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 501–600 of 736
Le, H. A. T., Luu, T. T., Dinh, D.-M., Rhee, K., Choi, K., Choi, D., & Park, Y. T. (2026) · Advanced Functional Materials · doi:10.1002/adfm.76403 · Cited by 3
Adaptive Shapelets Preservation for Time Series Augmentation
Pin, L., Guo, X., Chen, P., Shi, B., Wo, T., & Liu, X. (2022) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn55064.2022.9892768 · Cited by 4
“Existing works on time series augmentation mainly include the random transformations augmentation (e.g., jitter [11] and rotation [12] which perform geometric transformations on the time series) and the pattern mixing augmentation (e.g., DBA [13], [14] and DGW [15] which leverage Dynamic Time…”
BackgroundMethodInfluentialGu, B., & Sung, Y. (2023) · Sensors · doi:10.3390/s23031651 · Cited by 4
“improved CNN results by applying rotation, permutation, and time warping to 1D signal data [43].”
MethodMiss-placement Prediction of Multiple On-body Devices for Human Activity Recognition
Dönnebrink, R., Rueda, F. M., Grzeszick, R., & Stach, M. (2023) · International Workshop on Sensor-based Activity Recognition and Interaction (iWOAR) · doi:10.1145/3615834.3615838 · Cited by 2
“random sinusoidal curves are generated using a cubic spline to smoothly distort the time intervals between samples [20].”
MethodSwipe-it!: One-handed Thumb-tip Interaction for Text Editing in AR
Kim, J., Vanichvoranun, N., Jang, H., & Yoon, S. H. (2023) · IEEE International Symposium on Mixed and Augmented Reality Adjunct (ISMAR-Adjunct) · doi:10.1109/ismar-adjunct60411.2023.00127 · Cited by 2
“We use Magnitude warping suggested by [12], which is chaining the magnitude by convolving the data window with a smooth curve and with jittering as additive noise to 1,909 samples.”
MethodAdversarial Attack Driven Data Augmentation for Time Series Classification
Li, P., Hosseinzadeh, P., Bahri, O., Boubrahimi, S. F., & Hamdi, S. M. (2023) · International Conference on Machine Learning and Applications · doi:10.1109/icmla58977.2023.00096 · Cited by 3
“Transformations on time series can generally be categorized into three domains, the magnitude domain (jittering [13], rotation [14], scaling [13], and magnitude warping [14]), time domain (slicing [15], permutation [14], time warping [15]), and frequency domain (frequency warping [16], Fourier…”
BackgroundMethodInfluentialHuman Activity Recognition Using FixMatch-based Semi-supervised Learning with CSI
Teramoto, K., Haruyama, T., Shimoyama, T., Kato, F., & Mineno, H. (2024) · Journal of Information Processing · doi:10.2197/ipsjjip.32.596 · Cited by 3
BitLINK: Temporal Linkage of Address Clusters in Bitcoin Blockchain
Zhong, S., & Mueen, A. (2024) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/3637528.3672037 · Cited by 3
UWB/IMU-assisted Gesture Recognition Using Learning Approaches for VR/XR Applications
Oh, H. S., Bhattacharya, S., & Seo, S. (2024) · International Conference on Communications · doi:10.1109/icc51166.2024.10622800 · Cited by 2
“To tackle this, we augment our already collected data through adopting several augmentation techniques in [13] and [16].”
MethodThe SES framework and Frequency domain information fusion strategy for Human activity recognition
Wu, H., Feng, H., Shi, L., Zhang, H., & Xu, H. (2024) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn60899.2024.10650417 · Cited by 2
“In order to verify the stability and effect of the SES framework, we additionally conducted experiments by setting the most sensitive r in the framework to 16, 32, 128, 256 in the HHAR dataset, respectively.”
MethodTime-Series Representation Learning via Dual Reference Contrasting
Yu, R., Gong, Y., Wang, S., Si, J., Peng, X., Xu, B., & Lü, W. (2024) · ACM International Conference on Information and Knowledge Management (CIKM) · doi:10.1145/3627673.3679699 · Cited by 4
DeSmet, C., Greeley, C., & Cook, D. J. (2024) · IEEE Sensors Journal · doi:10.1109/jsen.2024.3483108 · Cited by 4
“Adding noise in the form of jitter is still popular [23], as is scaling by changing the magnitude of selected points in the series [24].”
Graph structure based data augmentation method
Kim, K. G., & Lee, B. T. (2024) · Biomedical Engineering Letters · doi:10.1007/s13534-024-00446-4 · Cited by 3
“Although performance gain achieved from normal augmentations comes from optimal parameter combination tuned by RandAugment, graph augmentation and normal augmentations show similar performance gain on average across every model and dataset.”
A Few-Shot Learning-Based Material Recognition Scheme Using Smartphones
Kim, Y., Yoon, J., & Kim, S. (2025) · Applied Sciences · doi:10.3390/app15010430 · Cited by 2
Lee, T. Y., Ha, S. H., & Lee�, H. (2024) · International Conference on Information and Communication Technology Convergence (ICTC) · doi:10.1109/ictc62082.2024.10827153 · Cited by 2
Yang, G., Luo, T.-J., & Zhang, X. (2025) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2025.110280 · Cited by 2
Ghosh, I., Chugh, G., Faridee, A. Z. M., & Roy, N. (2024) · Proceedings · doi:10.1109/icdm59182.2024.00079 · Cited by 3
“…fluctuations in the data, distinct from actual motion, thereby adding stochastic noise and rotation which involves angular displacement or orientation change, quantifying the degree and axis of rotational movement which are well-established for capturing real-world variations in wHAR tasks [25].”
Gupta, V., Broughton, J., Rukundo, A., & Pinky, L. Y. (2025) · Informatics in Medicine Unlocked · doi:10.1016/j.imu.2025.101627 · Cited by 5
ML-Aided Dynamic BSR Periodicity Adjustment for Enhanced UL Scheduling in Cellular Systems
Chukhno, N., Saafi, S., & Andreev, S. D. (2025) · IEEE Open Journal of the Communications Society · doi:10.1109/ojcoms.2025.3561002 · Cited by 2
“We employ the concept of the Sliding Window, a robust method used to process sequential data involving the systematic analysis of fixed-sized subsequences one at a time [31], [32].”
Hyunho, J., & Sukyung, P. (2025) · Journal of NeuroEngineering and Rehabilitation · doi:10.1186/s12984-025-01707-8 · Cited by 2
“When constructing the training data-set, we tried to mitigate overfitting by augmenting the data through virtual rotation of the IMU signals [29].”
Ge, Z., Fu, M., Cao, X., Tang, L., Zhang, H., Sun, H., Yuan, G., & Zheng, J. (2025) · Biomedical Signal Processing and Control · doi:10.1016/j.bspc.2025.108910 · Cited by 2
Chaudhry, F. F., Kim, T. W., Elemento, O., & Betel, D. (2026) · Frontiers in Aging Neuroscience · doi:10.3389/fnagi.2026.1730550 · Cited by 4
Forecasting High-risk Areas of COVID-19 Infection Through Socioeconomic and Static Spatial Analysis
Alhamadani, A., Sarkar, S., Zhang, L., Alkulaib, L., & Lu, C. (2021) · IEEE International Conference on Big Data (BigData) · doi:10.1109/bigdata52589.2021.9671896 · Cited by 4
“Moreover, data augmentation can increase the models’ prediction accuracies and enhance the robustness of a model toward overfitting [11], [12], [22], [26], [28], [49].”
Augmentation of Accelerometer and Gyroscope Signals in Biometric Gait Systems
Sawicki, A. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-10539-5_3 · Cited by 3
ATCN: Resource-efficient Processing of Time Series on Edge
Baharani, M., & Tabkhi, H. (2022) · ACM Transactions on Embedded Computing Systems · doi:10.1145/3524070 · Cited by 5
“For benchmarks whose training size is small, such as ECGFiveDays, we applied four types of data augmentation: jittering [23], magnitude warping [37], window warping [25], and scaling.”
MethodChallenges, Opportunities, and Future Prospects
Abdel‐Basset, M., Moustafa, N., Hawash, H., & Ding, W. (2021) · Studies in computational intelligence · doi:10.1007/978-3-030-89025-4_10 · Cited by 2
Small Sample Electromagnetic Signal Recognition based on Time Series Data Augmentation
Wang, X., & Lin, Y. (2021) · International Conference on Dependable Systems and Their Applications (DSA) · doi:10.1109/dsa52907.2021.00049 · Cited by 2
“In [16], the scaling ratio is formed by cubic spline interpolation, and then it enlarges and shrinks the numerical value in the amplitude domain and the time step of the selected window in the time domain.”
BackgroundStochastic Pairing for Contrastive Anomaly Detection on Time Series
Chambaret, G., Berti‐Équille, L., Bouchara, F., Bruno, E., Martin, V., & Chaillan, F. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-09282-4_26 · Cited by 3
“Note that, above parameters of different methods have been obtained with tuning in a range magnitude proposed proposed by Um et al.[17].”
BackgroundMethodEnd-to-end Gesture Recognition Framework for the Identification of Allergic Rhinitis Symptoms
Tzamalis, P., Bardoutsos, A., Markantonatos, D., Raptopoulos, C., Nikoletseas, S., Aggelidis, X., & Papadopoulos, N. G. (2022) · International Conference on Distributed Computing in Smart Systems and the Internet of Things (DCOSS-IoT) · doi:10.1109/dcoss54816.2022.00016 · Cited by 2
“Following the related literature on data augmentation for time-series data [12], a combination of the following techniques are applied to each window instance, which yield appropriate data transformations, and also help improve performance [13]: Data augmentation increases also the robustness…”
MethodZhou, H., Wang, L., Wang, B., & Yang, G. (2022) · IEEE International Conference on Industrial Informatics (INDIN) · doi:10.1109/indin51773.2022.9976121 · Cited by 2
“Jittering, scaling, and rotating strategies are used to augment data, which is similar to image augmentation [25].”
MethodVirtual Sensors With 3D Digital Human Motion for Interactive Simulation
Xia, C., & Sugiura, Y. (2023) · Computer · doi:10.1109/mc.2023.3239344 · Cited by 2
Predicting Group Work Performance from Physical Handwriting Features in a Smart English Classroom
Song, M., Qian, K., Chen, B., Okabayashi, K., Parada‐Cabaleiro, E., Yang, Z., Liu, S., Togami, K., Hidaka, I., Wang, Y., et al. (2021) · IEEE International Conference on Digital Signal Processing (DSP) · doi:10.1145/3458380.3458404 · Cited by 2
“Since the use of bigger datasets usually increases machine learning algorithm performance [38], we applied data augmentation to the 42 image samples.”
Bloch, V., Frondelius, L., Arcidiacono, C., Mancino, M., & Pastell, M. (2022) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2022.07.03.498612 · Cited by 2
“…components were rotated in 3D space by the transformation where 𝑎 𝑥 , 𝑎 𝑦 and 𝑎 𝑧 are the components of the tag acceleration measured along X, Y and Z axis, 𝛼 is a random rotational angle, as it was done by several authors (Um et al., 2017; Kalouris et al., 2019; Li at al., 2021).”
MethodWashington, P. Y. (2023) · medRxiv · doi:10.1101/2023.08.24.23294587 · Cited by 2
“For example, for accelerometer data, rotations simulate different sensor placements and cropping is used to diminish the dependency of event locations [57].”
BackgroundMethodOishi, N., Birch, P., Lago, P., & Roggen, D. (2026) · Knowledge-Based Systems · doi:10.1016/j.knosys.2026.116367 · Cited by 1
UniMTS: Unified Pre-training for Motion Time Series
Zhang, X., Teng, D., Chowdhury, R. R., Li, S., Hong, D., Gupta, R. K., & Shang, J. (2024) · arXiv · Cited by 9
“The physics engine generates sufficient motion time series data, which are subsequently encoded by graph networks and aligned with their corresponding text embeddings through contrastive learning.”
Deep learning approaches for the cardiovascular disease diagnosis using smartphone
Subaşı, A., Kontio, E., & Jafaritadi, M. (2022) · Elsevier eBooks · doi:10.1016/b978-0-323-90548-0.00010-3 · Cited by 2
An Augmentation-agnostic Semantic Preserving Technique for Data Generation
Choi, J., Chung, M., & Kim, J.-Y. (2023) · IEEE International Conference on Big Data (BigData) · doi:10.1109/bigdata59044.2023.10386610 · Cited by 1
“In the experiment, we selected jittering for time-series data augmentation, as it is simple yet effective method to simulate additive sensor noise [6].”
MethodBIMO: Bootstrap Inter–Intra Modality at Once Unsupervised Learning for Multivariate Time Series
Heo, S., Kim, S., & Lee, J. (2024) · Applied Sciences · doi:10.3390/app14093825 · Cited by 1
Technology and Information: Data Augmentation Applied to Industry in Modern Times
Vaz, G. C., de Oliveira, G. G., Arthur, R., & Iano, Y. (2024) · Smart innovation, systems and technologies · doi:10.1007/978-3-031-66961-3_26 · Cited by 2
Test Time Augmentation as a Defense Against Adversarial Attacks on Online Handwriting
Yamashita, Y., & Iwana, B. K. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-70536-6_10 · Cited by 1
Alexandrino, F. Q. D. L., Pacheco, C. C., Carvalho, D., & Ogasawara, E. (2024) · Anais do Brazilian e-Science Workshop (BreSci) · doi:10.5753/bresci.2024.244100 · Cited by 1
Trajectory Augmentation Method Based on Dynamic Movement Primitives
Asemanrafat, A., Taheri, A., & Meghdari, A. F. (2024) · Iranian Journal of Science and Technology Transactions of Mechanical Engineering · doi:10.1007/s40997-024-00809-3 · Cited by 1
“A study by Um et al. ( 2017 ) combined several of these methods to enhance wearable sensor data for Parkinson’s disease monitoring; increasing their classifi - cation accuracy from 77.52 to 85.88%.”
InfluentialFattahi, M., Esmaeil-Zadeh, M., Soltanian-Zadeh, H., Rostami, R., Mansouri, J., & Hossein‐Zadeh, G. (2025) · Frontiers in Human Neuroscience · doi:10.3389/fnhum.2024.1427532 · Cited by 3
Li, H., Zhang, Z., Kang, J., Jiang, X., Gong, X., Zhang, J., Sun, Z., & Cichocki, A. S. (2025) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp49660.2025.10889129 · Cited by 2
“) can be achieved through heuristic algorithms like Jittering, Scaling, Permutation [11], Dropout [12], Freq masking [13], and Freq adding [14], or through deep learning-based methods like ——Xiaoli Gong and Jin Zhang are the corresponding authors.”
S2C‐HAR: A Semi‐Supervised Human Activity Recognition Framework Based on Contrastive Learning
Li, X., Nie, L., Liu, M., Cheng, W., Zhu, H., Wu, X., & Zhan, D. (2025) · Concurrency and Computation Practice and Experience · doi:10.1002/cpe.70027 · Cited by 3
Data Augmentation of Electromyography Signal for Neuromuscular Disorders Diagnosis
Abdel-Maboud, N. F., Parusheva, S. S., Alfonse, M., & Salem, A.-B. M. (2024) · International Conference on Computer Theory and Applications (ICCTA) · doi:10.1109/iccta64612.2024.10974879 · Cited by 1
“In [12] authors used various data augmentation techniques (e.g. jittering, scaling, cropping, rotating, permutating, magnitude-warping and time-warping) for Parkinson’s disease monitoring through data recorded using wearable sensor.”
Sheng, Y., Zhang, B., Zhang, Z., Shao, Y., Yuan, G., Liu, J., & Liu, H. (2025) · Nondestructive Testing And Evaluation · doi:10.1080/10589759.2025.2541050 · Cited by 1
“These three enhancement strategies are Mask, Reorder and Jittering [30].”
Keskin, E., Özcan, O., & Yemez, Y. (2025) · The Visual Computer · doi:10.1007/s00371-025-04125-y · Cited by 1
P-Mix: A Data Augmentation Method for Contrastive Learning Based Human Activity Recognition
Chen, Y., Xie, Q., Cui, W., Chen, L., Song, H., & Zhu, T. (2025) · IEEE Transactions on Artificial Intelligence · doi:10.1109/tai.2025.3601599 · Cited by 1
Yang, S., Shi, H., Yin, J., Xu, X., Yao, J., & Liu, S. (2025) · IEEE Sensors Journal · doi:10.1109/jsen.2025.3602431 · Cited by 5
“Scaling can modify the global amplitude of time series through random scalar values, and time warping can be implemented using smooth warping paths [48], [49].”
Dutta, U. K., & Su, G. (2025) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw67362.2025.00556 · Cited by 1
“• In Table 16, we showcase the performances of different variations of our DAF method on IHS dataset, while also highlighting the impact of the shuffle technique [18].”
Advanced Data Augmentation Techniques for Biosignal Processing in Online Systems
Chen, P., Tang, Y., Li, D., Jiang, H. Y., Togo, S., Yokoi, H., & Jiang, Y. (2025) · IEEE/ASME International Conference on Advanced Intelligent Mechatronics · doi:10.1109/aim64088.2025.11175708 · Cited by 1
“Simulating biosignals variations in an online recognition system through data augmentation methods is crucial for enhancing the model’s accuracy and stability in real-time applications [26].”
Temporal-Aware and Intent Contrastive Learning for Sequential Recommendation
Zhang, Y., Fan, Y., Sheng, T., & Wang, A. (2025) · Symmetry · doi:10.3390/sym17101634 · Cited by 4
Thapa, D., & Rai, R. (2025) · Applied Sciences · doi:10.3390/app151910671 · Cited by 2
Zhang, Y., Bruijn, S. M., Punt, M., Helbostad, J. L., Pijnappels, M. A. G. M., & David, S. (2025) · Medical & Biological Engineering & Computing · doi:10.1007/s11517-025-03466-z · Cited by 2
“Um and colleagues found that adding rotated data can alleviate the effect of variability of sensor orientations, thereby improving accuracy by 11.9% on the testing dataset [20].”
InfluentialEarlyMix: Hierarchical Mixing for Early Time Series Classification
Hu, S., Hu, J., Lv, J., & Zhang, H. (2025) · IEEE International Conference on Multimedia and Expo (ICME) · doi:10.1109/icme59968.2025.11209452 · Cited by 1
Multi-scale Dynamic Preservation Mapping: A Time Series Data Augmentation Approach
Lu, Y., Yan, J., Zhang, W., & Huang, L. (2025) · Lecture notes in computer science · doi:10.1007/978-981-95-4384-7_26 · Cited by 1
RS-ScaleGAN: A Hierarchical Multi-Scale GAN Network for Data Augmentation in Remote Sensing
Lu, Y., & Li, J. (2025) · IEEE International Geoscience and Remote Sensing Symposium (IGARSS) · doi:10.1109/igarss55030.2025.11242550 · Cited by 1
“Most data augmentation techniques are based on random transformations of the training data, such as adding random noise [4], slicing or cropping [5], scaling [6], random warping [4, 6], and frequency warping [7].”
Optimizing Neural Inertial Classification: A Benchmark Study of Data-Driven Techniques
Kruzel, O., Yampolsky, Z., & Klein, I. (2026) · IEEE Journal of Indoor and Seamless Positioning and Navigation · doi:10.1109/jispin.2026.3654901 · Cited by 2
Abdullahi, S. S., Ali, M., Kashif, U. A., Hassan, F., Li, J., Yuhang, L., Omer, M. O. M., Chen, K., & Nie, Z. (2026) · IEEE Transactions on Antennas and Propagation · doi:10.1109/tap.2026.3657796 · Cited by 1
“Techniques like SpecAugment in speech recognition [29], label-preserving transformations in Parkinson’s detection [30], time-warping 2-”
Data Augmentation for Subject-Independent SSVEP-BCIs via Simultaneous Spatial-Energy Representation
Ding, W., Liu, A., Wu, L., Cui, H., Fang, B., & Chen, X. (2026) · IEEE Transactions on Biomedical Engineering · doi:10.1109/tbme.2026.3659606 · Cited by 1
Liu, Y., Zhang, Y., Zhang, X., Yang, Y., Li, Y., & Luo, J. (2025) · Proceedings · doi:10.1109/icdm65498.2025.00109 · Cited by 1
“When generating the positive samples, a lot of works mainly rely on approaches such as random cropping, rotation, color jittering, or noise addition for image data [9], [21], [5], [42], [35], or semantic alignment, event alignment, time-wrapping for sequential data [44], [18], [3], [49], [1].”
Shafiei, S. B., Shadpour, S., Dakwar, A., Xu, Z., & Mohler, J. L. (2026) · Journal of Robotic Surgery · doi:10.1007/s11701-026-03290-z · Cited by 2
“We applied time-series-specific data augmentation techniques to expand the dataset and improve classification performance [22, 23].”
Gao, E.-T., Yuan, H., Zhong, H.-Y., Yuan, R., & Chen, Z. (2026) · International Conference on Human Factors in Computing Systems · doi:10.1145/3772318.3790990 · Cited by 5
“Multi-scale temporal enhancements were applied to the saccade and fixation features by incorporating sliding-window statistics, differential features, and their combinations to construct enriched time-series representations [153].”
Trust-Centric Mitigation of Time-Warping Adversarial Attacks in Wearable Sensor Platforms
Sahay, R., Lakshmi, L., Sai, P. V. J., Althobaiti, M. M., Salhi, A., Khan, M. I., & Ismoilov, M. (2026) · IEEE Transactions on Consumer Electronics · doi:10.1109/tce.2026.3681806 · Cited by 1
“Um et al. [15] showed that the accuracy of HAR is sig-nificantly a ff ected by even basic temporal distortions added by means of augmentation or adversarial manipulation, which can be interpreted as evidence of the sensitivity of cadence-driven models.”
Devarapelly, A. R., Singh, A., Najeh, T., Kankar, P. K., Mandal, T. K., Karim, R., Kumar, U., & Varde, P. V. (2026) · MAPAN · doi:10.1007/s12647-026-00941-2 · Cited by 1
“Scaling simulates amplitude variations caused by changing operating conditions or sensor calibration differences [25].”
Dual Data-centric Separation with Circular Mixup for Noise-resistant Time Series Learning
Pei, Y., Meng, F., Ran, Q., Ren, T., Wang, Y., Ju, W., Wang, Z., Hua, X., & Luo, X. (2026) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/3770854.3780265 · Cited by 1
Li, C., Yang, H., Cheng, L., & Huang, F. (2022) · Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) · doi:10.1109/embc48229.2022.9871795 · Cited by 1
“In the work of [1], the methods of jittering, scaling, rotating, permutating, magnitude-warping and time-warping were used to augment Parkinson’s data recorded by accelerometer sensors, to assess the degree”
MethodSong, P., Dong, Y., & Zhang, K. (2023) · Communications in computer and information science · doi:10.1007/978-981-99-1642-9_1 · Cited by 1
Towards a New Multi-tasking Learning Approach for Human Fall Detection
Nguyen, D., Pham, C., Argent, R., Caulfield, B., & Le‐Khac, N. (2023) · Lecture notes in networks and systems · doi:10.1007/978-3-031-36886-8_5 · Cited by 2
Diagnosing Rolling-Element Bearing Faults in the Real World: Problem Solved?
Law, Y. W., Qiao, Y., Chow, C. W. K., Marian, R. M., Jolfaei, N. G., Gorjian, N., & Pan, J. (2023) · Annual Conference of the IEEE Industrial Electronics Society (IECON) · doi:10.1109/iecon51785.2023.10312154 · Cited by 1
“…see how Yu et al.’s DA method fares in comparison to more traditional time-domain DA methods, namely per-mutation [34], slicing [35], time warping [34] and window warping [35]; we tested all these methods, following [21], on all the 128 datasets from the 2018 UCR Time Series Archive repository in…”
BackgroundMethodInfluentialUse of Domain Labels during Pre-Training for Domain-Independent WiFi-CSI Gesture Recognition
van Berlo, B., Verhoeven, R., & Meratnia, N. (2023) · Sensors · doi:10.3390/s23229233 · Cited by 3
“The augmentations were directly taken from research studies involving Doppler Frequency Spectogram (DFS) and three-dimensional acceleration signal augmentation conducted by Zhang et al. and Um et al. [38,39].”
MethodA study on data augmentation optimization for data-centric health prognostics of industrial systems
Gay, A., Voisin, A., Iung, B., Do, P., Bonidal, R., & Khelassi, A. (2023) · IFAC-PapersOnLine · doi:10.1016/j.ifacol.2023.10.1754 · Cited by 1
A Weakly Supervised Learning Framework for Parkinson’s Disease Assessment Using Wearable Sensor
Li, Z., Peng, X., Zhao, Y., Wang, X., Yang, Y., & Yang, P. (2023) · International Conference on Mobility, Sensing and Networking (MSN) · doi:10.1109/msn60784.2023.00119 · Cited by 1
“In order to find the optimal number of cluster centers to determine the type of words, we set the number of cluster centers k ∈ [4,5,6,7,8,9,10,11,12,13,14,15,16,17].”
BackgroundThe Impact of Synthetic Data on Fall Detection Application
Debnath, M., Kabir, M. S., Ni, J., & Ngu, A. H. H. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-66538-7_21 · Cited by 5
X-Tacformer : Spatio-tempral Attention Model for Tactile Recognition
Hu, J., Zhou, Y., Wang, Z., Li, X., Jiang, Y., & He, B. (2024) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra57147.2024.10610365 · Cited by 1
“Therefore, we adopt the method proposed by T. Um et al. [20] to extend ALU-Textures.”
BackgroundMethodDomain Adaptation in Human Activity Recognition through Self-Training
Kfari, M. K. A., & Lüdtke, S. (2024) · ACM International Joint Conference on Pervasive and Ubiquitous Computing (UbiComp) · doi:10.1145/3675094.3678465 · Cited by 3
Kręcichwost, M., Badura, P., Piet, A., Hasan, M. A., Moćko, N., Miodońska, Z., Sage, A., & Grzegorzek, M. (2025) · Advances in intelligent systems and computing · doi:10.1007/978-3-031-95582-2_12 · Cited by 1
Hu, X., Li, N., Ghahramani, A., & Chen, Y. (2025) · Journal of Building Engineering · doi:10.1016/j.jobe.2025.113610 · Cited by 3
Shen, Y., Park, J. H., Huang, M., Benge, J. F., Rousseau, J. F., Lester‐Smith, R. A., & Thomaz, E. (2025) · IEEE-EMBS International Conference on Biomedical and Health Informatics (BHI ...) · doi:10.1109/bhi67747.2025.11269523 · Cited by 1
“Um et al. [20] applied various signal transformations to augment wearable accelerometer time series for monitoring Parkinson’s disease.”
Zhang, Z., Han, Y., Ma, B., & Geng, Z. (2025) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2025.3548177 · Cited by 2
“For continuous time series data, explicit augmentation methods usually include jittering, scaling, inverting, reversing, shuffling, scrambling, and warping [17].”
Wearable Devices and Privacy Concerns: Data Collection, Analysis, and Interpretation
Dincelli, E., Zhou, X., Yayla, A., & Jafarian, H. (2020) · Advances in information security, privacy, and ethics book series · doi:10.4018/978-1-7998-3487-8.ch004 · Cited by 1
Interactive Knowledge-Guided Learning
Nordsieck, R., & Hähner, J. (2020) · IEEE International Conference on Autonomic Computing and Self-Organizing Systems Companion (ACSOS-C) · doi:10.1109/acsos-c51401.2020.00066 · Cited by 1
An autonomous navigation system for pedestrian based on foot-bound MIMU and smartphone
Pan, X., & Zhang, S. (2020) · International Conference on Intelligent Human-Machine Systems and Cybernetics · doi:10.1109/ihmsc49165.2020.10124 · Cited by 1
Augmentation of Gait Cycles Using LSTM-MDN Networks in Person Identification System
Sawicki, A. (2021) · Lecture notes in computer science · doi:10.1007/978-3-030-84340-3_4 · Cited by 1
Hansen, T. F. (2022) · Environmental Monitoring and Assessment · doi:10.1007/s10661-021-09744-1 · Cited by 1
“Recently published works specifically using augmented motion sensor data to improve model training data efficiency of vibrational motion include a study of the monitoring of building site construction equipment (Rashid & Louis, 2019) and a very welldocumented study on using human-wearable motion sensors to monitor the…”
BackgroundMethodInfluentialA Novel Approach for Health Analysis Using Machine Learning Approaches
Bhattacharya, D., Rao, N. T., Vardhan, K. A., & Joshua, E. S. N. (2023) · Lecture notes in networks and systems · doi:10.1007/978-981-19-6880-8_19 · Cited by 1
DOD-Explainer: Explainable Drug Overdose Deaths Predictor from Crime and Socioeconomic Data
Alhamadani, A., Sarkar, S., Alkulaib, L., & Lu, C. (2022) · IEEE International Conference on Big Data (BigData) · doi:10.1109/bigdata55660.2022.10021054 · Cited by 1
Parkinson's desease detection by multimodal analysis combining handwriting and speech signals
Taleb, C. (2020) · theses.fr (ABES) · doi:10.70675/a9a3a82ezcf77z4897z8493zac34a08e613f · Cited by 2
Haresamudram, H., Tang, C. I., Suh, S., Lukowicz, P., & Ploetz, T. (2024) · arXiv · Cited by 31
“It relies on a set of eight data transformations / augmentations introduced by Um et al. [184] for increasing sensor data diversity, including adding random Gaussian noise, scaling, rotations, negation, flipping channels, permuting sub-segments of windows, time-warping and channel shuffling.”
Investigating Enhancements to Contrastive Predictive Coding for Human Activity Recognition
Haresamudram, H., Essa, I., & Ploetz, T. (2022) · arXiv · Cited by 27
“Another contrastive learning method includes SimCLR [24], [26], [27], which heavily relies on data transformations [32] as it contrasts two randomly augmented views of the data in a siamese setup, resulting in highly effective representations.”
MethodCalatrava-Nicolás, F. M., Stoyanov, T., & Martinez Mozos, O. (2026) · Scientific Reports · doi:10.1038/s41598-026-64846-5
“Prior work has shown that rotation augmentation, i.e., applying a 3-dimensional rotation (via a rotation matrix) to inertial measurements at each time step, is particularly effective for diversifying training samples in HAR 26,30 .”
Sensing-HH: A Deep Hybrid Attention Model for Footwear Recognition
Yao, Y., Wen, Y., & Wang, J. (2020) · Electronics · doi:10.3390/electronics9091552
“More and more researchers have used the motion characteristics of the human body for various tasks, which ranged from activity recognition [1–6], gesture categorization [7], clinical condition monitoring [8], BMI predication [9], to user gait recognition [10–13].”
BackgroundAldeer, M. M. N., Howard, R. E., Martin, R. P., & Ortiz, J. (2021) · doi:10.1145/3458648.3460011 · Cited by 1
“We start by using 10 % raw timeseries labeled data from the target domain (corresponding to 1 recurrence plot image per user) and feed them to an augmenter to produce some synthetic samples [24].”
MethodMwaniki, P., Kamanu, T., Akech, S., Dunsmuir, D. T., Ansermino, J. M., & Eijkemans, M. J. C. (2021) · Wellcome Open Research · doi:10.12688/wellcomeopenres.17148.1
Taccari, L. (2021) · IEEE International Conference on Computer Vision Workshops · doi:10.1109/iccvw54120.2021.00037
“In our experiments we use a custom set of transformations, inspired by those mentioned in [11, 22], that include axial rotations, additive noise, several kinds of filtering, warping, and cut-out of parts of the signal.”
MethodWearable Devices and Privacy Concerns: Data Collection, Analysis, and Interpretation
Dincelli, E., Zhou, X., Yayla, A., & Jafarian, H. (2021) · IGI Global eBooks · doi:10.4018/978-1-7998-8954-0.ch009
Light-Weight Permutation Generator for Efficient Convolutional Neural Network Data Augmentation
Kwan, B. P. Y., Guo, C., Luk, W. W., & Jiang, P. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-19983-7_11
Secure RFID Handwriting Recognition–Attacker Can Hear but Cannot Understand
Zhang, Q., Zhang, J., Liu, X., Tong, X., & Li, K. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-19208-1_34
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.