Shoulder physiotherapy exercise recognition: machine learning the inertial signals from a smartwatch
Open accessThe DOI matches its Crossref record; 5 of the 5 references it cites were found in a registry or an index.
Abstract
OBJECTIVE: Participation in a physical therapy program is considered one of the greatest predictors of successful conservative management of common shoulder disorders. However, adherence to these protocols is often poor and typically worse for unsupervised home exercise programs. Currently, there are limited tools available for objective measurement of adherence in the home setting. The goal of this study was to develop and evaluate the potential for performing home shoulder physiotherapy monitoring using a commercial smartwatch. APPROACH: Twenty healthy adult subjects with no prior shoulder disorders performed seven exercises from an evidence-based rotator cuff physiotherapy protocol, while 6-axis inertial sensor data was collected from the active extremity. Within an activity recognition chain (ARC) framework, four supervised learning algorithms were trained and optimized to classify the exercises: k-nearest neighbor (k-NN), random forest (RF), support vector machine classifier (SVC), and a convolutional recurrent neural network (CRNN). Algorithm performance was evaluated using 5-fold cross-validation stratified first temporally and then by subject. MAIN RESULTS: Categorical classification accuracy was above 94% for all algorithms on the temporally stratified cross validation, with the best performance achieved by the CRNN algorithm (99.4%). The subject stratified cross validation, which evaluated classifier performance on unseen subjects, yielded lower accuracies scores again with CRNN performing best (88.9%). SIGNIFICANCE: This proof of concept study demonstrates the technical feasibility of a smartwatch device and supervised machine learning approach to more easily monitor and assess the at-home adherence of shoulder physiotherapy exercise protocols.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Geurts, P., Ernst, D., & Wehenkel, L. (2006) · Machine Learning · doi:10.1007/s10994-006-6226-1 · Cited by 9,300
Deep Convolutional and LSTM Recurrent Neural Networks for Multimodal Wearable Activity Recognition
Ordóñez, F., & Roggen, D. (2016) · Sensors · doi:10.3390/s16010115 · Cited by 2,761 · cited 3×
“…convolutional input layers to generate a feature representation of the signal tensor which is learned by the recurrent layers utilizing long short term memory (LSTM) cells, an approach utilized recently for video (Yue-Hei Ng et al. 2015) and inertial signal (Ordonez and Roggen 2016) classification.”
BackgroundMethodA tutorial on human activity recognition using body-worn inertial sensors
Bulling, A., Blanke, U., & Schiele, B. (2014) · ACM Computing Surveys · doi:10.1145/2499621 · Cited by 1,668 · cited 4×
“Although other authors (Gonzalez et al. 2015; Bulling et al. 2014) have utilized inertial sensor fusion for HAR, only raw sensor data is utilized in this study.”
BackgroundMethodInfluentialKuhn, J. E., Dunn, W. R., Sanders, R., An, Q., Baumgarten, K. M., Bishop, J. Y., Brophy, R. H., Carey, J. L., Holloway, B. G., Jones, G. L., et al. (2013) · Journal of Shoulder and Elbow Surgery · doi:10.1016/j.jse.2013.01.026 · Cited by 351
Features and models for human activity recognition
González, S., Sedano, J., Villar, J. R., Corchado, E., Herrero, Á., & Baruque, B. (2015) · Neurocomputing · doi:10.1016/j.neucom.2015.01.082 · Cited by 73
Research works printed but not linked: 22
- Attal, 2015, Physical human activity recognition using wearable sensors, Sensors, 15, 31314, 10.3390/s151229858 · DOI · Paper page
- Banos, 2014, Window size impact in human activity recognition, Sensors, 14, 6474, 10.3390/s140406474 · DOI · Paper page
- Crabolu, 2017, In vivo estimation of the shoulder joint center of rotation using magneto-inertial sensors: MRI-based accuracy and repeatability assessment, Biomed. Eng. Online, 16, 10.1186/s12938-017-0324-0 · DOI
- Crema, 2017, IMU-based solution for automatic detection and classification of exercises in the fitness scenario, 1, 10.1109/SAS.2017.7894068 · DOI
- De Baets, 2017, Shoulder assessment according to the international classification of functioning by means of inertial sensor technologies: a systematic review, Gait Posture, 57, 278, 10.1016/j.gaitpost.2017.06.025 · DOI
- Dietterich, 2002, Machine learning for sequential data: a review, 10.1007/3-540-70659-3_2 · DOI · Paper page
- Garcia-Ceja, 2014, Long-term activity recognition from wristwatch accelerometer data, Sensors, 14, 22500, 10.3390/s141222500 · DOI
- Holden, 2014, Recommendations for exercise adherence measures in musculoskeletal settings: a systematic review and consensus meeting (protocol), Systematic Rev., 3, 10, 10.1186/2046-4053-3-10 · DOI · Paper page
- Jack, 2010, Barriers to treatment adherence in physiotherapy outpatient clinics: a systematic review, Manual Ther., 15, 220, 10.1016/j.math.2009.12.004 · DOI · Paper page
- Korver, 2014, Inertia based functional scoring of the shoulder in clinical practice, Physiol. Meas., 35, 167, 10.1088/0967-3334/35/2/167 · DOI
- Louppe, 2013, Understanding variable importances in forests of randomized trees, Adv. Neural Inf. Process. Syst., 1, 431
- McGirr, 2015, An elastic exercise band mounted with a Bandcizer can differentiate between commonly prescribed home exercises for the shoulder, Int. J. Sports Phys. Ther., 10, 332
- Mitchell, 2005, Shoulder pain: diagnosis and management in primary care, BMJ Br. Med. J., 331, 1124, 10.1136/bmj.331.7525.1124 · DOI · Paper page
- Pan, 2013, Intelligent shoulder joint home-based self-rehabilitation monitoring system, Int. J. Smart Home, 7, 395, 10.14257/ijsh.2013.7.5.38 · DOI
- Pedregosa, 2011, Scikit-learn: machine learning in Python, J. Mach. Learn. Res., 12, 2825
- Picerno, 2015, Ambulatory assessment of shoulder abduction strength curve using a single wearable inertial sensor, J. Rehabil. Res. Dev., 52, 171, 10.1682/JRRD.2014.06.0146 · DOI
- Pichonnaz, 2015, Enhancing clinically-relevant shoulder function assessment using only essential movements, Physiol. Meas., 36, 547, 10.1088/0967-3334/36/3/547 · DOI
- Roldán-Jiménez, 2016, Age-related changes analyzing shoulder kinematics by means of inertial sensors, Clin. Biomech., 37, 70, 10.1016/j.clinbiomech.2016.06.004 · DOI
- Rose, 2017, Wearable inertial sensors allow for quantitative assessment of shoulder and elbow kinematics in a cadaveric knee arthroscopy model, Arthroscopy J. Arthroscopic Relat. Surg., 33, 2110, 10.1016/j.arthro.2017.06.042 · DOI
- Tashjian, 2012, Epidemiology, natural history, and indications for treatment of rotator cuff tears, Clin. Sports Med., 31, 589, 10.1016/j.csm.2012.07.001 · DOI
- Werner, 2014, Validation of an innovative method of shoulder range-of-motion measurement using a smartphone clinometer application, J. Shoulder Elbow Surg., 23, e275, 10.1016/j.jse.2014.02.017 · doi:10.1016/j.jse.2014.02.017
- Yang, 2008, Using acceleration measurements for activity recognition: an effective learning algorithm for constructing neural classifiers, Pattern Recognit. Lett., 29, 2213, 10.1016/j.patrec.2008.08.002 · DOI · Paper page
Cited by
Alsobhi, M., Sachdev, H. S., Chevidikunnan, M. F., Basuodan, R., K U, D. K., & Khan, F. (2022) · International Journal of Environmental Research and Public Health · doi:10.3390/ijerph192315919 · Cited by 53
Correia, F. D., Molinos, M., Luís, S., Carvalho, D., Carvalho, C., Costa, P., Seabra, R., Francisco, G., Bento, V., & Lains, J. (2022) · American Journal of Physical Medicine & Rehabilitation · doi:10.1097/phm.0000000000001780 · Cited by 52
Janela, D., Costa, F., Molinos, M., Moulder, R. G., Lains, J., Francisco, G. E., Bento, V., Cohen, S. P., & Correia, F. D. (2022) · Journal of Pain Research · doi:10.2147/jpr.s343308 · Cited by 33
“37,38 47 Most papers report the feasibility of different hardware/platforms 48–53 or are pilot trials assessing pertinence of telerehabilitation strategies to deliver exercise programs, and only a few include built-in feedback/communication between patients and providers.”
BackgroundCote, M. P., & Lubowitz, J. H. (2024) · Arthroscopy · doi:10.1016/j.arthro.2023.12.027 · Cited by 8
Metadata
- Authors
- Burns, D. M., Leung, N., Hardisty, M., Whyne, C. M., Henry, P., & McLachlin, S.
- Venue
- Physiological Measurement
- Publisher
- IOP Publishing
- Published
- 2018-07-23
- Volume
- 39
- Issue
- 7
- Pages
- 075007
- ISSN
- 0967-3334 (ISSN-L)
- 1361-6579
- OpenAlex
- W2786289268
- Cited by
- 103
- References
- 5
- FWCI
- 2.49
- Type
- journal article
- Language
- en
- Open access
- bronze
- Licence
- publishingsupport.iopscience.iop.org/iop-standard/v1 (vor, 22 d)
- iopscience.iop.org/info/page/text-and-data-mining (tdm, 22 d)
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Shoulder physiotherapy exercise recognition: machine learning the inertial signals from a smartwatch. https://app.cochat.ai/papers/10.1088/1361-6579/aacfd9
- The paper
- Burns, D. M., Leung, N., Hardisty, M., Whyne, C. M., Henry, P., & McLachlin, S. (2018). Shoulder physiotherapy exercise recognition: machine learning the inertial signals from a smartwatch. Physiological Measurement. https://doi.org/10.1088/1361-6579/aacfd9
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.