Data augmentation of wearable sensor data for parkinson’s disease monitoring using convolutional neural networks
Open accessThe DOI matches its Crossref record; 13 of the 13 references it cites were found in a registry or an index.
Abstract
While convolutional neural networks (CNNs) have been successfully applied to many challenging classification applications, they typically require large datasets for training. When the availability of labeled data is limited, data augmentation is a critical preprocessing step for CNNs. However, data augmentation for wearable sensor data has not been deeply investigated yet.
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.
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?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“We apply the jittering method to simulate additive sensor noise using random parameters as described in [47] for label-preserving augmentation.”
From Sensor Data Acquisition and Multimodal Sensor Fusion for Human Activity Recognition Using Deep Learning (2019) · cites this paper as [47]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“We considered three self-supervised tasks from ref. 27, which were first used in ref. 34 as data augmentation techniques.”
From Self-supervised learning for human activity recognition using 700,000 person-days of wearable data (2024) · cites this paper as [34]
Experimental — accuracy not yet measuredThe cited source says what the claim attributes to it.
In the citing paper
“Um et al. [193] tackle the scarcity of labeled motor states in Parkinson's disease patients by introducing a variety of augmentation techniques, including jittering, scaling, and rotation.”
From Past, Present, and Future of Sensor-based Human Activity Recognition Using Wearables: A Surveying Tutorial on a Still Challenging Task (2025) · cites this paper as [193]
The cited source says what the claim attributes to it.
In the citing paper
“In sensor-based HAR leveraging the temporal nature of IMU signals allows for the application of augmentation methods grounded in signal processing [32,34,99,132,141,178,193].”
From Past, Present, and Future of Sensor-based Human Activity Recognition Using Wearables: A Surveying Tutorial on a Still Challenging Task (2025) · cites this paper as [193]
Experimental — accuracy not yet measuredThe cited source says what the claim attributes to it.
In the citing paper
“The choice of augmentations is key to performance, and in this tutorial, we utilize all pairwise combinations of augmentations introduced by Um et al. [193].”
From Past, Present, and Future of Sensor-based Human Activity Recognition Using Wearables: A Surveying Tutorial on a Still Challenging Task (2025) · cites this paper as [193]
7 more claims about this paper
Create a free account to see every claim other papers make about this one.
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
Hochreiter, S., & Schmidhuber, J. (1997) · Neural Computation · doi:10.1162/neco.1997.9.8.1735 · Cited by 102,318 · cited 1×
“A reason for employing a CNN rather than a Long Short-term Memory (LSTM) [13], which is often used for learning time-series data, is to reduce the number of parameters to work with the small-scale PD data.”
MethodLeCun, Y., Bengio, Y., & Hinton, G. (2015) · Nature · doi:10.1038/nature14539 · Cited by 84,126
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2015) · arXiv · doi:10.48550/arxiv.1512.03385 · Cited by 4,196 · cited 2×
“…(CNNs) have shown excellent performance on classification problems when large-scale labeled datasets are available. successes of CNNs include static image classification (e.g. [17], [12]), and extend to other nonstatic domains, for example, video [15], speech [26], and sentence [16] classification.”
BackgroundReturn of the Devil in the Details: Delving Deep into Convolutional Nets
Chatfield, K., Simonyan, K., Vedaldi, A., & Zisserman, A. (2014) · Proceedings of the British Machine Vision Conference 2014 · doi:10.5244/c.28.6 · Cited by 3,335
Return of the Devil in the Details: Delving Deep into Convolutional NetsPreprint
Chatfield, K., Simonyan, K., Vedaldi, A., et al. (2014) · arXiv (Cornell University) · doi:10.48550/arxiv.1405.3531 · Cited by 663 · cited 1×
“Because of its practical importance, data augmentation has been investigated with systematic experiments, e.g., for image [2] and speech recognition [4].”
BackgroundMonitoring Motor Fluctuations in Patients With Parkinson's Disease Using Wearable Sensors
Patel, S., Lorincz, K., Hughes, R., Huggins, N., Growdon, J., Standaert, D., Akay, M., Dy, J., Welsh, M., & Bonato, P. (2009) · IEEE Transactions on Information Technology in Biomedicine · doi:10.1109/titb.2009.2033471 · Cited by 623 · cited 3×
“Several researchers have attempted to automate PD motor state classification using machine learning techniques based on wearable sensor data (e.g. [25], [10], [7]).”
BackgroundMethodClinical Approach to Parkinson's Disease: Features, Diagnosis, and Principles of Management
Massano, J., & Bhatia, K. P. (2012) · Cold Spring Harbor Perspectives in Medicine · doi:10.1101/cshperspect.a008870 · Cited by 422 · cited 2×
“…accurate evaluation of a patient’s phenomenology is needed for determining the right dose of medication to alleviate the symptoms of bradykinesia and prevent overtreatment-induced dyskinesia. current evaluation process relies on patient’s self-reports and visual observation by the clinician [22].”
BackgroundData Augmentation for Deep Neural Network Acoustic Modeling
Cui, X., Goel, V., & Kingsbury, B. (2015) · IEEE/ACM Transactions on Audio, Speech, and Language Processing · doi:10.1109/taslp.2015.2438544 · Cited by 376
Free‐living monitoring of Parkinson's disease: Lessons from the field
Del Din, S., Godfrey, A., Mazzà, C., Lord, S., & Rochester, L. (2016) · Movement Disorders · doi:10.1002/mds.26718 · Cited by 370 · cited 1×
“…for a large-scale PD monitoring [18]. ability to assess motor state from free-living conditions has been identified as an important need for improved assessment and treatment; but to date there is no fully validated system to monitor clinical features or activities in free-living environments [6].”
BackgroundLin, M., Chen, Q., & Yan, S. (2013) · arXiv · doi:10.48550/arxiv.1312.4400 · Cited by 252 · cited 1×
“Fully connected layers which are frequently-used in CNN architectures for classification are substituted with a global averaging pooling (GAP) layer [21] to reduce the number of parameters.”
Method
Research works printed but not linked: 8
- [6] Ian Goodfellow, Yoshua Bengio, and Aaron Courville. 2016. Deep Learning . MIT Press. -
- [7] Robert I. Griffiths, Katya Kotschet, Sian Arfon, Zheng Ming Xu, William Johnson, John Drago, Andrew Evans, Peter Kempster, Sanjay Raghav, and Malcolm K. Horne. 2012. Automated assessment of bradykinesia and dyskinesia in Parkinson's disease. Journal of Parkinson's disease 2, 1 (2012), 47-55. -
- [11] Sergey Ioffe and Christian Szegedy. 2015. Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift. ArXiv e-prints (Feb. 2015). arXiv:cs.LG/1502.03167 -
- [12] Alex Krizhevsky, Ilya Sutskever, and Geoffrey E. Hinton. 2012. ImageNet Classification with Deep Convolutional Neural Networks. In Advances in Neural Information Processing Systems 25 , F. Pereira, C. J. C. Burges, L. Bottou, and K. Q. Weinberger (Eds.). Curran Associates, Inc., 1097-1105. -
- [14] Arthur Le Guennec, Simon Malinowski, and Romain Tavenard. 2016. Data Augmentation for Time Series Classification using Convolutional Neural Networks. In ECML/PKDD Workshop on Advanced Analytics and Learning on Temporal Data . Riva Del Garda, Italy. https://halshs.archives-ouvertes.fr/halshs-01357973 -
- [15] Yann LeCun, Yoshua Bengio, and Geoffrey Hinton. 2015. Deep learning. Nature 521, 7553 (28 May 2015), 436-444. Insight. -
- [19] Shyamal Patel, Konrad Lorincz, Richard Hughes, Nancy Huggins, John Growdon, David Standaert, Metin Akay, Jennifer Dy, Matt Welsh, and Paolo Bonato. 2009. Monitoring Motor Fluctuations in Patients With Parkinson's Disease Using Wearable Sensors. IEEE Transactions on Information Technology in Biomedicine 13, 6 (Nov 2009), 864-873. -
- [20] Terry Taewoong Um, Vahid Babakeshizadeh, and Dana Kulić. 2017. Exercise Motion Classification from Large-Scale Wearable Sensor Data Using Convolutional Neural Networks. In 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) . 1-6.
Cited by
Items 301–400 of 736
Zhou, B., Feng, N., Wang, H., Lu, Y., Wei, C., Jiang, D., & Li, Z. (2022) · Journal of Neural Engineering · doi:10.1088/1741-2552/ac89b4 · Cited by 11
Davidashvilly, S., Hssayeni, M. D., Chi, C., Shahed, J. J., & Ghoraani, B. (2022) · Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) · doi:10.1109/embc48229.2022.9871181 · Cited by 14
“Um et al. [1] utilized the same augmentation techniques for detecting medication states in PwPD.”
BackgroundMethodInfluentialA Study on the Prediction of Cancer Using Whole-Genome Data and Deep Learning
Lee, Y.-J., Park, J., & Lee, S.-H. (2022) · International Journal of Molecular Sciences · doi:10.3390/ijms231810396 · Cited by 12
“[21] in 2017 measured the accuracy of hand tremors by applying wearable data sensors on the hands of people with Parkinson’s disease.”
Clustering of Human Activities from Wearables by Adopting Nearest Neighbors
Ahmed, A., Haresamudram, H., & Plötz, T. (2022) · ACM International Symposium on Wearable Computers (ISWC) · doi:10.1145/3544794.3558477 · Cited by 14
“Data augmentation transforms from [30, 31] are used in accordance with contrastive training.”
MethodWei, S., Sun, Z., Wang, Z., Liao, F., Li, Z., & Mi, H. (2023) · Applied Sciences · doi:10.3390/app13053177 · Cited by 17
“…relies on domain knowledge, as modulated signal data are a type of time series data, unlike electrocardiogram signal data in the medical field [42], which are not used to distinguish an event in a regular signal sequence but prefers a pattern of signal sample points over a period of time,…”
BackgroundGesture Correctness Estimation with Deep Neural Networks and Rough Path Descriptors
Sadawi, N., Miron, A., Ismail, W., Hussain, H., & Groşan, C. (2019) · IEEE ... International Conference on Data Mining workshops · doi:10.1109/icdmw.2019.00090 · Cited by 11
“While existing approaches employed GAN networks (as mentioned in section II), we have used some of the techniques reported in [18].”
MethodResultSmartGe: Identifying Pen-Holding Gesture With Smartwatch
Bi, H., Zhang, J., & Chen, Y. (2020) · IEEE Access · doi:10.1109/access.2020.2967770 · Cited by 11
“Therefore, we transform the data using time-wraping, magnitude-warping and scaling respectively and add these new datasets to our sample for enhancing the robustness of our model and ensuring a good generalization ability [19].”
MethodXiang, L., Gu, Y., Wang, A., Shim, V., Gao, Z., & Fernandez, J. (2023) · Journal of Human Kinetics · doi:10.5114/jhk/163059 · Cited by 11
“Data augmentation leverages limited data by transforming the original samples to create new ones (Um et al., 2017).”
BackgroundSun, Y., Kargarandehkordi, A., Slade, C. R., Jaiswal, A., Busch, G., Guerrero, A. P. S., Phillips, K. T., & Washington, P. Y. (2023) · JMIR Research Protocols · doi:10.2196/46493 · Cited by 14
Wang, H., Zhang, X., Long, H., Yao, S., & Zhu, P. (2024) · Neural Processing Letters · doi:10.1007/s11063-024-11478-3 · Cited by 23
“In response to the above problems, research has focused on four schemes: (1) Time-domain to frequency-domain conversion, (2) Unbalanced class enhancement, (3) Gaussian processing methods, and (4) A deeper neural network enhancement model method [57].”
VAE-LSTM Data Augmentation for Cattle Behavior Classification Using a Wearable Inertial Sensor
Dang, T.-H., Park, J., Tran, T. V., & Chung, W. (2024) · IEEE Sensors Letters · doi:10.1109/lsens.2024.3385418 · Cited by 11
“Um et al. [8], for instance, used rotation (ROT), scaling, jittering (JIT), permutation, slicing, time warping, and magnitude warping to improve wearable sensor data for deep temporal CNNs. Rashid and Louis [9] recently integrated recurrent neural network (RNN) with ROT, scaling, and time warping…”
Siargkas, C., Papapanagiotou, V., & Delopoulos, A. N. (2024) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2024.3387834 · Cited by 11
“Typically, accelerometer signals are augmented using signal processing transformations such as jittering, scaling, rotation, and permutation [47], [48].”
MethodNouman, M., Khoo, S., Mahmud, M. A. P., & Kouzani, A. Z. (2024) · Journal of Ambient Intelligence and Humanized Computing · doi:10.1007/s12652-024-04827-6 · Cited by 12
“The proposed radar-based model uses a dataset collected with the XeThru X4M03 UWB sensor from Novelda (XeThru X4M03 Development Kit—SensorLogic 2021 ) applied to the training dataset, and further details can be seen in Piriyajitakonkij et al. ( 2020 ) and Um et al. 2017.”
Shuqair, M., Shahed, J. J., & Ghoraani, B. (2024) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2024.3423708 · Cited by 11
Wang, H., Ullah, Z., Gazit, E., Brozgol, M., Tan, T., Hausdorff, J. M., Shull, P. B., & Ponger, P. (2024) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2024.3470310 · Cited by 11
“However, applying these methods to wearable sensor data poses a challenge due to the difficulty in preserving accurate labels [46], [47].”
BackgroundMethodInfluentialAccurate and Efficient Human Activity Recognition Through Semi-Supervised Deep Learning
Long, X., Wang, P., Gu, F., Chen, C., & Guo, S. (2025) · IEEE Sensors Journal · doi:10.1109/jsen.2025.3561418 · Cited by 11
“Finally, scaling, which is similar to the scaling used in weak augmentation, also increases sample diversity by adjusting the scale of the data [27].”
Enhancing Wearable Fall Detection System via Synthetic Data
Debnath, M., Alamgeer, S., Kabir, M. S., & Ngu, A. H. H. (2025) · Sensors · doi:10.3390/s25154639 · Cited by 13
“Various augmentation techniques have been employed to generate synthetic time-series data to enhance the training and performance of machine learning models [21].”
Zheng, W., & Arakawa, Y. (2025) · IEEE Access · doi:10.1109/access.2025.3598088 · Cited by 13
“For instance, Um et al. [103] introduced techniques such as time-warping, magnitude-warping, jittering, and cropping to generate acceleration data for monitoring Parkinson’s disease.”
White Noise Windows: Data Augmentation for Time Series
Goubeaud, M., Jousen, P., Gmyrek, N., Ghorban, F., & Kummert, A. (2021) · International Conference on Optimization and Applications (ICOA) · doi:10.1109/icoa51614.2021.9442656 · Cited by 17
“Different straight-forward methods to augment time series datasets such as scaling, jittering, and rotating are presented in [4].”
MethodPersonalized and Nonparametric Framework for Detecting Changes in Gait Cycles
Lamooki, S. R., Kang, J., Cavuoto, L. A., Megahed, F. M., & Jones‐Farmer, L. A. (2021) · IEEE Sensors Journal · doi:10.1109/jsen.2021.3090985 · Cited by 10
Jeon, H., & Lee, D. (2021) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2021.3103648 · Cited by 12
Deep Semi-supervised Learning for Time-Series Classification
Goschenhofer, J. (2022) · Advances in intelligent systems and computing · doi:10.1007/978-981-19-6153-3_15 · Cited by 12
A Study on Diffusion Modelling For Sensor-based Human Activity Recognition
Shao, S., & Sánchez, V. (2023) · International Workshop on Biometrics and Forensics (IWBF) · doi:10.1109/iwbf57495.2023.10157482 · Cited by 20
“Data Transformation Data Scaling [15] 0.”
MethodMeng, Q., & Zhu, S. (2022) · Journal of Construction Engineering and Management · doi:10.1061/(asce)co.1943-7862.0002359 · Cited by 12
Sikandar, T., Rabbi, M. F., Ghazali, K. H., Altwijri, O., Almijalli, M., & Ahamed, N. U. (2022) · Physical and Engineering Sciences in Medicine · doi:10.1007/s13246-022-01195-3 · Cited by 10
How Much Unlabeled Data is Really Needed for Effective Self-Supervised Human Activity Recognition?
Dhekane, S. G., Haresamudram, H., Thukral, M., & Plötz, T. (2023) · Proceedings of the 2023 International Symposium on Wearable Computers · doi:10.1145/3594738.3611366 · Cited by 9
“SimCLR , which contrasts augmented versions [35] of input windows against other pairs [4, 34].”
Skaramagkas, V., Boura, I., Karamanis, G., Kyprakis, I., Fotiadis, D. I., Kefalopoulou, Z. M., Spanaki, C., & Tsiknakis, M. (2025) · npj Digital Medicine · doi:10.1038/s41746-025-01630-1 · Cited by 9
“Although high performance was achieved with previous research 16,44 , utilization of facial videos offers competitive advantages in terms of accessibility and feasibility for both clinical and real-world settings over wearable sensors, which are often numerous, intricate and potentially…”
Donié, C., Das, N., Endo, S., & Hirche, S. (2025) · Scientific Reports · doi:10.1038/s41598-025-04263-2 · Cited by 10
“All high-performing related work with raw data either used semi-supervised learning 57 or data augmentation with domain-specific knowledge 48,66 .”
Self-citationArtificial Intelligence-Enhanced Flexible Sensors for Human Motion and Posture Sensing
Jiang, Y., & He, T. (2026) · Sensors · doi:10.3390/s26051562 · Cited by 10
“Under such conditions, data augmentation and synthetic data generation can serve as complementary strategies, introducing additional variability to better approximate the potential test distribution [94,95].”
Mehrang, S., Tadi, M. J., Knuutila, T., Jaakkola, J., Jaakkola, S., Kiviniemi, T., Vasankari, T., Airaksinen, J. K. E., Koivisto, T., & Pänkäälä, M. (2022) · Physiological Measurement · doi:10.1088/1361-6579/ac66ba · Cited by 14
“Specifically,wedeployed segmentationwithoverlap (Tadi et al2018,Mehrang et al2019) and rotation augmentation technique (Um et al2017)which is suitable for thedata that hold a geometric description.”
MethodChoi, A., Park, E., Kim, T. H., Im, G. J., & Mun, J. H. (2022) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2022.3186436 · Cited by 11
“Various previous studies reported that a sufficient amount of data is essential to prevent overfitting of multi-layered deep learning models and achieve adequate training performance [38].”
BackgroundA bidirectional trajectory contrastive learning model for driving intention prediction
Zhou, Y. F., Wang, H., Ning, N., Wang, Z., Zhang, Y., & Liu, F. (2023) · Complex & Intelligent Systems · doi:10.1007/s40747-022-00945-w · Cited by 10
“[33]designedcontrastivelearningmeth-ods of biological signals, such as EEG and ECG.”
BackgroundCaramaschi, S., Papini, G. B., & Caiani, E. G. (2023) · Applied Sciences · doi:10.3390/app13074175 · Cited by 11
On the Advantages of Hand Gesture Recognition with Data Gloves for Gaming Applications
Schade, A., Schulz, J., Nguyen, V., Scheunert, C., Bodenstedt, S., Nguyen, G. T. K., Speidel, S., & Fitzek, F. H. P. (2023) · IEEE International Conference on Pervasive Computing and Communications Workshops (PerCom Workshops) · doi:10.1109/percomworkshops56833.2023.10150283 · Cited by 9
“One common approach is to modify already existing and labeled data, e.g., by adding noise or by warping time series in duration, so-called data augmentation [5], resulting in more available training data.”
BackgroundSikandar, T., Rabbi, M. F., Ghazali, K. H., Altwijri, O., Almijalli, M., & Ahamed, N. U. (2023) · Scientific Reports · doi:10.1038/s41598-023-43428-9 · Cited by 11
Oh, Y., Choi, S.-A., Shin, Y., Jeong, Y., Lim, J., & Kim, S. (2023) · Sensors · doi:10.3390/s24010210 · Cited by 20
“In addition, the e ff ect of movement types (UNI, BIA, and BIS) in the ADL task on F1-scores across the training groups was studied via linear regression, where the factors were movement types (BIA, UNI, and BIS) and training groups (ND and Stroke).”
MethodInfluentialAugmentation Techniques for Sequential Clinical Data to Improve Deep Learning Prediction Techniques
Florez, A. Y. C., Scabora, L. C., Amer-Yahia, S., & Junior, J. F. R. (2020) · Proceedings - IEEE Symposium on Computer-Based Medical Systems · doi:10.1109/cbms49503.2020.00118 · Cited by 12
“…the realm of Medical clinical data, data augmentation techniques include works to supplement cancer outcome-adaptive clinical trials [21], works to increase the performance of alcoholism detection algorithms based on brain images [22], and works to assist in the Parkinson’s disease evaluation [23].”
MethodEmpirical Mode Decomposition Based Data Augmentation for Time Series Prediction Using NARX Network
Abayomi‐Alli, O., Sidekerskienė, T., Damaševičius, R., Siłka, J., & Połap, D. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-61401-0_65 · Cited by 9
Classifying In-Place Gestures with End-to-End Point Cloud Learning
Zhao, L., Lu, X., Zhao, M., & Wang, M. (2021) · Proceedings - International Symposium on Mixed and Augmented Reality, ISMAR · doi:10.1109/ismar52148.2021.00038 · Cited by 9
Liu, D., Wu, Y., Hong, D., & Wang, S. (2022) · Concurrency and Computation Practice and Experience · doi:10.1002/cpe.7331 · Cited by 14
Exploring Multimodal Fusion for Continuous Protective Behavior Detection
Cen, G., Wang, C., Olugbade, T., Williams, A. C. D. C., & Bianchi‐Berthouze, N. (2022) · International Conference on Affective Computing and Intelligent Interaction and workshops · doi:10.1109/acii55700.2022.9953851 · Cited by 11
“To manage the small size (7629 frames of which 5231 from people with chronic pain, in which 4219 are not PB and 1012 are PB) of the EmoPain dataset, jittering and cropping were used as data augmentation [30].”
Smartwatch Sensors with Deep Learning to Predict the Purchase Intentions of Online Shoppers
Chang, R.-I., Tsai, C.-Y., & Chung, P. (2022) · Sensors · doi:10.3390/s23010430 · Cited by 10
SensiX: A System for Best-effort Inference of Machine Learning Models in Multi-device Environments
Min, C., Mathur, A., Montanari, A., & Kawsar, F. (2022) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2022.3173914 · Cited by 10
“One research thread is to retrain and re-engineer ML models by employing ML techniques such as data augmentation [9], domain adaptation, and transfer learning [10].”
BackgroundSemi-Supervised Federated Analytics for Heterogeneous Household Characteristics Identification
Chen, W., Bu, S., Zhang, X., Tao, Y., Zhang, Y., & Han, Z. (2024) · IEEE Transactions on Smart Grid · doi:10.1109/tsg.2024.3415504 · Cited by 10
“For weak data augmentation, we utilize jittering [62] to x ur , i which adds Gaussian noise to the original data.”
MethodBidirectional consistency with temporal-aware for semi-supervised time series classification
Liu, H., Zhang, F., Huang, X., Wang, R., & Xi, L. (2024) · Neural Networks · doi:10.1016/j.neunet.2024.106709 · Cited by 9
Pal, D., Mukhopadhyay, S., & Gupta, R. (2024) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2024.3460887 · Cited by 10
“Um et al. [19] describes three simple meth-ods, viz., magnitude-warping, time-warping, and cropping for wearable sensors dataset toward Parkinson’s disease detection.”
BackgroundOpenFi: Open-Set WiFi Human Sensing via Virtual Embedding Confidence-Aware
Liu, H., Xi, L., Wei, W., Zhang, F., & Haas, Z. J. (2024) · IEEE Transactions on Network Science and Engineering · doi:10.1109/tnse.2024.3496496 · Cited by 9
MobHAR: Source-free Knowledge Transfer for Human Activity Recognition on Mobile Devices
Xue, M., Zhu, Y., Xie, W., Wang, Z., Chen, Y., Jiang, K., & Zhang, Q. (2025) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3712620 · Cited by 14
“Therefore, as shown in Figure 2, for each unlabeled data, we first augment unlabeled data using six kinds of strategies [6, 35, 40, 46, 47, 51, 55, 56], including normalization, rotation, wrapping, permutation, adversarial examples, and jittering.”
TFC: Time–frequency contrasting network for wearable-based human activity recognition
Huang, Z., Deng, J., Wang, S., Tang, C., & Xiao, S. (2025) · Knowledge-Based Systems · doi:10.1016/j.knosys.2025.113373 · Cited by 9
Arunan, A., Qin, Y., Li, X., & Yuen, C. (2025) · Expert Systems with Applications · doi:10.1016/j.eswa.2025.127889 · Cited by 8
“For the remaining techniques, we follow Um et al. [56]’s recommendations.”
Li, Y., Wang, T., Xie, J., Yang, J. N., Pan, T., Niu, B., & Yu, X. (2025) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2025.111537 · Cited by 8
Exploring data augmentation methods to enhance EEG measures for epilepsy seizure detection
Guo, Y., Zhang, X., & Dai, C. (2025) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2025.110512 · Cited by 8
Arce-Sáenz, L. A., Izquierdo-Reyes, J., & Bustamante-Bello, R. (2025) · Scientific Reports · doi:10.1038/s41598-025-10573-2 · Cited by 8
“The data augmentation process, inspired by the methods described by Um et al. 35 , involved applying time-series transformations to generate synthetic samples.”
Qi, L., Li, F., Shang, J., Ge, D., Wang, S., & Yuan, S. (2025) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2025.3605810 · Cited by 9
“Permutation-basedaugmentation[31] canalsodestroysequentialdependencies,weakeningtempo-ral dynamics.”
A Machine Learning Approach for the Automatic Estimation of Fixation-Time Data Signals’ Quality
Gabrieli, G., Balagtas, J. P. M., Esposito, G., & Setoh, P. (2020) · Sensors · doi:10.3390/s20236775 · Cited by 7
“The technique has been shown to be suitable for ML analysis and classification of different type of signals, including neurophysiological signals [27,46], and it has been successfully employed to enhance the accuracy of ML classifiers [39,47].”
MethodFew-Shot and Weakly Supervised Repetition Counting With Body-Worn Accelerometers
Nishino, Y., Maekawa, T., & Hara, T. (2022) · Frontiers in Computer Science · doi:10.3389/fcomp.2022.925108 · Cited by 8
“For each subsegment, we perform the following operations according to (Morris et al., 2017; Yu et al., 2017): • Scaling: We change the scale of the sensor data magnitude by multiplying a random value sampled from the uniform distribution U (1 − c s , 1 + c s ) for each subsegment.”
MethodCROMOSim: A Deep Learning-Based Cross-Modality Inertial Measurement Simulator
Hao, Y., Lou, X., Wang, B., & Zheng, R. (2022) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2022.3230370 · Cited by 14
“In [15], the authors proposed a systematic way to augment the IMU data via rotation, permutation, time-warping, scaling, magnitude-warping and jittering.”
BackgroundTN-GAN-Based Pet Behavior Prediction through Multiple-Dimension Time-Series Augmentation
Kim, H., & Moon, N. (2023) · Sensors · doi:10.3390/s23084157 · Cited by 8
“A machine learning algorithm has been used to achieve human activity recognition [22,23].”
MethodWang, T., Li, J., Wang, R., Kara, D., Liu, S., Wertheimer, D., Martin, A. V., Ganti, R., Srivatsa, M., & Abdelzaher, T. (2023) · ACM Conference on Embedded Networked Sensor Systems (SenSys) · doi:10.1145/3625687.3625785 · Cited by 13
“Traditional time-series data augmentation carried out in the time domain includes scaling, jittering, rotating, permuting, noise injection, and others [8, 18, 30, 30, 36].”
BackgroundPerformance Analysis of Data Augmentation Approaches for Improving Wrist-Based Fall Detection System
Tu, Y.-C., Lin, C.-H., Liu, C.-P., & Chan, C.-T. (2025) · Sensors · doi:10.3390/s25072168 · Cited by 9
“By tuning the physical properties of IMU signals, including rotation, permutation, time-warping, magnitude-warping, jittering, and scaling [30], data transformation can synthetically generate a large quantity of training data, expecting to solve the class imbalance issue, as shown in Figure 1.”
Deep Learning: Predicting Environments From Short-Time Observations of Postural Balance
Phan, V., & Lee, H. (2022) · IEEE Transactions on Biomedical Engineering · doi:10.1109/tbme.2022.3170850 · Cited by 8
Ma, W., Song, G., Zeng, Q., Zhang, H., Zou, M., & Zhao, Z. (2024) · IEEE Sensors Journal · doi:10.1109/jsen.2024.3371588 · Cited by 9
Data Augmentation-Based Enhancement for Efficient Network Traffic Classification
Shin, C.-Y., Choi, Y., & Kim, M. (2024) · IEEE Access · doi:10.1109/access.2024.3525000 · Cited by 12
“Jittering, a classic but very simple and effective method of transformation-based data augmentation [72], [73], was applied to increase the robustness of individual data values by varying the amount of variation in the range of each value. p 1 denotes the first single packet which is extracted from…”
Leafeon: Toward Accurate Sensing of Leaf Water Content for Protected Cropping With mmWave Radar
Cardamis, M., Jia, H. Y., Qian, H., Chen, W., Yan, Y., Ghannoum, O., Quigley, A. J., Chou, C. T., & Hu, W. S. (2025) · IEEE Internet of Things Journal · doi:10.1109/jiot.2025.3541976 · Cited by 7
“Jittering, a type of noise that simulates additive sensor noise [48], is challenging for robustness analysis and thus is considered in our experiment.”
A Lightweight CNN for Real-Time Pre-Impact Fall Detection
Turetta, C., Ali, M. T., Demrozi, F., & Pravadelli, G. (2025) · Design, Automation, and Test in Europe · doi:10.23919/date64628.2025.10993022 · Cited by 10
“The time warping stretches and compresses the signal, making the model more robust to the sampling of the signal [16], while window warping warps a randomly selected falling time series segment by speeding it up or down [17].”
Tasdemir, A. T., Ozkat, E. C., Yalçın, G., & Gül, F. (2026) · Sensors · doi:10.3390/s26123877 · Cited by 8
“The augmentations were Gaussian jitter ( σ = 0.03), per-channel random scaling ( σ = 0.1), magnitude warping (4 control knots, σ = 0.2) [48], temporal permutation, and 90%-crop-resize.”
InfluentialThe Time-Sequence Prediction via Temporal and Contextual Contrastive Representation Learning
Liu, Y., & Liu, J. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-20862-1_34 · Cited by 7
Exploring the Benefits of Time Series Data Augmentation for Wearable Human Activity Recognition.
Hasan, M. A., Li, F., Piet, A., Gouverneur, P., Irshad, M. T., & Grzegorzek, M. (2023) · International Workshop on Sensor-based Activity Recognition and Interaction (iWOAR) · doi:10.1145/3615834.3615842 · Cited by 10
“The majority of TSDA techniques, including for example, jittering [2, 32], slicing or cropping [20], time warping [9], and convolve or frequency warping [29] is part of the RT family since they involve applying random transformations to the training data like for image data augmentation.”
BackgroundStröbel, R., Mau, M., Puchta, A., & Fleischer, J. (2024) · Machine Learning and Knowledge Extraction · doi:10.3390/make6020049 · Cited by 12
“The implementation for this method followed [37].”
Shuqair, M., Shahed, J. J., & Ghoraani, B. (2024) · Bioengineering · doi:10.3390/bioengineering11070689 · Cited by 10
“We generated three signal transformations [36]: ) .”
BackgroundMethodA Time Series Data Augmentation Method based on SMOTE
Qu, H., & Zhang, Z. (2024) · Chinese Control and Decision Conference (CCDC) · doi:10.1109/ccdc62350.2024.10587816 · Cited by 8
“Jittering (Jit), which involves adding random noise from a Gaussian distribution with mean μ=0 and standard deviation σ=0.03 to the original time series for data augmentation [15].”
BackgroundInfluentialWang, T., Chen, Y., Yang, Q., Sun, D., Wang, R., Li, J., Kimura, T., & Abdelzaher, T. (2024) · International Conference on Computer Communications and Networks (ICCCN) · doi:10.1109/icccn61486.2024.10637566 · Cited by 11
“…approach simply ignores the discrepancy between the data distribution between the training set and 2) Conventional augmentation (convn-aug) : Conventional time-series augmentation methods proposed in [16], including jittering, scaling, rotation, permutation, magnitude-warping, and time-warping.”
BackgroundMethodEmotion Recognition on the Go: Utilizing Wearable IMUs for Personalized Emotion Recognition
Leng, Z., Jung, M., Hwang, S., Oh, S., Zhang, L., Plötz, T., & Kim, K. K. (2024) · ACM International Joint Conference on Pervasive and Ubiquitous Computing (UbiComp) · doi:10.1145/3675094.3678452 · Cited by 10
“To amass more data, future studies could implement data augmentation techniques such as rotations, permutations, time-warping, and scaling [37], or generate additional data through motion synthesis methods [45, 46].”
Huang, Y., Song, Y., & Cai, Z. (2024) · Reliability Engineering & System Safety · doi:10.1016/j.ress.2024.110716 · Cited by 8
Augmentation of Segmented Motion Capture Data for Improving Generalization of Deep Neural Networks
Sawicki, A., & Zieliński, S. K. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-47679-3_24 · Cited by 7
A Well-Overflow Prediction Algorithm Based on Semi-Supervised Learning
Liu, W., Fu, J., Liang, Y., Cao, M., & Han, X. (2022) · Energies · doi:10.3390/en15124324 · Cited by 10
“Terry et al., use methods such as Jitter, Scale, MagWarp, and TimeWarp to enhance wearable sensor data; appropriate enhancement can improve classification performance from 77.54% to 86.88% [19].”
MethodInfluentialA Novel Data Augmentation Method for Improving the Accuracy of Insulator Health Diagnosis
Li, Z., Song, Y., Li, R., Gu, S., & Fan, X. (2022) · Sensors · doi:10.3390/s22218187 · Cited by 7
“Other techniques involve window slicing [22] and time warping [10] which are implemented by changing the time steps of the sequence.”
BackgroundMethodInfluentialExplainable Detection of Microplastics Using Transformer Neural Networks
Barker, M., Willans, M., Pham, D.-S., Krishna, A., & Hackett, M. J. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-22695-3_8 · Cited by 7
Savalia, J., Desai, S., Geddam, R., Shah, P., & Chhikaniwala, H. (2022) · Communications in computer and information science · doi:10.1007/978-3-031-23092-9_26 · Cited by 8
Moshkova, A. A., Samorodov, A. V., Ivanova, E., & Fedotova, E. Y. (2020) · Ural Symposium on Biomedical Engineering, Radioelectronics and Information Technology (USBEREIT) · doi:10.1109/usbereit48449.2020.9117736 · Cited by 7
“However, CNN is more suitable for small datasets, because CNN generally use a smaller number of parameters then LSTM [5].”
BackgroundMethodTime Series Modeling for Phenotypic Prediction and Phenotype-Genotype Mapping Using Neural Networks
Choudhury, S. D. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-65414-6_17 · Cited by 9
The Effects of Data Augmentation Methods on the Performance of Human Activity Recognition
Bakhshayesh, P. R., Ejtehadi, M., Taheri, A., & Behzadipour, S. (2022) · Iranian Conference on Signal Processing and Intelligent Systems (ICSPIS) · doi:10.1109/icspis56952.2022.10043959 · Cited by 7
“…there exist a few time-series data augmentation methods such as jittering, scaling, time warping, amplitude warping, encoder-decoder, linear combination, etc. that are mostly categorized into three classes, namely: random transformation, pattern mixing and generative models [5], [10], [16], [17].”
BackgroundMethodInfluentialImproving Deep Learning Parkinson’s Disease Detection Through Data Augmentation Training
Taleb, C., Likforman-Sulem, L., & Mokbel, C. (2019) · Communications in computer and information science · doi:10.1007/978-3-030-37548-5_7 · Cited by 7
Differentiable Prior-Driven Data Augmentation for Sensor-Based Human Activity Recognition
Zhang, Y., Gao, Q. H., Hu, R., Ding, Q., Li, B., & Guo, Y. (2025) · IEEE Transactions on Computational Social Systems · doi:10.1109/tcss.2025.3565414 · Cited by 6
“These operations comprise Timewarp [17], Timenoise [17], Phaseperturb [16], and Timecrop [16].”
InfluentialStefanou, E. E., Bitilis, P., Bouchouras, G., & Kotis, K. (2025) · Applied Sciences · doi:10.3390/app151910541 · Cited by 7
Khaked, A. A., Oishi, N., Roggen, D., & Lago, P. (2023) · ACM International Joint Conference on Pervasive and Ubiquitous Computing (UbiComp) · doi:10.1145/3594739.3610742 · Cited by 9
Two-Stage Convolutional Neural Network for Classification of Movement Patterns in Tremor Patients
Weede, P., Smietana, P. D., Kuhlenbäumer, G., Deuschl, G., & Schmidt, G. U. (2024) · Information · doi:10.3390/info15040231 · Cited by 6
He, H., Huan, X., Wang, J., Luo, Y., Hu, H., & An, J. (2024) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2024.3459944 · Cited by 8
Bott, A., Liu, B., Puchta, A., & Fleischer, J. (2024) · Journal of Machine Engineering · doi:10.36897/jme/192681 · Cited by 8
“The subsequent data augmentation can enable coverage of the unexplored input space, avoid overfitting, and improve model generalization [39].”
BackgroundMethodSelf-Supervised Learning for Activity Recognition Based on Datasets With Imbalanced Classes
Hamad, R. A., Woo, W. L., Wei, B. W., & Yang, L. (2025) · IEEE Transactions on Emerging Topics in Computational Intelligence · doi:10.1109/tetci.2025.3526504 · Cited by 11
Kraft, D., Srinivasan, K., & Bieber, G. (2020) · International Conference on PErvasive Technologies Related to Assistive Environments (PETRA) · doi:10.1145/3389189.3397983 · Cited by 8
“88 % in the domain of Parkinson Disease classication in Alzheimer patients [44].”
BackgroundMiller, E., MacFarlane, Z. T., Martin, S. S., Banerjee, N., & Zhu, T. (2021) · Smart Health · doi:10.1016/j.smhl.2021.100245 · Cited by 6
ALAE-TAE-CutMix+: Beyond the State-of-the-Art for Human Activity Recognition Using Wearable Sensors
Ahmad, N., & Leung, H.-F. (2023) · IEEE International Conference on Pervasive Computing and Communications · doi:10.1109/percom56429.2023.10099138 · Cited by 13
ADL-GAN: Data Augmentation to Improve In-the-Wild ADL Recognition Using GANs
Ditthapron, A., Lammert, A. C., & Agu, E. (2022) · IEEE Access · doi:10.1109/access.2023.3271409 · Cited by 7
“For data augmentations of triaxial accelerometer and gyroscope features, we simply applied three transformations previously proposed for wearable sensor data [30]: rotation, permutation, and time-warping.”
MethodResultUse of Artificial Intelligence for Analyzing Kidney Stone Composition: Are We There Yet?
Isha, S., & Shah, S. Z. (2023) · Mayo Clinic Proceedings Digital Health · doi:10.1016/j.mcpdig.2023.06.007 · Cited by 6
“20,21 On the contrary, transfer learning uses the existing solution for a similar problem as a starting point and tries to solve a new classi fi - cation problem, thus requiring training data to obtain a robust solution.”
SimMix: Local similarity-aware data augmentation for time series
Liu, P., Guo, Y., Chen, P., Chen, Z., Wang, R., Wang, Y., & Shi, B. (2024) · Expert Systems with Applications · doi:10.1016/j.eswa.2024.124793 · Cited by 6
FPGA-Based Realtime Detection of Freezing of Gait of Parkinson Patients
Langer, P., Esfahani, A. H., Dyka, Z., & Langendörfer, P. (2022) · Lecture notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering · doi:10.1007/978-3-030-95593-9_9 · Cited by 7
A Multi-View learning approach to enhance automatic 12-Lead ECG diagnosis performance
Choi, J.-W., Hong, D.-Y., Jung, C., Hwang, E., Park, S.-H., & Roh, S. (2024) · Biomedical Signal Processing and Control · doi:10.1016/j.bspc.2024.106214 · Cited by 8
Vlachostergiou, A., Tagaris, A., Stafylopatis, A., & Kollias, S. (2018) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2018.8451276 · Cited by 6
“Another recent research approach is that of [6], where the motor state Parkinson’s disease prediction is treated as a CNN-based classification problem.”
BackgroundPacifici, I., Sernani, P., Falcionelli, N., Tomassini, S., & Dragoni, A. F. (2020) · Data in Brief · doi:10.1016/j.dib.2020.106455 · Cited by 6
“The data included in the dataset can be augmented with various techniques, including time-warping, scaling, jittering [8] .”
Metadata
- Authors
- Um, T. T., Pfister, F. M. J., Pichler, D., Endo, S., Lang, M., Hirche, S., Fietzek, U., & Kulić, D.
- Affiliations
- University of Waterloo (CA)
- Ludwig-Maximilians-Universität München (DE)
- Technical University of Munich (DE)
- Schön Klinik München Schwabing (DE)
- Venue
- Proceedings of the 19th ACM International Conference on Multimodal Interaction
- Publisher
- ACM
- Published
- 2017-11-03
- Pages
- 216-220
- OpenAlex
- W2620664872
- Cited by
- 729
- References
- 8
- Citations by year
- 2026: 80 · 2025: 131 · 2024: 128 · 2023: 112 · 2022: 101 · 2021: 77 · 2020: 64 · 2019: 26 · 2018: 10
- FWCI
- 7.99
- Type
- conference paper
- Language
- en
- Open access
- green
- arxiv.org
- Locations
- INTERNATIONAL CONFERENCE ON MULTIMODAL INTERACTION (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- mediaTUM (Technical University of Munich) (submittedVersion, cc-by)
- arXiv (Cornell University)
- Funders
- FP7 Ideas: European Research Council (337654)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Context-Aware Activity Recognition Systems
- Health Sciences › Health Professions › Health Information Management › Artificial Intelligence in Healthcare
- Life Sciences › Neuroscience › Cognitive Neuroscience › EEG and Brain-Computer Interfaces
- Keywords
- data augmentation, wearable sensor data, Parkinson's disease, disease surveillance, convolutional neural networks, limited labeled data
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.