title":"Probability of regaining dexterity in the flaccid upper limb: impact of severity of paresis and time since onset in acute stroke","volume":"34","author":"Kwakkel","year":"2003","journal-title":"Stroke"},{"key":"B33","doi-asserted-by":"publisher","first-page":"173","DOI":"10.1016\/j.rehab.2019.10.009","article-title":"Rehabilitation of the upper arm early after stroke: video games versus conventional rehabilitation: a randomized controlled trial","volume":"63","author":"Laffont","year":"2020","journal-title":"Ann. Phys. Rehabilit. Med"},{"key":"B34","doi-asserted-by":"publisher","first-page":"CD008349","DOI":"10.1002\/14651858.CD008349.pub4","article-title":"Virtual reality for stroke rehabilitation","volume":"11","author":"Laver","year":"2017","journal-title":"Cochrane Database Syst. Rev"},{"key":"B35","doi-asserted-by":"publisher","first-page":"635","DOI":"10.1177\/1545968315613863","article-title":"Compensatory versus noncompensatory shoulder movements used for reaching in stroke","volume":"30","author":"Levin","year":"2016","journal-title":"Neurorehabil. Neur. Rep"},{"key":"B36","doi-asserted-by":"publisher","first-page":"768","DOI":"10.1109\/IAECST54258.2021.9695873","article-title":"\u201cA vision-based compensation detection approach during robotic stroke rehabilitation therapy,\u201d","author":"Lin","year":"2021","journal-title":"2021 3rd International Academic Exchange Conference on Science and Technology Innovation (IAECST)"},{"key":"B37","doi-asserted-by":"publisher","first-page":"1250","DOI":"10.3389\/fnins.2019.01250","article-title":"semg-based trunk compensation detection in rehabilitation training","volume":"13","author":"Ma","year":"2019","journal-title":"Front. Neurosci"},{"key":"B38","doi-asserted-by":"publisher","first-page":"2421","DOI":"10.1097\/00007632-199511001-00009","article-title":"Motion characteristics of the lumbar spine in the normal population","volume":"20","author":"McGregor","year":"1995","journal-title":"Spine"},{"key":"B39","doi-asserted-by":"publisher","first-page":"291","DOI":"10.3109\/09638288.2012.691942","article-title":"Predicting health related quality of life 6 months after stroke: the role of anxiety and upper limb dysfunction","volume":"35","author":"Morris","year":"2013","journal-title":"Disab. Rehabilit"},{"key":"B40","doi-asserted-by":"publisher","first-page":"32055","DOI":"10.1007\/s11042-019-08020-2","article-title":"Continues online exercise monitoring and assessment system with visual guidance feedback for stroke rehabilitation","volume":"78","author":"Mortazavi","year":"2019","journal-title":"Multim. Tools Applic"},{"key":"B41","doi-asserted-by":"publisher","first-page":"15288","DOI":"10.5555\/3495724.3497006","article-title":"Calibrating deep neural networks using focal loss","volume":"33","author":"Mukhoti","year":"2020","journal-title":"Adv. Neural Inf. Proces. Syst"},{"key":"B42","doi-asserted-by":"publisher","first-page":"2197","DOI":"10.1016\/S0140-6736(12)61689-4","article-title":"Disability-adjusted life years (dalys) for 291 diseases and injuries in 21 regions, 1990-2010: a systematic analysis for the global burden of disease study 2010","volume":"380","author":"Murray","year":"2012","journal-title":"Lancet"},{"key":"B43","doi-asserted-by":"publisher","first-page":"611","DOI":"10.1191\/0269215500cr369oa","article-title":"Barriers to achieving evidence-based stroke rehabilitation","volume":"14","author":"Pollock","year":"2000","journal-title":"Clin. Rehabilit"},{"key":"B44","doi-asserted-by":"publisher","first-page":"737","DOI":"10.1111\/j.1365-2648.2008.04925.x","article-title":"Task-oriented training in rehabilitation after stroke: systematic review","volume":"65","author":"Rensink","year":"2009","journal-title":"J. Adv. Nurs"},{"key":"B45","doi-asserted-by":"publisher","first-page":"218898","DOI":"10.1109\/ACCESS.2020.3041822","article-title":"Sensor-based human activity recognition using deep stacked multilayered perceptron model","volume":"8","author":"Rustam","year":"2020","journal-title":"IEEE Access"},{"key":"B46","doi-asserted-by":"publisher","first-page":"1627","DOI":"10.1021\/ac60214a047","article-title":"Smoothing and differentiation of data by simplified least squares procedures","volume":"36","author":"Savitzky","year":"1964","journal-title":"Analyt. Chem"},{"key":"B47","doi-asserted-by":"publisher","first-page":"111","DOI":"10.3390\/s22010111","article-title":"Detection of typical compensatory movements during autonomously performed exercises preventing low back pain (lbp)","volume":"22","author":"Sellmann","year":"2022","journal-title":"Sensors"},{"key":"B48","author":"Senior","year":"2004","journal-title":"Qualisys track manager"},{"key":"B49","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/DICTA.2016.7797053","article-title":"\u201cImpact of automatic feature extraction in deep learning architecture,\u201d","author":"Shaheen","year":"2016","journal-title":"2016 International Conference on Digital Image Computing: Techniques and Applications (DICTA)"},{"key":"B50","doi-asserted-by":"publisher","first-page":"105524","DOI":"10.1016\/j.asoc.2019.105524","article-title":"Investigating the impact of data normalization on classification performance","volume":"97","author":"Singh","year":"2020","journal-title":"Appl. Soft Comput"},{"key":"B51","doi-asserted-by":"publisher","first-page":"1607","DOI":"10.1109\/BioRob.2012.6290668","article-title":"\u201cVision-based posture assessment to detect and categorize compensation during robotic rehabilitation therapy,\u201d","author":"Taati","year":"2012","journal-title":"2012 4th IEEE RAS &EMBS International Conference on Biomedical Robotics and Biomechatronics (BioRob)"},{"key":"B52","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1186\/s12984-022-00986-9","article-title":"Impact of the robotic-assistance level on upper extremity function in stroke patients receiving adjunct robotic rehabilitation: sub-analysis of a randomized clinical trial","volume":"19","author":"Takebayashi","year":"2022","journal-title":"J. NeuroEng. Rehabilit"},{"key":"B53","doi-asserted-by":"publisher","first-page":"359728","DOI":"10.1155\/2012\/359728","article-title":"Maladaptive plasticity for motor recovery after stroke: mechanisms and approaches","volume":"2012","author":"Takeuchi","year":"2012","journal-title":"Neural Plast"},{"key":"B54","doi-asserted-by":"publisher","first-page":"1613","DOI":"10.1016\/j.apmr.2004.01.028","article-title":"Rehabilitation of reaching after stroke: task-related training versus progressive resistive exercise","volume":"85","author":"Thielman","year":"2004","journal-title":"Arch. Phys. Med. Rehabilit"},{"key":"B55","doi-asserted-by":"publisher","first-page":"457","DOI":"10.1177\/030802269005301104","article-title":"The use of home therapy programmes for improving recovery of the upper limb following stroke","volume":"53","author":"Turton","year":"1990","journal-title":"Br. J. Occup. Ther"},{"key":"B56","doi-asserted-by":"crossref","first-page":"432","DOI":"10.1109\/ICAIIC48513.2020.9064992","article-title":"\u201cAnalysis of detecting compensation for robotic stroke rehabilitation therapy using imbalanced learning and outlier detection,\u201d","volume-title":"2020 International Conference on Artificial Intelligence in Information and Communication (ICAIIC)","author":"Uy","year":"2020"},{"key":"B57","doi-asserted-by":"publisher","first-page":"28","DOI":"10.3389\/frobt.2015.00028","article-title":"A review of human activity recognition methods","volume":"2","author":"Vrigkas","year":"2015","journal-title":"Front. Robot. AI"},{"key":"B58","doi-asserted-by":"publisher","first-page":"381","DOI":"10.1016\/S0893-6080(01)00026-0","article-title":"Quantitative examinations for multi joint arm trajectory planning\u2014using a robust calculation algorithm of the minimum commanded torque change trajectory","volume":"14","author":"Wada","year":"2001","journal-title":"Neural Netw"},{"key":"B59","doi-asserted-by":"publisher","first-page":"743","DOI":"10.1007\/s11036-019-01445-x","article-title":"Deep learning models for real-time human activity recognition with smartphones","volume":"25","author":"Wan","year":"2020","journal-title":"Mobile Netw. Applic"},{"key":"B60","doi-asserted-by":"publisher","first-page":"e68013","DOI":"10.7554\/eLife.68013.sa2","article-title":"The energetic basis for smooth human arm movements","volume":"10","author":"Wong","year":"2021","journal-title":"Elife"},{"key":"B61","doi-asserted-by":"publisher","first-page":"3521","DOI":"10.3934\/mbe.2021177","article-title":"Multi-feature gait recognition with dnn based on semg signals","volume":"18","author":"Yao","year":"2021","journal-title":"Math. Biosci. Eng"},{"key":"B62","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1155\/2021\/6630492","article-title":"A control and posture recognition strategy for upper-limb rehabilitation of stroke patients","volume":"2021","author":"Yu","year":"2021","journal-title":"Wirel. Commun. Mobile Comput"},{"key":"B63","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/JTEHM.2017.2780836","article-title":"Automatic detection of compensation during robotic stroke rehabilitation therapy","volume":"6","author":"Zhi","year":"2017","journal-title":"IEEE J. Transl. Eng. Health Med"},{"key":"B64","doi-asserted-by":"publisher","first-page":"11","DOI":"10.4173\/mic.2017.1.2","article-title":"Energy optimal trajectories in human arm motion aiming for assistive robots","volume":"38","author":"Zhou","year":"2017","journal-title":"Model. Identific. Control"},{"key":"B65","doi-asserted-by":"publisher","first-page":"197","DOI":"10.1109\/TPAMI.2022.3147841","article-title":"Consistency and diversity induced human motion segmentation","volume":"45","author":"Zhou","year":"2022","journal-title":"IEEE Trans. Patt. Analy. Mach. Intell"}],"container-title":["Frontiers in Neurorobotics"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fnbot.2023.1155826\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,7,15]],"date-time":"2023-07-15T08:47:43Z","timestamp":1689410863000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fnbot.2023.1155826\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,7,14]]},"references-count":65,"alternative-id":["10.3389\/fnbot.2023.1155826"],"URL":"https:\/\/doi.org\/10.3389\/fnbot.2023.1155826","relation":{},"ISSN":["1662-5218"],"issn-type":[{"value":"1662-5218","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,7,14]]},"article-number":"1155826"}}