Rehab-Net: Deep Learning Framework for Arm Movement Classification Using Wearable Sensors for Stroke Rehabilitation
The DOI matches its Crossref record; 60 of the 62 references it cites were found in a registry or an index.
Abstract
In this paper, we present a deep learning framework “Rehab-Net” for effectively classifying three upper limb movements of the human arm, involving extension, flexion, and rotation of the forearm, which, over the time, could provide a measure of rehabilitation progress. The proposed framework, Rehab-Net is formulated with a personalized, light weight and low-complex, customized convolutional neural network (CNN) model, using two-layers of CNN, interleaved with pooling layers, followed by a fully connected layer that classifies the three movements from tri-axial acceleration input data collected from the wrist. The proposed Rehab-Net framework was validated on sensor data collected in two situations: 1) semi-naturalistic environment involving an archetypal activity of “making-tea” with four stroke survivors and 2) natural environment, where ten stroke survivors were free to perform any desired arm movement for the duration of 120 min. We achieved an overall accuracy of 97.89% on semi-naturalistic data and 88.87% on naturalistic data which exceeded state-of-the-art learning algorithms namely, linear discriminant analysis, support vector machines, and k-means clustering with an average accuracy of 48.89%, 44.14%, and 27.64%. Subsequently, a computational complexity analysis of the proposed model has been discussed with an eye toward hardware implementation. The clinical significance of this study is to accurately monitor the clinical progress of the rehabilitated subjects under t…
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“In addition, in order to design efficient ML-based assessment systems that generalises to new users a large number of participants should take part with different and variant degrees of severity [43,48,49,69,70], which is not always available and taken into account.”
From Wearable sensors and machine learning in post-stroke rehabilitation assessment: A systematic review (2022) · cites this paper as [43]
In the citing paper
“Another ANN architecture that was employed is Convolutional Neural Network (CNN) Architectures [43,45,50,54].”
From Wearable sensors and machine learning in post-stroke rehabilitation assessment: A systematic review (2022) · cites this paper as [43]
In the citing paper
“many studies proved the possibility to design effective systems with a very small number of sensors (sometimes a single sensor is sufficient) as is the case in [43,51].”
From Wearable sensors and machine learning in post-stroke rehabilitation assessment: A systematic review (2022) · cites this paper as [43]
Citations
Cites
ImageNet classification with deep convolutional neural networks
Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2017) · Communications of the ACM · doi:10.1145/3065386 · Cited by 112,155
Adam: A Method for Stochastic Optimization
Kingma, D. P., & Ba, J. (2014) · arXiv · doi:10.48550/arxiv.1412.6980 · Cited by 98,946 · cited 1×
“Sparse categorical cross-entropy loss function and ‘Adam optimizer’ [56] is used to update the weights faster than the popularly used stochastic gradient descent (SGD).”
MethodHeart Disease and Stroke Statistics—2017 Update: A Report From the American Heart Association
Benjamin, E. J., Blaha, M. J., Chiuve, S. E., Cushman, M., Das, S. R., Deo, R., de Ferranti, S. D., Floyd, J., Fornage, M., Gillespie, C., et al. (2017) · Circulation · doi:10.1161/cir.0000000000000485 · Cited by 12,591 · cited 1×
“S TROKE is a global public health issue, being the second leading cause of death worldwide, besides post-stroke outcomes such as disability affects the quality of life [1], [2].”
BackgroundHan, S., Mao, H., & Dally, W. J. (2015) · arXiv · doi:10.48550/arxiv.1510.00149 · Cited by 4,662
Deep learning for healthcare: review, opportunities and challenges
Miotto, R., Wang, F., Wang, S., Jiang, X., & Dudley, J. T. (2018) · Briefings in Bioinformatics · doi:10.1093/bib/bbx044 · Cited by 3,076 · cited 1×
“[61].”
Deep learning for sensor-based activity recognition: A survey
Wang, J., Chen, Y., Hao, S., Peng, X., & Hu, L. (2019) · Pattern Recognition Letters · doi:10.1016/j.patrec.2018.02.010 · Cited by 2,014 · cited 1×
“been covered in detail in the review [30], [45].”
BackgroundAPI design for machine learning software: experiences from the scikit-learn project
Buitinck, L., Louppe, G., Blondel, M., Pedregosa, F., Mueller, A., Grisel, O., Niculae, V., Prettenhofer, P., Gramfort, A., Grobler, J., et al. (2013) · arXiv (Cornell University) · doi:10.48550/arxiv.1309.0238 · Cited by 1,871
Activity recognition from accelerometer data
Ravi, N., Dandekar, N., Mysore, P., & Littman, M. L. (2005) · Innovative Applications of Artificial Intelligence · Cited by 1,775
A 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 1×
“nition is to identify human behavior in real time scenario to provide proactive assistance to users even outside the conventional clinical setting [10].”
BackgroundWearable Sensors for Human Activity Monitoring: A Review
Mukhopadhyay, S. C. (2015) · IEEE Sensors Journal · doi:10.1109/jsen.2014.2370945 · Cited by 1,401
Cited by
Items 101–158 of 158
Li, F., Tao, Z., Li, R., & Qu, Z. (2021) · Personal and Ubiquitous Computing · doi:10.1007/s00779-021-01520-9 · Cited by 4
RetractedAn Improved SVM Method for Movement Recognition of Lower Limbs by MIMU and sEMG
Yun, X., Xu, L., Gao, L., Zhanhao, L., & Bohan, S. (2023) · International Journal of Advanced Computer Science and Applications · doi:10.14569/ijacsa.2023.0140113 · Cited by 3
“INTRODUCTION Along with the development of artificial intelligence technology, many researchers have focused on the study of human posture recognition [1-4].”
BackgroundCuffless Non-invasive Blood Pressure Measurement Using CNN-LSTM Model: A Correlation Study
Gan, K. B., & Leong, J. S. V. (2023) · International Journal on Robotics Automation and Sciences · doi:10.33093/ijoras.2023.5.2.3 · Cited by 6
Tong, L., Liu, D., Peng, L., Hao, H., & Wang, C. (2023) · IEEE Sensors Journal · doi:10.1109/jsen.2023.3325344 · Cited by 3
“A big amount of training data can support deep learning models to improve the stability of the network and make the network more robust to unseen or untrained data [38], [39].”
BackgroundXia, C., & Sugiura, Y. (2021) · IEEE Access · doi:10.1109/access.2021.3134262 · Cited by 4
“The accelerometer can be not only embedded in consumer electronics but also specialized to monitor specific people's activity and conduct related applications [25, 26].”
BackgroundUnsupervised Learning of Wearable Sensor Data for Assessing Upper Limb Function
Samweslin, S., Janani, S., & Rose, S. G. H. (2024) · International Conference on Communication, Computing and Energy Efficient Technologies (I3CEET) · doi:10.1109/i3ceet61722.2024.10994002 · Cited by 2
Bilgin, T. T., Avcı, M. F., Günay, S. M., Şahin, B., Sayaca, Ç., Altan, L., Özkal, Ö., Coşkun, T., & Özkaynak, H. (2025) · The European Research Journal · doi:10.18621/eurj.1672422 · Cited by 2
Theerthagiri, P., Ruby, A. U., & George Chellin Chandran, J. (2025) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2025.111184 · Cited by 2
Machine Learning Based Robotic-Assisted Upper Limb Rehabilitation Therapies: A Review
Shymala Gowri, S., Priya, N. H., Rathika, P. D., & Lakshmi, M. M. (2023) · Algorithms for intelligent systems · doi:10.1007/978-981-19-7892-0_6 · Cited by 2
Pahwa, B., Tayal, A., & Garg, K. (2024) · World Neurosurgery · doi:10.1016/j.wneu.2024.01.059 · Cited by 2
An Advanced Smart Glove for Stroke Rehabilitation Using IOT
Giri, K., Logeshwari, M., Abhishek, A. V., Dharun, R., Jeevitha, S., Mohanaprasath, K. V., & Mangai, S. (2025) · International Conference on Bio Signals, Images, and Instrumentation (ICBSII) · doi:10.1109/icbsii65145.2025.11013342 · Cited by 2
Mekruksavanich, S., Phaphan, W., & Jitpattanakul, A. (2025) · Lobachevskii Journal of Mathematics · doi:10.1134/s1995080224608166 · Cited by 5
“Panwar et al. [18] introduced a technique using two 1D-CNN layers to handle a single IMU signal from the forearm, achieving an accuracy of 97.89 % for three semi-naturalistic forearm motions.”
InfluentialAnalysis and synthesis of function data of human movement
Ma, Y., & Sharaf, S. A. (2021) · Applied Mathematics and Nonlinear Sciences · doi:10.2478/amns.2021.2.00086 · Cited by 2
“It is similar to traditional algorithms [10].”
Machine learning for healthcare using wearable sensors
Rad, N. M., & Marchiori, E. (2021) · Elsevier eBooks · doi:10.1016/b978-0-12-818914-6.00007-7 · Cited by 3
Zuo, C., Yin, Y., Wang, H., Zheng, Z., Ma, X., Yang, Y., Wang, J., Wang, S., Huang, Z., & Ye, C. (2025) · Scientific Data · doi:10.1038/s41597-025-05767-2 · Cited by 5
“Publicly available MI datasets, including those from BCI competitions 30 , BCI 2000 31 , and various journal publications 29–32 , provide valuable resources but are constrained by small sample sizes, short recording durations, and a predominant focus on healthy participants 13,33,34 .”
Brain Stroke Prediction Using the Artificial Intelligence
Maheshwari, H., Yadav, D., & Chandra, U. (2022) · Communications in computer and information science · doi:10.1007/978-3-031-23647-1_1 · Cited by 3
Babu, B. M., & Sandhya, B. (2024) · MATEC Web of Conferences · doi:10.1051/matecconf/202439201161 · Cited by 1
“These techniques can be adapted and combined to address the specific challenges of time series classification tasks [25].”
BackgroundLee, S., Lee, S., Yun, S., Ryu, S., & Jeong, I. C. (2025) · IEEE Sensors Journal · doi:10.1109/jsen.2025.3581211 · Cited by 4
Palta, P., Palta, A., & Kumar, V. (2025) · ECS Journal of Solid State Science and Technology · doi:10.1149/2162-8777/ae2c1c · Cited by 3
Chen, C.-H., Tang, L.-H., Yeh, C.-H., Wu, E. H., & Yeh, S. (2026) · Electronics · doi:10.3390/electronics15071459 · Cited by 1
Body Motion Capture and Applications
Akman, Ç., & Sönmez, T. (2021) · Advances in computational intelligence and robotics book series · doi:10.4018/978-1-7998-7468-3.ch009 · Cited by 2
Virtual Reality, Robotics, and Artificial Intelligence
Kanade, A., Sharma, M., & Manivannan, M. (2022) · Advances in medical technologies and clinical practice book series · doi:10.4018/978-1-6684-3533-5.ch005 · Cited by 1
Classification of pathological and healthy individuals for computer-aided physical rehabilitation
Kramer, I., Memmesheimer, R., & Paulus, D. W. R. (2023) · IEEE/SICE International Symposium on System Integration (SII) · doi:10.1109/sii55687.2023.10039185 · Cited by 1
“al [15] introduced the ”Rehab-Net” framework, which utilizes a CNN to classify basic upper limb movements of the human arm.”
MethodMassoud, M., Mahmoud, G., Ali, W. Y., & Ahmed, W. (2024) · International Journal of Telecommunications · doi:10.21608/ijt.2024.262285.1042 · Cited by 1
Wearable Loops for Dynamic Monitoring of Joint Flexion: A Machine Learning Approach
Saltzman, H., Rajaram, R., Zhang, Y., Islam, M. A., & Kiourti, A. (2024) · Electronics · doi:10.3390/electronics13122245 · Cited by 1
“During the grid searching process, the following hyperparameters were chosen to be variable: learning rate, batch size, number of epochs, and layer size [32].”
Motion Analysis Tool for Diagnosis of Orthopedic Disorders
Jičinský, P., & Mareš, J. (2020) · Transactions of the VŠB - Technical University of Ostrava Mechanical Series · doi:10.22223/tr.2020-1/2058 · Cited by 1
PPG-Based Non-invasive Methodologies for Pervasive Monitoring of Vitals: BP and HR
Panwar, M., Gautam, A., & Acharyya, A. (2021) · Springer eBooks · doi:10.1007/978-981-16-5324-7_3 · Cited by 1
Self-citationPost-Stroke Virtual Assessment Using Deep Learning
Razfar, N., Kashef, R. F., & Mohammadi, F. A. (2023) · IEEE Conference on Artificial Intelligence (CAI) · doi:10.1109/cai54212.2023.00056 · Cited by 1
Li, Z., Fathima, G. S., & Kautish, S. (2021) · Journal of Intelligent & Fuzzy Systems · doi:10.3233/jifs-219010
Identifying Stroke Indicators Using Rough SetsPreprint
Pathan, M. S., Zhang, J., John, D., Nag, A., & Dev, S. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2110.10152
“Models like support vector machine [18], decision trees [19], deep neural network [20], and ensemble methods [21] have achieved notable results to classify stroke outcomes when provided with relevant data.”
BackgroundAn Investigation of Paralysis Attack Using Machine Learning Approach
Surya, S., & Ramamoorthy, S. (2022) · Advances in intelligent systems and computing · doi:10.1007/978-981-16-5652-1_12
2019 Index IEEE Transactions on Biomedical Engineering Vol. 66
(2019) · IEEE Transactions on Biomedical Engineering · doi:10.1109/tbme.2020.2964087
TMCA-Net: A Compact Convolution Network for Monitoring Upper Limb Rehabilitation
Liu, Q., Wu, Z., & Liu, X. (2022) · Journal on internet of things · doi:10.32604/jiot.2022.040368
Multi Model Fusion Based Prediction of Human Joint Angles Using sEMG and Historical Angles
Han, J., Wang, H., & Tian, Y. (2023) · Chinese Control Conference (CCC) · doi:10.23919/ccc58697.2023.10240374 · Cited by 1
An Effective Application of Deep Learning in the Early Diagnosis of Stroke
Babu, C. R., & Pushpalatha, N. (2023) · International Conference on Pervasive Computing and Social Networking (ICPCSN) · doi:10.1109/icpcsn58827.2023.00085
“[8] introduced a deep learning method called "Rehab-Net" for identifying 3 movements of the person's arm that involve elongation, flexion, and rotations of the wrist.”
MethodAzizan, A. (2024) · Research Square · doi:10.21203/rs.3.rs-3923026/v1 · Cited by 2
Feature representation learning for human activity recognition
Khaertdinov, B. (2024) · Research Publications (Maastricht University) · doi:10.26481/dis.20240618bk · Cited by 1
Han, S., Park, C., & YOU, S. (. H. (2024) · Journal of Mechanics in Medicine and Biology · doi:10.1142/s0219519424400785
SPVD: A Stroke Patient Motor Movements Video Dataset for AI Augmented Rehabilitation
Sherazi, T. J., Ahmad, W., & Ushaq, M. (2023) · International Bhurban Conference on Applied Sciences and Technology (IBCAST) · doi:10.1109/ibcast59916.2023.10713023
“Stroke is among those chronic diseases, which are the leading causes of disability in the world [2].”
BackgroundMethodOh, W., Shin, J., & Park, C. (2025) · Journal of Mechanics in Medicine and Biology · doi:10.1142/s0219519425400019
Martínez-Pascual, D., Vales, Y., Batanero, R. M., Úbeda, P. R., Catalan, J. M., Pérez, L. D. L., & García-Aracil, N. (2025) · Jornadas de Automática · doi:10.17979/ja-cea.2025.46.12166
Akbari, S., Horemans, H. L. D., Bussmann, J. B. J., & Zgonnikov, A. (2025) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2025.111167 · Cited by 3
Akbarifar, F., Maghsoodi, N., Dukelow, S. P., Scott, S. H., & Mousavi, P. (2026) · Journal of NeuroEngineering and Rehabilitation · doi:10.1186/s12984-026-02012-8
Lee, K., Kim, J.-H., Hong, H., Jeong, Y., Ryu, H., Kim, H., & Lee, S.-U. (2024) · Figshare · doi:10.6084/m9.figshare.26698568
Lee, K., Kim, J.-H., Hong, H., Jeong, Y., Ryu, H., Kim, H., & Lee, S.-U. (2024) · Figshare · doi:10.6084/m9.figshare.25495186
Lee, K., Kim, J.-H., Hong, H., Jeong, Y., Ryu, H., Kim, H., & Lee, S.-U. (2024) · Figshare · doi:10.6084/m9.figshare.26698571
Lee, K., Kim, J.-H., Hong, H., Jeong, Y., Ryu, H., Kim, H., & Lee, S.-U. (2024) · Figshare · doi:10.6084/m9.figshare.26698565
Lee, K., Kim, J.-H., Hong, H., Jeong, Y., Ryu, H., Kim, H., & Lee, S.-U. (2024) · Figshare · doi:10.6084/m9.figshare.26698562
Lee, K., Kim, J.-H., Hong, H., Jeong, Y., Ryu, H., Kim, H., & Lee, S.-U. (2024) · Figshare · doi:10.6084/m9.figshare.25495248
Towards a Methodology for Using EEG Motor Imagery to Control Video Games
Moschella, F., Giudice, N. D., Loreti, M., & Vilhjálmsson, H. H. (2025) · SeGAH ... IEEE ... International Conference on Serious Games and Applications for Health/IEEE International Conference on Serious Games and Applications for Health · doi:10.1109/segah65397.2025.11168439
“Several gamification systems have been developed for rehabilitation, such as the RehabNet [14], a deep learning framework for classifying extension, flexion and rotation of the forearm, that can in time provide a measure of the rehabilitation progress, the RehabGlove [15], a robotic glove that can…”
Human Activity Recognition Reimagined: The CGRNHAR Self-Supervised Approach
Challa, A. K. R., Raavi, V. K., Bandari, H., Gude, D. K., & Kumar, S. (2025) · Global AI Summit - International Conference on Artificial Intelligence and Emerging Technology (AI Summit) · doi:10.1109/aisummit66170.2025.11411187
An Edge-Enabled Web of Things Framework for Home Rehabilitation Monitoring of Older Adults
Anuraj, B., Aerts, J., & Calbimonte, J.-P. (2026) · IEEE Access · doi:10.1109/access.2026.3671678
“…devices, such as wearable and environmental sensors, including smartwatches, motion-sensing wristbands, air quality monitors, and temperature sensors, can capture the movement and environmental conditions of older adults while they are at home and conducting rehabilitation exercises [3], [4].”
MobileRehab: Smartphone-Based Rehabilitation Monitoring Using Active Acoustic Sensing
Feng, X., Dong, K., Yang, X., & Xia, H. (2026) · Lecture notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering · doi:10.1007/978-3-032-21168-2_19
Kamal, Neog, D. R., Bhuyan, M. K., MacDorman, K. F., Karsh, R. K., & Laskar, R. H. (2026) · IET Computer Vision · doi:10.1049/cvi2.70066
6G-powered human motion synthesis framework for artificial intelligence–enabled pose guidance
Sharanya, S., Simran, Drusilla, S., Usman, M., & Pongiannan, K. (2026) · Elsevier eBooks · doi:10.1016/b978-0-443-45875-0.00001-6
Intelligent Rehabilitation-Action Evaluation System via Enhanced DTW and Deep Learning
Han, Y., Xu, Y., Lu, H., & Gan, Y. (2026) · Smart innovation, systems and technologies · doi:10.1007/978-3-032-25613-3_22
Islam, A., Rahman, A., Salgado, A. A. V., Scharf, R. J., Blemker, S. S., & Barnes, L. E. (2026) · IEEE International Conference on Healthcare Informatics (ICHI) · doi:10.1109/ichi69079.2026.00045
“Data augmentation provides only modest gains [17], [24], motivating transfer learning approaches that leverage representations learned from large-scale motion datasets.”
Ling, X., & Yin, H. (2026) · Discover Artificial Intelligence · doi:10.1007/s44163-026-02038-w
Metadata
- Authors
- Panwar, M., Biswas, D., Bajaj, H., Jobges, M., Turk, R., Maharatna, K., & Acharyya, A.
- Affiliations
- Indian Institute of Technology Hyderabad (IN)
- IMEC (BE)
- National Institute of Technology Tiruchirappalli (IN)
- University of Southampton (GB)
- Venue
- IEEE Transactions on Biomedical Engineering
- Published
- 2019-11
- Volume
- 66
- Issue
- 11
- Pages
- 3026-3037
- ISSN
- 0018-9294 (ISSN-L)
- 1558-2531
- OpenAlex
- W2916879374
- Cited by
- 158
- References
- 62
- Citations by year
- 2026: 9 · 2025: 22 · 2024: 26 · 2023: 24 · 2022: 28 · 2021: 29 · 2020: 18 · 2019: 2
- FWCI
- 8.62
- Type
- journal article
- Language
- en
- Open access
- closed
- Licence
- Locations
- IEEE Transactions on Biomedical Engineering (publishedVersion)
- PubMed (publishedVersion)
- ePrints Soton (University of Southampton) (acceptedVersion)
- Research Archive of Indian Institute of Technology Hyderabad (Indian Institute of Technology Hyderabad) (submittedVersion)
- Funders
- Science and Engineering Research Board (ECR/2015/000148)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Context-Aware Activity Recognition Systems
- Health Sciences › Medicine › Rehabilitation › Stroke Rehabilitation and Recovery
- Life Sciences › Neuroscience › Cognitive Neuroscience › EEG and Brain-Computer Interfaces
- Keywords
- stroke rehabilitation, upper limb movement classification, wearable sensors, triaxial accelerometer, convolutional neural networks, rehabilitation progress monitoring, outpatient rehabilitation, forearm rotation
- MeSH terms
- Stroke Rehabilitation, Wearable Electronic Devices, Deep Learning, Aged, Algorithms, Arm, Female, Humans, Human Activities, Male, Middle Aged, Movement, Cluster Analysis, Support Vector Machine
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
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.