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 101–200 of 736
Hilbert sEMG data scanning for hand gesture recognition based on deep learning
Tsinganos, P., Cornelis, B., Cornelis, J. P., Jansen, B., & Skodras, A. N. (2020) · Neural Computing and Applications · doi:10.1007/s00521-020-05128-7 · Cited by 45
“In addition, sEMG signals are augmented with the magnitude-warping method described in [50, 51].”
MethodGuerra, B. M. V., Torti, E., Marenzi, E., Schmid, M., Ramat, S., Leporati, F., & Danese, G. (2023) · Frontiers in Neuroscience · doi:10.3389/fnins.2023.1256682 · Cited by 61
“A recent trend is to exploit data augmentation strategies to create synthetic data both to increase the training set size and to balance it (Um et al., 2017; Mathur et al., 2018; Steven Eyobu and Han, 2018).”
Wang, C., Olugbade, T., Mathur, A., Williams, A. C. D. C., Lane, N. D., & Bianchi‐Berthouze, N. (2019) · International Symposium on Wearable Computers · doi:10.1145/3341163.3347728 · Cited by 44
“15 and globally add each to the original data; ii) Random Discarding (similar to cropping [12]), which is to simulate unexpected data loss at some points during the recording.”
BackgroundMethodMekruksavanich, S., & Jitpattanakul, A. (2020) · Advances in intelligent systems and computing · doi:10.1007/978-981-15-8289-9_72 · Cited by 41
Leveraging Activity Recognition to Enable Protective Behavior Detection in Continuous Data
Wang, C., Gao, Y., Mathur, A., De C. Williams, A. C., Lane, N. D., & Bianchi-Berthouze, N. (2021) · Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies · doi:10.1145/3463508 · Cited by 41
“In order to fully leverage the existing data for our model development, we followed the experience of previous studies [25, 40, 64] using data augmentations [77].”
MethodMohammad, Z., Anwary, A. R., Mridha, M. F., Shovon, M. S. H., & Vassallo, M. (2023) · Sensors · doi:10.3390/s23104774 · Cited by 50
“Nevertheless, in this work, we used arbitrary rotation, scaling, and jitter transformation processes defined in [30] to augment the data.”
MethodRavindra, N. G., Espinosa, C. A., Berson, E., Phongpreecha, T., Zhao, P., Becker, M., Chang, A. L., Shome, S., Marić, I., Francesco, D. D., et al. (2023) · npj Digital Medicine · doi:10.1038/s41746-023-00911-x · Cited by 50
“36 .”
BackgroundInfluentialLiu, Y., Zhang, S., & Gowda, M. (2021) · Proceedings of the International Conference on Internet-of-Things Design and Implementation · doi:10.1145/3450268.3453537 · Cited by 39
“Data augmentation techniques have been explored in the context of wearable sensing for parkinson disease gait monitoring [68] and construction activity monitoring [53].”
BackgroundWu, X., Zhao, Z., Tian, R., Gao, S., Niu, Y., & Liu, H. (2020) · Food Chemistry · doi:10.1016/j.foodchem.2020.127640 · Cited by 40
Sensor-data augmentation for human activity recognition with time-warping and data masking
Jeong, C. Y., Shin, H. C., & Kim, M. (2021) · Multimedia Tools and Applications · doi:10.1007/s11042-021-10600-0 · Cited by 38
“However, because accelerometer signals have characteristics of time-series data, augmentation methods [4, 12, 18, 24, 33] using signal processing (e.g., jittering, scaling, rotation, and random sampling) are widely used in the field of HAR.”
BackgroundMethodResultInfluentialZhan, X., Kou, L., Xue, M., Zhang, J., & Zhou, L. (2022) · IEEE Transactions on Reliability · doi:10.1109/tr.2022.3174093 · Cited by 40
“In order to solve the abovementioned problem, the promising research will focus on five directions: 1) time-frequency domain enhancement, 2) imbalance class enhancement, 3) selection and combination of enhancement, 4) Gaussian enhancement process, and 5) more deep neural network enhancement models [31] approaches.”
BackgroundIMU Based Deep Stride Length Estimation With Self-Supervised Learning
Sui, J.-D., & Chang, T. (2021) · IEEE Sensors Journal · doi:10.1109/jsen.2021.3049523 · Cited by 40
“[26] also proposed many sensor data augmentation methods, like jittering, permutating, and time-warping methods.”
BackgroundFalls Risk Classification of Older Adults Using Deep Neural Networks and Transfer Learning
Martinez, M., & Leon, P. L. D. (2019) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2019.2906499 · Cited by 41
“Deep learning has recently been applied to gait analysis [11], [12] and gait disorder classification [13], [14].”
MethodJayarathne, I., Cohen, M., & Amarakeerthi, S. (2020) · PLoS ONE · doi:10.1371/journal.pone.0238872 · Cited by 45
“Therefore, three signal augmentation techniques which have been used to improve classification performance of wearable sensor data was used [24].”
MethodKim, Y., & Lee, W. (2022) · Sensors · doi:10.3390/s22249900 · Cited by 35
“DA is a technique to generate larger amounts of new data by changing existing data [32].”
MethodA CNN-Based Animal Behavior Recognition Algorithm for Wearable Devices
Pan, Z., Chen, H., Zhong, W., Wang, A., & Zheng, C. (2023) · IEEE Sensors Journal · doi:10.1109/jsen.2023.3239015 · Cited by 34
“A variety of data augmentation methods were proposed for wearable sensor data in [15], which were applied to the classification of motor state of Parkinson’s disease patients.”
MethodInfluentialOperational prediction of solar flares using a transformer-based framework
Abduallah, Y., Wang, J. T. L., Wang, H., & Xu, Y. (2023) · Scientific Reports · doi:10.1038/s41598-023-40884-1 · Cited by 34
“Here, we adopt the Gaussian white noise (GWN) data augmentation scheme because it has shown a significant improvement in model performance 43,44 .”
MethodChu, X., Kim, S., Cho, Y., & Park, K.-S. (2024) · Electronics · doi:10.3390/electronics13040747 · Cited by 42
Otsuka, R., Yoshimura, N., Tanigaki, K., Koyama, S., Mizutani, Y., Yoda, K., & Maekawa, T. (2024) · Methods in Ecology and Evolution · doi:10.1111/2041-210x.14294 · Cited by 39
“Similar to the HAR study (Um et al., 2017 ), Experiment 1 also showed that differences in attachment position could be handled by data augmentation, rotation for black-tailed gulls in particular.”
MethodResultData Augmentation for Deep Learning-Based ECG Analysis
Pan, Q., Li, X., & Fang, L. (2020) · Springer eBooks · doi:10.1007/978-981-15-3824-7_6 · Cited by 48
SenseCollect: We Need Efficient Ways to Collect On-body Sensor-based Human Activity Data!
Chen, W., Lin, S., Thompson, E., & Stankovic, J. A. (2021) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3478119 · Cited by 52
“Um et al. demonstrated that combining three basic ways (rotation, permutation, and time warping) yields better performance than using a single method [48].”
MethodPPGface: Like What You Are Watching? Earphones Can "Feel" Your Facial Expressions
Choi, S., Gao, Y., Jin, Y., Kim, S. J., Li, J., Xu, W., & Jin, Z. (2022) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3534597 · Cited by 48
“Typical data augmentation methods rotate, scale, time-warp, and crop the original data to create new data [86], and we utilize these techniques for our self-collected PPG and accelerometer data.”
MethodData Augmentation for Human Activity Recognition With Generative Adversarial Networks
Lupión, M., Cruciani, F., Cleland, I., Nugent, C., & Ortigosa, P. M. (2024) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2024.3364910 · Cited by 44
Echabarri, S., Do, P., Vu, H. C., & Liegeois, P.-Y. (2025) · Reliability Engineering & System Safety · doi:10.1016/j.ress.2025.111297 · Cited by 39
Learning acoustic features to detect Parkinson’s disease
Wu, K., Zhang, D. D., Lu, G., & Guo, Z. (2018) · Neurocomputing · doi:10.1016/j.neucom.2018.08.036 · Cited by 42
Identifying mental health status using deep neural network trained by visual metrics
Shafiei, S. B., Lone, Z. M., Elsayed, A. S., Hussein, A. A., & Guru, K. A. (2020) · Translational Psychiatry · doi:10.1038/s41398-020-01117-5 · Cited by 39
“To prevent happening this problem, we applied data augmentation techniques to time-series data to enlarge the size of the dataset and increase classi fi cation accuracy 62 – 64 .”
MethodPham, C., Nguyen, L., Nguyen, A., Nguyen, C.-N., & Nguyen, V.-T. (2021) · Multimedia Tools and Applications · doi:10.1007/s11042-021-11058-w · Cited by 38
“In addition, we utilize data augmentation for more diversified training data by adopting three augmentation techniques from [35].”
MethodNi, Z., Wu, T., Wang, T., Sun, F., & Li, Y. (2022) · IEEE Transactions on Biomedical Engineering · doi:10.1109/tbme.2022.3163429 · Cited by 39
“[40]; 2) Swap the 3-axis of accelerometer data or gyroscope data with random permutations and rotate them by a random angle to simulate scenarios where inertial sensors were placed”
BackgroundLai, D. K.-H., Yu, Z., Leung, T. Y.-N., Lim, H.-J., Tam, A. Y.-C., So, B. P.-H., Mao, Y.-J., Cheung, D. S. K., Wong, D. W., & Cheung, J. C. (2023) · Sensors · doi:10.3390/s23052475 · Cited by 38
“Data augmentation techniques, including scaling, magnitude warping, flipping, time Data augmentation techniques, including scaling, magnitude warping, flipping, time warping, and window warping [51,52] were applied (Figure 3). each frame of the signal by a random scalar.”
MethodInfluentialLabel-preserving data augmentation for mobile sensor data
Kim, M., & Jeong, C. Y. (2021) · Multidimensional Systems and Signal Processing · doi:10.1007/s11045-020-00731-2 · Cited by 32
“In recent years, research on the use of deep learning in a wide range of applications using mobile sensor data has received much attention (Ohashi et al. 2017; Um et al. 2017; Chung et al. 2019; Jeong and Kim 2019).”
BackgroundMethodInfluentialLost circulation monitoring using bi-directional LSTM and data augmentation
Sun, W., Li, W., Zhang, D., Liu, K., Chen, W., Dai, Y., & Huang, W. (2023) · Geoenergy Science and Engineering · doi:10.1016/j.geoen.2023.211660 · Cited by 35
Wang, H., Lin, J., & Zhang, Z. (2024) · Energy and Buildings · doi:10.1016/j.enbuild.2024.114192 · Cited by 32
Conditional GAN based augmentation for predictive modeling of respiratory signals
Jayalakshmy, S., & Sudha, G. F. (2021) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2021.104930 · Cited by 38
“The authors [26 – 29] explored DA techniques in the speech repository, auditory signals and sensor based models and classified enhanced data set using CNN with good accuracy.”
BackgroundLiu, Y., Chen, Z., Wei, L., Wang, X., & Li, L. (2022) · IEEE Transactions on Industrial Electronics · doi:10.1109/tie.2022.3231287 · Cited by 32
“independent solution, can alleviate such problems by transforming existing data into new ones [33].”
BackgroundHaresamudram, H., Tang, C. I., Suh, S., Lukowicz, P., & Plötz, T. (2025) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3729467 · Cited by 33
“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.”
Toward Robust Detection of Puppet Attacks via Characterizing Fingertip-Touch Behaviors
Wu, C., He, K., Chen, J., Zhao, Z., & Du, R. (2021) · IEEE Transactions on Dependable and Secure Computing · doi:10.1109/tdsc.2021.3116552 · Cited by 35
“Previous work has demonstrated that combinations of various data augmentation methods show better performance than using a single data augmentation method [27], [28], therefore we use the combination of the four transformations to perform sensor data augmentation.”
BackgroundMethodMining knowledge from unlabeled data for fault diagnosis: A multi-task self-supervised approach
Kong, D., Huang, W., Zhao, L., Ding, J., Wu, H., & Yang, G. (2024) · Mechanical Systems and Signal Processing · doi:10.1016/j.ymssp.2024.111189 · Cited by 33
Automated identification of abnormal infant movements from smart phone videos
Passmore, E., Kwong, A., Greenstein, S., Olsen, J. E., Eeles, A. L., Cheong, J. L. Y., Spittle, A. J., & Ball, G. (2024) · PLOS Digital Health · doi:10.1371/journal.pdig.0000432 · Cited by 41
Bui, V., Pham, T. L., Nguyen, H. Q., & Jang, Y. M. (2021) · Applied Sciences · doi:10.3390/app11052166 · Cited by 37
“In these studies, the traditional data transformation methods do not significantly improve the accuracy of the model [9–11 , 14].”
MethodSigcha, L., Domínguez, B., Borzì, L., Costa, N., da Costa, S. P., Arezes, P., López, J. M., de Arcas, G., & Pavón, I. (2022) · Electronics · doi:10.3390/electronics11233879 · Cited by 40
Kwon, H., Abowd, G. D., & Plötz, T. (2021) · Sensors · doi:10.3390/s21248337 · Cited by 30
“To tackle this “small data” problem, the research community has explored many different avenues, ranging from, for example, data augmentation [10,11], transfer learning [12], to self-supervised learning techniques [13].”
BackgroundMethodInfluentialUchitomi, H., Ming, X., Zhao, C., Ogata, T., & Miyake, Y. (2023) · Scientific Reports · doi:10.1038/s41598-023-39862-4 · Cited by 35
“In contrast, in augmentation via time-domain transformation, the elements of the time series are displaced to time steps that differ from the original sequence.”
BackgroundMethodInfluentialZhang, J., Srivatsa, P., Ahmadzai, F. H., Liu, Y., Song, X., Karpatne, A., Kong, Z., & Johnson, B. N. (2023) · ACS Sensors · doi:10.1021/acssensors.3c01258 · Cited by 32
Simulating Time-Series Data for Improved Deep Neural Network Performance
Yeomans, J., Thwaites, S., Robertson, W. S. P., Booth, D., Ng, B. W. ‐., & Thewlis, D. (2019) · IEEE Access · doi:10.1109/access.2019.2940701 · Cited by 37
Lee, U. H., Bi, J., Patel, R., Fouhey, D. F., & Rouse, E. J. (2020) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2020.3007455 · Cited by 35
“number of parameters in these architectures, it may be challenging to determine the optimal parameters from relatively small datasets [16].”
BackgroundKim, K., & Jeong, J. (2020) · Sensors · doi:10.3390/s20247099 · Cited by 35
“In this paper, we increased data by applying jittering and scaling techniques to existing data [4].”
MethodWang, Y., Li, Z., Wang, X., Yu, H., Liao, W., & Arifoğlu, D. (2021) · Machines · doi:10.3390/machines9120367 · Cited by 33
MSFF-Net: Multi-Stream Feature Fusion Network for surface electromyography gesture recognition
Peng, X., Zhou, X. N., Zhu, H., Ke, Z., & Pan, C. (2022) · PLoS ONE · doi:10.1371/journal.pone.0276436 · Cited by 30
“Furthermore, the amplitude and duration of sEMG cannot be fully replicated during motion repetition acquisition, we timewarped the training set data [32] and expanded the training data to twice the original size.”
BackgroundQin, X., Wang, J., Ma, S., Lu, W., Zhu, Y., Xie, X., & Chen, Y. (2023) · Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining · doi:10.1145/3580305.3599360 · Cited by 29
“Specifically, we employ seven data transformation techniques [42] to get the augmented data.”
MethodYu, X., Ma, T., Jang, J., & Xiong, S. (2022) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2022.3228598 · Cited by 32
“the normal distribution instead of the commonly used uniform distribution [33] to describe sensor rotation error in actual”
MethodSynthetic ECG signals generation: A scoping review
Zanchi, B., Monachino, G., Fiorillo, L., Conte, G., Auricchio, A., Tzovara, A., & Faraci, F. D. (2024) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2024.109453 · Cited by 44
Hu, B. H., Simon, A. M., & Hargrove, L. J. (2019) · IEEE Transactions on Medical Robotics and Bionics · doi:10.1109/tmrb.2019.2952148 · Cited by 34
“recognition using wearable sensors, applying analogous transformations for data augmentation not only preserved class labels but also improved binary classification accuracy for detecting Parkinson’s disease symptoms using a single wearable accelerometer [25].”
MethodPérez, J., Arroba, P., & Moya, J. M. (2022) · Applied Intelligence · doi:10.1007/s10489-022-03557-6 · Cited by 30
“The most commonly used traditional methods of time-series data augmentation include Gaussian noise addition, rotation, scaling, warping, and permutation [25].”
MethodImproving biosensor accuracy and speed using dynamic signal change and theory-guided deep learning
Zhang, J., Srivatsa, P., Ahmadzai, F. H., Liu, Y., Song, X., Karpatne, A., Kong, Z., & Johnson, B. N. (2023) · Biosensors and Bioelectronics · doi:10.1016/j.bios.2023.115829 · Cited by 32
Improving CNN-based activity recognition by data augmentation and transfer learning
Kalouris, G., Zacharaki, E. I., & Megalooikonomou, V. (2019) · 2019 IEEE 17th International Conference on Industrial Informatics (INDIN) · doi:10.1109/indin41052.2019.8972135 · Cited by 28
“Augmentation [26] [27] [28] is typically performed in data space or, as shown more recently, in feature space.”
BackgroundMethodInfluentialAbir, F. F., Alyafei, K. A., Chowdhury, M. E. H., Khandakar, A. A., Ahmed, R., Hossain, M., Mahmud, S., Rahman, A., Abbas, T. O., Zughaier, S. M., et al. (2022) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2022.105682 · Cited by 34
“We employed seven techniques to augment the original train dataset [47, 48].”
CubeNet: Array-Based Seismic Phase Picking with Deep Learning
Chen, G., & Li, J. (2022) · Seismological Research Letters · doi:10.1785/0220220147 · Cited by 38
iTimes: Investigating Semisupervised Time Series Classification via Irregular Time Sampling
Liu, X., Zhang, F., Liu, H., & Fan, H. (2022) · IEEE Transactions on Industrial Informatics · doi:10.1109/tii.2022.3199374 · Cited by 29
“We use time warping, and magnitude warping data augmentation [29] for all models.”
MethodAdaimi, R., & Thomaz, E. (2022) · Sensors · doi:10.3390/s22186881 · Cited by 35
“To do so, with every incoming batch, we apply four data augmentation techniques [60] enlarging the batch size four times by: (1) adding random jitter or noise, (2) scaling by a constant amount, (3) time warping, and (4) magnitude warping.”
BackgroundMethodNeural networks generative models for time series
Gatta, F., Giampaolo, F., Prezioso, E., Mei, G., Cuomo, S., & Piccialli, F. (2022) · Journal of King Saud University - Computer and Information Sciences · doi:10.1016/j.jksuci.2022.07.010 · Cited by 41
Ott, F., Rügamer, D., Heublein, L., Hamann, T., Barth, J., Bischl, B., & Mutschler, C. (2022) · International Journal on Document Analysis and Recognition (IJDAR) · doi:10.1007/s10032-022-00415-6 · Cited by 37
“, scaling of acceleration signals [93].”
Detection and prediction of freezing of gait with wearable sensors in Parkinson’s disease
Zhang, W., Sun, H., Huang, D., Zhang, Z., Li, J., Wu, C., Sun, Y., Gong, M., Wang, Z., Sun, C., et al. (2023) · Neurological Sciences · doi:10.1007/s10072-023-07017-y · Cited by 36
“However, some studies [96, 108] suggested that data augmentation performed in time domain could also improve system performance of FoG detection.”
Resource-Efficient Pet Dog Sound Events Classification Using LSTM-FCN Based on Time-Series Data
Kim, Y., Sa, J., Chung, Y., Park, D., & Lee, S.-J. (2018) · Sensors · doi:10.3390/s18114019 · Cited by 27
“[28] proposed a method for applying convolutional neural networks (CNNs) to Parkinson’s disease data acquired from wearable sensors.”
MethodInfluentialMachine Learning for Detection and Risk Assessment of Lifting Action
Thomas, B., Lu, M., Jha, R., & Bertrand, J. (2022) · IEEE Transactions on Human-Machine Systems · doi:10.1109/thms.2022.3212666 · Cited by 30
“For the NIOSH dataset, LP augmentation was performed using the augmentation strategies defined in [31] to determine which augmentations and how much of them could be applied to the training data without changing the label.”
MethodSingh, A., Chakraborty, S., He, Z., Tian, S., Zhang, S., Lustria, M. L. A., Charness, N. H., Roque, N. A., Harrell, E. R., & Boot, W. R. (2022) · Frontiers in Psychology · doi:10.3389/fpsyg.2022.980778 · Cited by 29
“Data augmentation has been used in conjunction with time series data with promising empirical results (Um et al., 2017; Wen et al., 2020).”
BackgroundMethodInfluentialUsing Artificial Intelligence Methods to Classify Different Seismic Events
Wang, T., Bian, Y., Zhang, Y., & Hou, X. (2022) · Seismological Research Letters · doi:10.1785/0220220055 · Cited by 27
Arı, E., & Taçgın, E. (2024) · Biomedical Signal Processing and Control · doi:10.1016/j.bspc.2024.106081 · Cited by 29
Investigating Enhancements to Contrastive Predictive Coding for Human Activity Recognition
Haresamudram, H., Essa, I., & Plötz, T. (2023) · 2023 IEEE International Conference on Pervasive Computing and Communications (PerCom) · doi:10.1109/percom56429.2023.10099197 · 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.”
MethodWang, Y., Cheng, X., Zhou, P., Li, B., & Yuan, X. (2019) · IEEE Sensors Journal · doi:10.1109/jsen.2019.2907051 · Cited by 24
“Then, we randomly permute the segments to create a new sequential data [28].”
MethodHybrid optimization enabled deep learning technique for multi-level intrusion detection
Selvan, G. S. R. E., Maria, A., Vinodkumar, C. R., & Parthasarathy, G. (2022) · Advances in Engineering Software · doi:10.1016/j.advengsoft.2022.103197 · Cited by 25
Longitudinal Trend Monitoring of Multiple Sclerosis Ambulation Using Smartphones
Creagh, A. P., Dondelinger, F., Lipsmeier, F., Lindemann, M., & Vos, M. D. (2022) · IEEE Open Journal of Engineering in Medicine and Biology · doi:10.1109/ojemb.2022.3221306 · Cited by 31
“Smartphone orientation augmentation was performed by randomly rotating sensor-channel axis during training [36].”
MethodAutoAugHAR: Automated Data Augmentation for Sensor-based Human Activity Recognition
Zhou, Y., Zhao, H., Huang, Y., Röddiger, T., Kurnaz, M., Riedel, T., & Beigl, M. (2024) · Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies · doi:10.1145/3659589 · Cited by 25
“DA Combination Challenge : Research [32, 67, 70] has demonstrated notable improvements in HAR tasks by training models combining multiple DA methods.”
BackgroundMethodInfluentialOptimizing Sensor Position with Virtual Sensors in Human Activity Recognition System Design
Xia, C., & Sugiura, Y. (2021) · Sensors · doi:10.3390/s21206893 · Cited by 29
Role of Data Augmentation Strategies in Knowledge Distillation for Wearable Sensor Data
Jeon, E. S., Som, A., Shukla, A., Hasanaj, K., Buman, M. P., & Turaga, P. K. (2021) · IEEE Internet of Things Journal · doi:10.1109/jiot.2021.3139038 · Cited by 25
“tion methods in the time domain helps to improve performance in classification [42].”
TimeDRL: Disentangled Representation Learning for Multivariate Time-Series
Chang, C., Chan, C.-T., Wang, W., Peng, W.-C., & Chen, T.-F. (2024) · IEEE International Conference on Data Engineering (ICDE) · doi:10.1109/icde60146.2024.00054 · Cited by 56
“Remarkably, the benefits of TimeDRL’s pre-training stage are evident even when 100% of the labels are available.”
BackgroundLópez-Nava, I. H., Valentín-Coronado, L. M., García-Constantino, M., & Favela, J. (2020) · Sensors · doi:10.3390/s20174756 · Cited by 28
“These techniques that are commonly used in the field of activity recognition using time-series were analyzed [28,29], implementing those that best fit the data types.”
MethodZhang, S., Nemati, E., Dinh, M., Folkman, N., Ahmed, T., Rahman, M., Kuang, J., Alshurafa, N., & Gao, A. (2022) · ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · doi:10.1109/icassp43922.2022.9746334 · Cited by 23
Yhdego, H. G., Paolini, C. P., & Audette, M. A. (2023) · Applied Sciences · doi:10.3390/app13084988 · Cited by 35
“Wearable sensor data can be transformed in a label-preserving way, but this is not intuitive [39].”
Data Augmentation Strategies for Human Activity Data Using Generative Adversarial Neural Networks
Hoelzemann, A., Sorathiya, N., & Laerhoven, K. V. (2021) · IEEE International Conference on Pervasive Computing and Communications Workshops (PerCom Workshops) · doi:10.1109/percomworkshops51409.2021.9431046 · Cited by 26
“Traditional transformations of images, e.g. scale, zoom, crop or add noise to the data were adapted and transferred to time-series data by [22].”
MethodHeidarivincheh, F., McConville, R., Morgan, C., McNaney, R., Masullo, A., Mirmehdi, M., Whone, A. L., & Craddock, I. J. (2021) · Sensors · doi:10.3390/s21124133 · Cited by 32
SensorSCAN: Self-supervised learning and deep clustering for fault diagnosis in chemical processes
Golyadkin, M., Pozdnyakov, V., Евгеньевич, Ж. Л., & Makarov, I. (2023) · Artificial Intelligence · doi:10.1016/j.artint.2023.104012 · Cited by 29
Chen, L., Hu, R., Wu, M., & Zhou, X. (2023) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3610909 · Cited by 36
A Shapelet-Based Framework for Unsupervised Multivariate Time Series Representation Learning
Liang, Z., Zhang, J., Liang, C., Wang, H., Liang, Z., & Pan, L. (2023) · Proceedings of the VLDB Endowment · doi:10.14778/3632093.3632103 · Cited by 33
“We illustrate how these methods are performed using the running examples in Figure 2, and we refer interested readers to [20, 48] for more details.”
MethodEkwaro-Osire, H., Ponugupati, S. L., Noman, A. A., Bode, D., & Thoben, K. (2025) · Industrial Artificial Intelligence · doi:10.1007/s44244-024-00021-x · Cited by 26
“[12–14].”
Kang, M., Lee, K. H., & Lee, Y. (2021) · Applied Sciences · doi:10.3390/app11083668 · Cited by 23
“Additionally, the data augmentation method was considered for time-series data, such as signal data [15].”
MethodXia, C., & Sugiura, Y. (2022) · Proceedings of the 2022 ACM International Symposium on Wearable Computers · doi:10.1145/3544794.3558460 · Cited by 22
“In addition, traditional IMU-based signal augmentation is a direct secondary modification of collected signals [20], which lacks specific evidence.”
BackgroundSemi-Supervised Learning for Wearable-based Momentary Stress Detection in the Wild
Yu, H., & Sano, A. (2023) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3596246 · Cited by 68
“To perform the consistency regularization, we adopted four types of data augmentation techniques for time-series data from [47], including jittering, scaling, time warping, and magnitude warping.”
MethodWearables-Driven Freeform Handwriting Authentication
Griswold-Steiner, I., Matovu, R., & Serwadda, A. (2019) · IEEE Transactions on Biometrics Behavior and Identity Science · doi:10.1109/tbiom.2019.2912401 · Cited by 25
“[32] applied a wide variety of augmentation operations (e.”
BackgroundMethodInfluentialA closer look at quality-aware runtime assessment of sensing models in multi-device environments
Min, C., Montanari, A., Mathur, A., & Kawsar, F. (2019) · ACM Conference on Embedded Networked Sensor Systems (SenSys) · doi:10.1145/3356250.3360043 · Cited by 30
“In [36], the authors introduced seven augmentation methods for IMU data on wearables: rotation, permutation, time-warping, scaling, magnitude-warping, jittering, and cropping.”
BackgroundMethodInfluentialRespWatch: Robust Measurement of Respiratory Rate on Smartwatches with Photoplethysmography
Dai, R., Lu, C., Avidan, M. S., & Kannampallil, T. (2021) · IEEE/ACM International Conference on Internet-of-Things Design and Implementation (IoTDI) · doi:10.1145/3450268.3453531 · Cited by 34
“Deep learning with wearable data has drawn great attentions in activity recognition [2, 40], Parkinson Disease monitoring [12, 51], atrial fibrillation detection [36, 46] and other mobile health applications [25, 54].”
BackgroundContrastive learning method for leak detection in water distribution networks
Liu, R., Zayed, T. M., & Xiao, R. (2024) · npj Clean Water · doi:10.1038/s41545-024-00406-6 · Cited by 27
“For time-series signals, the basic augmentation tool includes Gaussian noise,timeshiftaugmentation,pitchshiftaugmentation,jitter,andadjacent sample augmentation 34,35 .”
Deep generative cross-modal on-body accelerometer data synthesis from videos
Zhang, S., & Alshurafa, N. (2020) · Adjunct Proceedings of the 2020 ACM International Joint Conference on Pervasive and Ubiquitous Computing and Proceedings of the 2020 ACM International Symposium on Wearable Computers · doi:10.1145/3410530.3414329 · Cited by 21
“Sensor Data Synthesis: Besides the long standing problem for general time series data augmentation, recently there has been growing interest in sensor data augmentation techniques [2, 13, 17, 23, 24].”
BackgroundResultInfluentialOgunseiju, O. R., Akinniyi, A., Gonsalves, N. J., Khalid, M., & Akanmu, A. A. (2023) · Journal of Computing in Civil Engineering · doi:10.1061/jccee5.cpeng-5169 · Cited by 23
Abir, F. F., Chowdhury, M. E. H., Tapotee, M. I., Mushtak, A., Khandakar, A. A., Mahmud, S., & Hasan, A. (2023) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2023.106130 · Cited by 30
ConvBoost: Boosting ConvNets for Sensor-based Activity Recognition
Shao, S., Guan, Y., Zhai, B., Missier, P., & Plötz, T. (2023) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3596234 · Cited by 30
“(2) Experiment 2: we additionally add scaling operation [37] in the Resilient Layer.”
MethodA deep learning based health index construction method with contrastive learning
Wang, H., Li, X., Zhang, Z., Deng, X., & Jiang, W. R. (2023) · Reliability Engineering & System Safety · doi:10.1016/j.ress.2023.109799 · Cited by 23
Ashfaq, N., Khan, M. H., & Nisar, M. A. (2024) · Information · doi:10.3390/info15060343 · Cited by 26
Trustworthy AI-based Performance Diagnosis Systems for Cloud Applications: A Review
Xin, R., Wang, J., Chen, P., & Zhao, Z. (2025) · ACM Computing Surveys · doi:10.1145/3701740 · Cited by 30
“For time-series data, random warping in the time dimension [193] and frequency warping [90] are applied.”
A Hilbert Curve Based Representation of sEMG Signals for Gesture Recognition
Tsinganos, P., Cornelis, B., Cornelis, J. P., Jansen, B., & Skodras, A. N. (2019) · International Conference on Systems, Signals and Image Processing · doi:10.1109/iwssip.2019.8787290 · Cited by 26
“In addition, sEMG signals are augmented with the magnitude-warping method described in [37, 38].”
MethodLi, Y., Luo, J., Deng, S., & Zhou, G. (2023) · ACM Transactions on Sensor Networks · doi:10.1145/3599727 · Cited by 24
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.