Pith. sign in

REVIEW 4 major objections 5 minor 1 cited by

Freezing of Gait Detection Using Gramian Angular Fields and Federated Learning from Wearable Sensors

T0 review · 4 major / 5 minor · reviewed 2026-08-12 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read A system called FOGSense claims to detect freezing of gait in Parkinson's patients from the vertical axis of a single lower-back accelerometer, reaching a 96.27% F1-score.

desk verdict The headline F1 and FPR come from different experimental runs, and the channel is selected on the test set, so the paper's central claim is not supported by any single configuration. read the letter →

arxiv 2411.11764 v3 pith:XR6C35PQ submitted 2024-11-18 cs.LG eess.SP

Shovito Barua Soumma , S M Raihanul Alam , Rudmila Rahman , Umme Niraj Mahi , Abdullah Mamun , Sayyed Mostafa Mostafavi , Hassan Ghasemzadeh This is my paper · ORCID
classification cs.LGeess.SP
keywords freezingofgaitParkinson'sdiseaseGramianangularfieldfederatedlearningwearableaccelerometerconvolutionalneuralnetworkfalsepositiveratesingle-sensormonitoring
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

FOGSense claims that freezing of gait in Parkinson's disease can be detected accurately in free-living conditions from just one axis of a single lower-back accelerometer, using image-style encoding of the signal rather than hand-built features. The paper shows that the vertical acceleration axis alone reaches 96.25% accuracy and a 96.27% F1-score, with a false positive rate of 3.2%, and that a federated-learning version still reaches 90.47% F1 while keeping patient data on-device. If true, this would let FOG monitoring run on a single inexpensive wearable, reduce sensor-failure points, and avoid unnecessary intervention alerts, which matter because too many false alarms can desensitize patients.

What carries the argument

Gramian Angular Field (GAF) transformation: a 1D time series is normalized to $[-1,1]$, mapped to polar coordinates (angle $\theta_i = \cos^{-1}(x'_i)$, radius as normalized time index), and then the pairwise cosines of summed angles form a 2D matrix $\text{GASF}(i,j) = \cos(\theta_i + \theta_j)$ that preserves temporal order and correlations. This matrix is fed to a CNN as an image. The paper also uses an oversampling step that creates overlapping FOG windows in training, a weighted-averaging aggregation strategy for federated learning, and dynamic weight transfer, where per-channel models are ranked by test F1 and a failed channel is replaced by the best functional one.

What would settle it

Re-run the evaluation on held-out subjects with no overlapping FOG-window augmentation in training, and also compute F1 per FOG episode rather than per window; if the single-axis model's F1 falls far below 0.96 or the per-episode metric is materially worse, the headline claim is inflated.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central discovery is that a single channel—the vertical (AccV) axis of a waist-worn triaxial accelerometer—carries enough information to detect FOG episodes, and that converting 4-second windows of that signal into Gramian Angular Field images lets a multi-branch CNN outperform all three-channel and two-channel configurations. The reported numbers are 96.25% accuracy, 96.27% F1, 95.8% sensitivity, and a 3.2% false positive rate on a public dataset with 62 Parkinson's patients in uncontrolled settings. A federated variant using weighted averaging of client model updates reaches 86.98% accuracy and 90.47% F1 while preserving data privacy, and a dynamic weight-transfer scheme lets the system fall back to any functional channel if one sensor axis fails.

Load-bearing premise

It assumes that the 50%-overlap oversampling of FOG training windows mimics real test conditions and that three random splits of the 62 patients are representative, so the reported F1 is not inflated by memorized overlapping segments or an unrepresentative split.

Editorial extensions

If this is right

  • A waist-worn single-axis accelerometer is sufficient for FOG detection, so multi-sensor anklet or insole systems may be unnecessary for most monitoring.
  • Low false-positive rates make automated cueing interventions less likely to lose effectiveness from habituation.
  • Federated training lets the model improve across patients without moving raw sensor data off the phone, supporting long-term monitoring as symptoms evolve.
  • If any sensor axis fails, the system can switch to another single-channel model, so reliability does not depend on all channels working.
  • The GAF-to-CNN pipeline needs only about 11 seconds per training epoch for the single-channel model, making it feasible for resource-limited deployment.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The claim that one axis is enough suggests a testable extension: other wearable sites, such as the wrist or chest, may also have a single dominant axis, and the GAF-plus-CNN recipe could transfer with minimal changes.
  • The reported numbers are window-level; a per-episode or per-patient aggregation could show different performance, so that is where an independent reader should look first.
  • Because the oversampling of FOG windows is applied only to training, a benchmark with no overlapping augmentation and strictly held-out subjects would clarify how much of the F1 gain comes from the augmentation rather than from the GAF representation.
  • The federated result is simulated across synthetic clients; deployment on real phones with real battery and connectivity constraints is an open extension.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 5 minor

Summary. FOGSense is a FOG detection pipeline that converts lower-back accelerometer windows from the tdcsfog dataset into Gramian Angular Field images, classifies them with a multi-branch 2D CNN, and optionally trains under a federated averaging scheme over five simulated clients. The paper reports that the vertical-axis (AccV) single-channel model achieves 96.25% accuracy and 96.27% F1, that a federated variant reaches 90.47% F1, and that FOGSense reduces false-positive rate by 74.53% compared with state-of-the-art methods. A dynamic weight transfer mechanism is proposed to switch to a different channel when one fails, and code is made publicly available.

Significance. The intended contribution addresses a real clinical problem and has a plausible practical impact: single-sensor, low-false-positive FOG detection with a privacy-preserving learning option would be useful for home monitoring. The use of a public dataset, free-living conditions, and the release of code are clear strengths. However, the current evaluation does not support the headline quantitative claims: channel selection is performed on the test set, the abstract and Table IV combine metrics from different detection levels and channel configurations, and the comparison with prior work is confounded by an augmentation applied to FOGSense but not to the baselines. Because these problems are methodological and fixable, the result cannot be accepted as reported.

major comments (4)
  1. [Section IV.A, Algorithm 1, Table III] The best channel (AccV) is selected by evaluating models on the test set: Algorithm 1 trains one model per channel, sorts by test F1, and the paper then reports the top-ranked channel as the system. This is test-set model selection, so the reported 96.27% F1 is an optimistic estimate whose bias is unquantified. The authors should use a held-out validation split (or nested cross-validation) for channel selection and report the performance of the pre-selected channel on the test set.
  2. [Abstract, Tables II-IV] The headline figure mixes incompatible runs: Table III's AccV row is a window-level single-channel model with no FPR reported, while the 3.2% FPR quoted in the abstract and Table IV comes from the episode-level all-channel model in Table II (F1 90.86). Consequently, no single experiment supports the claim that using only AccV attains both 96.27% F1 and 3.2% FPR. The authors should present a single consistent configuration (e.g., the AccV model with its own window- and episode-level FPR) and revise the comparative claims accordingly.
  3. [Section III.D and Table IV] The proposed method trains on an augmented distribution in which FOG windows are oversampled with 50% overlap (DHWT), raising the FOG window count by 82%. Table IV compares FOGSense with methods that (with the possible exception of LIFT-PD [29]) are not described as using this augmentation. The reported gains may then be due in part to the augmented training data rather than to GAF or CNN choices. The authors should retrain the comparators under the same DHWT-augmented distribution, or report FOGSense without DHWT, and discuss the augmentation's contribution to the F1 improvement.
  4. [Section III.C] Results are averaged over three subject-based splits, but no variance (standard deviation, range, or confidence interval) is reported for any metric in Tables II-IV. Given the small number of patients (62) and the large-scale subject splits, the sensitivity of the headline F1 to the split should be quantified; the authors should report per-split metrics and a measure of dispersion for at least the main configurations.
minor comments (5)
  1. [Section III.A] The description of the federated clients is too vague: saying clients are assigned 'different combinations of gait data' does not specify whether the partition is by subject, by recording session, or by window, and without this detail the simulated non-IID setting cannot be assessed.
  2. [Section III.E, Equations (1)-(2)] The normalization description is inconsistent: Eq. (1) maps to [0,1] and Eq. (2) to [-1,1], but θ_i is defined from x'_i, yielding angles in [0,π/2] rather than the usual GAF angle range; the intended normalization should be clarified.
  3. [Section IV.A text after Table III] The dual-channel AccML-AccV result is described as 95.30% accuracy and 90.75% F1-score, while Table III reports 95.3 and 90.8; these numbers should be reconciled.
  4. [Table IV] The FOGSense row reports '–' for DFE but the FPR 3.2% is taken from the episode-level configuration; use consistent column definitions and indicate the detection level for each metric.
  5. [Abstract] The '22.2% improvement in F1-score' is not precisely consistent with the 0.963 versus 0.79 comparison in Table IV (a relative gain of about 21.9%); specify whether the gain is absolute or relative.

Circularity Check

0 steps flagged · score 2.0 of 10

No significant circularity: FOGSense's headline metrics are empirical measurements on a public benchmark, not derivations; the self-citations are minor and non-load-bearing.

full rationale

The paper is an empirical evaluation, not a derivation from first principles. GAF (Eqs. 1-4) is a standard time-series-to-image transform, and the federated learning uses standard FedAvg; neither reduces to the reported results by construction. The DHWT is attributed to the authors' prior LIFT-PD [29], but the procedure is fully specified in the text ('varying overlap sizes: no overlap (0%) for no-FOG windows and an increased 50% overlap for FOG windows'), so the citation is attribution rather than load-bearing proof. The comparison against LIFT-PD [29] is a self-citation, but it is one of several baselines, and the claimed 96.27% F1 is a measured value from the AccV model, not forced by that citation. Two protocol concerns are visible but are not circularity: Algorithm 1 selects the input channel by test F1 ('Sort the N models and channels in descending order of their test F-1 scores'), which is test-set model selection and can inflate the reported AccV F1 via winner's curse; and Table IV's FOGSense row mixes the AccV window-level SEN/PRE/F1 with the all-channel episode-level FPR from Table II, so the abstract's 'single sensor' + '3.2% FPR' claim is not supported by a single run. These are statistical-validity and reporting issues, not definitional equivalences between inputs and outputs. Therefore no significant circularity is present; the score of 2 reflects only the occurrence of minor, non-load-bearing self-citations.

Assumptions & free parameters 7 free parameters · 5 assumptions · 0 invented entities

The system's performance rests on hand-chosen preprocessing (4s windows, 64 Hz, GAF 64x64), the DHWT augmentation from the authors' prior work, hand-set CNN hyperparameters, and the post-hoc selection of the AccV channel. The axioms include standard GAF math and the domain assumption that the tdcsfog dataset with its annotations is representative of free-living FOG; the most fragile assumption is that DHWT-augmented training transfers to un-augmented test data.

free parameters (7)
  • Window length = 4 seconds
    Chosen based on prior work [36] to capture gait features; affects temporal context.
  • Downsampling rate = 64 Hz
    Set from prior work [29] to reduce computation.
  • GAF image size = 64x64
    Not justified in the paper; affects CNN input resolution.
  • DHWT FOG window overlap = 50%
    Inherited from the authors' prior LIFT-PD [29]; controls class balance and training set size.
  • CNN hyperparameters = 32/64/128 filters, dropout 20/40%, L2 0.001
    Hand-chosen architecture; fixed across experiments but not optimized or justified.
  • Number of federated clients = 5
    Simulated clients with different data splits; not grounded in real device counts or non-IID distributions.
  • Channel selection = AccV
    Selected after evaluating all channel configurations on the test set (Table III).
assumptions (5)
  • standard math GAF transformation equations (Eq. 1-3) are correct and preserve temporal structure
    Standard result from [15], used as preprocessing.
  • domain assumption The tdcsfog dataset's annotations and lower-back accelerometer placement are reliable for FOG detection
    The system is evaluated only on this public dataset; no external validation.
  • ad hoc to paper DHWT-augmented training distribution generalizes to un-augmented test windows
    The 50% overlap for FOG windows is an ad-hoc augmentation from the authors' prior work; if it causes episode memorization, F1 is inflated.
  • domain assumption Subject-based 70/10/20 split of 62 subjects yields a representative evaluation
    Only three random splits, no variance reported; test set size is about 12 subjects.
  • ad hoc to paper Simulated federated clients approximate real non-IID deployment
    The FL experiment uses simulated data distributions, not real edge devices; real-world validation is deferred.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Freezing of Gait Detection Using Gramian Angular Fields and Federated Learning from Wearable Sensors." pith.science (2026). https://pith.science/paper/XR6C35PQ

@misc{pith2026241111764,
  author       = {Pith},
  title        = {Pith review of: Freezing of Gait Detection Using Gramian Angular Fields and Federated Learning from Wearable Sensors},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/XR6C35PQ}},
  note         = {Machine review of arXiv:2411.11764}
}
read the original abstract

Freezing of gait (FOG) is a debilitating symptom of Parkinson's disease that impairs mobility and safety by increasing the risk of falls. An effective FOG detection system must be accurate, real-time, and deployable in free-living environments to enable timely interventions. However, existing detection methods face challenges due to (1) intra- and inter-patient variability, (2) subject-specific training, (3) using multiple sensors in FOG dominant locations (e.g., ankles) leading to high failure points, (4) centralized, non-adaptive learning frameworks that sacrifice patient privacy and prevent collaborative model refinement across populations and disease progression, and (5) most systems are tested in controlled settings, limiting their real-world applicability for continuous in-home monitoring. Addressing these gaps, we present FOGSense, a real-world deployable FOG detection system designed for uncontrolled, free-living conditions using only a single sensor. FOGSense uses Gramian Angular Field (GAF) transformations and privacy-preserving federated deep learning to capture temporal and spatial gait patterns missed by traditional methods with a low false positive rate. We evaluated our system using a public Parkinson's dataset collected in a free-living environment. FOGSense improves accuracy by 10.4% over a single-axis accelerometer, reduces failure points compared to multi-sensor systems, and demonstrates robustness to missing values. The federated architecture allows personalized model adaptation and efficient smartphone synchronization during off-peak hours, making it effective for long-term monitoring as symptoms evolve. Overall, FOGSense achieved a 22.2% improvement in F1-score and a 74.53% reduction in false positive rate compared to state-of-the-art methods, along with enhanced sensitivity for FOG episode detection.

Figures

Figures reproduced from arXiv: 2411.11764 by the authors.

Figure 1
Figure 1. Proposed distributed federated system design. [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 3
Figure 3. Class distribution normal and FOG events before (a) [PITH_FULL_IMAGE:figures/full_fig_p003_3.png] view at source ↗
Figure 4
Figure 4. GAF transformation of accelerometer signals. [PITH_FULL_IMAGE:figures/full_fig_p003_4.png] view at source ↗
Figures from the paper (3 more)
Figure 2
Figure 2. Figure 2: System architecture illustrating the federated learning workflow: local models are trained and uploaded by devices, [PITH_FULL_IMAGE:figures/full_fig_p004_2.png]
Figure 6
Figure 6. Figure 6: The multichannel CNN architecture. The proposed architecture ( [PITH_FULL_IMAGE:figures/full_fig_p004_6.png]
Figure 7
Figure 7. Figure 7: Comparison of channel configurations in the [PITH_FULL_IMAGE:figures/full_fig_p006_7.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Trustworthy AI in Digital Health: A Comprehensive Review of Robustness and Explainability

    cs.AI 2026-08 conditional novelty 3.0 of 10

    A survey of robustness and explainability methods for digital health AI, proposing a taxonomy and illustrating known XAI tools, without new empirical or theoretical results.

Reference graph

Works this paper leans on

40 extracted references · 32 canonical work pages · cited by 1 Pith paper

  1. [29]

    Self-supervised learning and opportunistic infer- ence for continuous monitoring of freezing of gait in parkinson’s disease,

    S. B. Soumma, K. Mangipudi, D. Peterson, S. Mehta, and H. Ghasemzadeh, “Self-supervised learning and opportunistic infer- ence for continuous monitoring of freezing of gait in parkinson’s disease,” arXiv preprint arXiv:2410.21326, 2024

    arXiv

  2. [1]

    J. Brederecke, “Freezing of gait prediction from accelerometer data using a simple 1d-convolutional neural network–8th place solution for kaggle’s parkinson’s freezing of gait prediction competition,” arXiv preprint arXiv:2307.03475, 2023

    work page Pith review arXiv 2023

  3. [2]

    Freezing of gait in parkinson’s disease: risk factors, their interactions, and associated nonmotor symptoms,

    D. G. Lichter, R. H. B. Benedict, and L. A. Hershey, “Freezing of gait in parkinson’s disease: risk factors, their interactions, and associated nonmotor symptoms,” Parkinson’s Disease, vol. 2021, no. 1, p. 8857204, 2021

    work page 2021

  4. [3]

    Detecting freezing of gait with a tri-axial accelerometer in parkinson’s disease patients,

    C. Ahlrichs, A. Sam `a, M. Lawo, J. Cabestany, D. Rodr ´ıguez-Mart´ın, C. P´erez-L´opez, D. Sweeney, L. R. Quinlan, G. `O. Laighin, T. Couni- han, et al., “Detecting freezing of gait with a tri-axial accelerometer in parkinson’s disease patients,” Medical & biological engineering & computing, vol. 54, pp. 223–233, 2016

    work page 2016

  5. [4]

    Prediction of freezing of gait in parkinson’s disease using wearables and machine learning,

    L. Borz `ı, I. Mazzetta, A. Zampogna, A. Suppa, G. Olmo, and F. Irrera, “Prediction of freezing of gait in parkinson’s disease using wearables and machine learning,” Sensors, vol. 21, no. 2, p. 614, 2021

    2021

  6. [5]

    Gait analysis using wearable sensors,

    W. Tao, T. Liu, R. Zheng, and H. Feng, “Gait analysis using wearable sensors,” Sensors, vol. 12, no. 2, pp. 2255–2283, 2012

    2012

  7. [6]

    Cyclepro: A robust framework for domain-agnostic gait cycle detec- tion,

    Y . Ma, Z. Esna Ashari, M. Pedram, N. Amini, D. Tarquinio, K. Nouri- Mahdavi, M. Pourhomayoun, R. D. Catena, and H. Ghasemzadeh, “Cyclepro: A robust framework for domain-agnostic gait cycle detec- tion,” IEEE Sensors Journal, vol. 19, no. 10, pp. 3751–3762, 2019

    work page 2019

  8. [7]

    Smartsock: A wearable platform for context-aware assessment of ankle edema,

    R. Fallahzadeh, M. Pedram, and H. Ghasemzadeh, “Smartsock: A wearable platform for context-aware assessment of ankle edema,” in 2016 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), pp. 6302–6306, IEEE, 2016

    work page 2016

Show all 40 references
  1. [8]

    Gait-guard: Turn-aware freezing of gait detection for non-intrusive intervention systems,

    K. Koltermann, J. Clapham, G. Blackwell, W. Jung, E. N. Burnet, Y . Gao, H. Shao, L. Cloud, I. Pretzer-Aboff, and G. Zhou, “Gait-guard: Turn-aware freezing of gait detection for non-intrusive intervention systems,” in 2024 IEEE/ACM Conference on Connected Health: Applications,...

    2024

  2. [9]

    Detection of freezing of gait in parkinson’s dis- ease from foot-pressure sensing insoles using a temporal convolutional neural network,

    J.-M. Park, C.-W. Moon, B. C. Lee, E. Oh, J. Lee, W.-J. Jang, K. H. Cho, and S.-H. Lee, “Detection of freezing of gait in parkinson’s dis- ease from foot-pressure sensing insoles using a temporal convolutional neural network,” Frontiers in aging neuroscience, vol. 16, p. 1437707, 2024

    2024

  3. [10]

    Designing deep neural networks robust to sensor failure in mobile health environ- ments,

    A. Mamun, S. I. Mirzadeh, and H. Ghasemzadeh, “Designing deep neural networks robust to sensor failure in mobile health environ- ments,” in 2022 44th Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC), pp. 2442–2446, IEEE, 2022

    2022

  4. [11]

    Adaptive detection in real-time gait analysis through the dynamic gait event identifier,

    Y . Liu, X. Liu, Q. Zhu, Y . Chen, Y . Yang, H. Xie, Y . Wang, and X. Wang, “Adaptive detection in real-time gait analysis through the dynamic gait event identifier,” Bioengineering, vol. 11, no. 8, p. 806, 2024

    2024

  5. [12]

    Wearable- sensor-based detection and prediction of freezing of gait in parkinson’s disease: a review,

    S. Pardoel, J. Kofman, J. Nantel, and E. D. Lemaire, “Wearable- sensor-based detection and prediction of freezing of gait in parkinson’s disease: a review,” Sensors, vol. 19, no. 23, p. 5141, 2019

    2019

  6. [13]

    Domain-informed label fusion surpasses llms in free-living activity classification (stu- dent abstract),

    S. B. Soumma, A. Mamun, and H. Ghasemzadeh, “Domain-informed label fusion surpasses llms in free-living activity classification (stu- dent abstract),” Proceedings of the AAAI Conference on Artificial Intelligence, vol. 39, pp. 29495–29497, Apr. 2025

    2025

  7. [14]

    Ecg classification using deep cnn and gramian angular field,

    Y . Elmir, Y . Himeur, and A. Amira, “Ecg classification using deep cnn and gramian angular field,” in 2023 IEEE Ninth International Conference on Big Data Computing Service and Applications (BigDataService), pp. 137–141, IEEE, 2023

    2023

  8. [15]

    Imaging time-series to improve classification and imputation,

    Z. Wang and T. Oates, “Imaging time-series to improve classification and imputation,” arXiv preprint arXiv:1506.00327, 2015

    2015 arXiv

  9. [16]

    Gaformer: Wearable imu-based human activity recognition with gramian angular field and transformer,

    T.-H. Le, T.-K. Nguyen, T.-K. Tran, T.-H. Tran, and C. Pham, “Gaformer: Wearable imu-based human activity recognition with gramian angular field and transformer,” in2023 Asia Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC), pp. 2...

    2023

  10. [17]

    Ai benchmark: Running deep neural networks on android smartphones,

    A. Ignatov, R. Timofte, W. Chou, K. Wang, M. Wu, T. Hartley, and L. Van Gool, “Ai benchmark: Running deep neural networks on android smartphones,” in Proceedings of the European Conference on Computer Vision (ECCV) Workshops, pp. 0–0, 2019

    2019

  11. [18]

    On-device machine learning for diagnosis of parkinson’s disease from hand drawn artifacts,

    S. V . P. Venkata, S. Sabat, C. A. Deshpande, A. Arefeen, D. Peterson, and H. Ghasemzadeh, “On-device machine learning for diagnosis of parkinson’s disease from hand drawn artifacts,” in 2022 IEEE-EMBS International Conference on Wearable and Implantable Body Sensor Networks (...

    2022

  12. [19]

    Convergence of edge computing and deep learning: A comprehensive survey,

    X. Wang, Y . Han, C. Wang, Q. Zhao, X. Chen, and M. Chen, “Convergence of edge computing and deep learning: A comprehensive survey,” IEEE Communications Surveys & Tutorials, vol. 22, no. 2, pp. 869–904, 2020

    2020

  13. [20]

    Federated learning for edge networks: Resource opti- mization and incentive mechanism,

    L. U. Khan, S. R. Pandey, N. H. Tran, W. Saad, Z. Han, M. N. Nguyen, and C. S. Hong, “Federated learning for edge networks: Resource opti- mization and incentive mechanism,” IEEE Communications Magazine, vol. 58, no. 10, pp. 88–93, 2020

    2020

  14. [21]

    Prediction of freezing of gait in parkinson’s from physio- logical wearables: an exploratory study,

    S. Mazilu, A. Calatroni, E. Gazit, A. Mirelman, J. M. Hausdorff, and G. Tr¨oster, “Prediction of freezing of gait in parkinson’s from physio- logical wearables: an exploratory study,” IEEE journal of biomedical and health informatics, vol. 19, no. 6, pp. 1843–1854, 2015

    2015

  15. [22]

    Hfl: Hybrid fed- erated learning for healthcare applications on edge devices,

    J. Luo, B. Song, M. Wen, H. Dong, and Y . Zhang, “Hfl: Hybrid fed- erated learning for healthcare applications on edge devices,” in 2020 IEEE International Conference on Big Data (Big Data), pp. 3172– 3181, IEEE, 2020

    2020

  16. [23]

    Federated learning: Challenges, methods, and future directions,

    T. Li, A. K. Sahu, A. Talwalkar, and V . Smith, “Federated learning: Challenges, methods, and future directions,” IEEE signal processing magazine, vol. 37, no. 3, pp. 50–60, 2020

    2020

  17. [24]

    The future of digital health with federated learning,

    N. Rieke, J. Hancox, W. Li, F. Milletari, H. R. Roth, S. Albarqouni, S. Bakas, M. N. Galtier, B. A. Landman, K. Maier-Hein, et al., “The future of digital health with federated learning,” NPJ digital medicine, vol. 3, no. 1, pp. 1–7, 2020

    2020

  18. [25]

    A hybrid spatio-temporal model for detection and severity rating of parkinson’s disease from gait data,

    A. Zhao, L. Qi, J. Li, J. Dong, and H. Yu, “A hybrid spatio-temporal model for detection and severity rating of parkinson’s disease from gait data,” Neurocomputing, vol. 315, pp. 1–8, 2018

    2018

  19. [26]

    Dyskinesia and motor state detection in parkinson’s disease patients with a single movement sensor,

    A. Sama, C. P ´erez-L´opez, J. Romagosa, D. Rodriguez-Martin, A. Catala, J. Cabestany, D. A. Perez-Martinez, and A. Rodr ´ıguez- Molinero, “Dyskinesia and motor state detection in parkinson’s disease patients with a single movement sensor,” in 2012 Annual International Confere...

    2012

  20. [27]

    Towards real-time prediction of freezing of gait in patients with parkinson’s disease: a novel deep one- class classifier,

    N. Naghavi and E. Wade, “Towards real-time prediction of freezing of gait in patients with parkinson’s disease: a novel deep one- class classifier,” IEEE Journal of Biomedical and Health Informatics, vol. 26, no. 4, pp. 1726–1736, 2021

    2021

  21. [28]

    Fog-finder: Real-time freezing of gait detection and treatment,

    K. Koltermann, W. Jung, G. Blackwell, A. Pinney, M. Chen, L. Cloud, I. Pretzer-Aboff, and G. Zhou, “Fog-finder: Real-time freezing of gait detection and treatment,” in Proceedings of the 8th ACM/IEEE International Conference on Connected Health: Applications, Systems and Engin...

    2023

  22. [30]

    Prediction of freezing of gait based on self-supervised pretraining via contrastive learning,

    Y . Xia, H. Sun, B. Zhang, Y . Xu, and Q. Ye, “Prediction of freezing of gait based on self-supervised pretraining via contrastive learning,” Biomedical Signal Processing and Control, vol. 89, p. 105765, 2024

    2024

  23. [31]

    Wearable sensors for parkinson’s disease: which data are worth collecting for training symptom detection models,

    L. Lonini, A. Dai, N. Shawen, T. Simuni, C. Poon, L. Shimanovich, M. Daeschler, R. Ghaffari, J. A. Rogers, and A. Jayaraman, “Wearable sensors for parkinson’s disease: which data are worth collecting for training symptom detection models,” NPJ digital medicine, vol. 1, no. 1, ...

    2018

  24. [32]

    Fedhealth: A federated transfer learning framework for wearable healthcare,

    Y . Chen, X. Qin, J. Wang, C. Yu, and W. Gao, “Fedhealth: A federated transfer learning framework for wearable healthcare,” IEEE Intelligent Systems, vol. 35, no. 4, pp. 83–93, 2020

    2020

  25. [33]

    Waf- fle: Weighted averaging for personalized federated learning,

    M. Beaussart, F. Grimberg, M.-A. Hartley, and M. Jaggi, “Waf- fle: Weighted averaging for personalized federated learning,” arXiv preprint arXiv:2110.06978, 2021

    2021 arXiv

  26. [34]

    Parkinson’s freezing of gait prediction

    A. Howard, amit salomon, eran gazit, J. Hausdorff, L. Kirsch, Maggie, P. Ginis, R. Holbrook, and Y . F. Karim, “Parkinson’s freezing of gait prediction.” https://kaggle.com/competitions/ tlvmc-parkinsons-freezing-gait-prediction , 2023. Kaggle

    2023

  27. [35]

    Deep, convolutional, and recurrent models for human activity recognition using wearables,

    N. Y . Hammerla, S. Halloran, and T. Pl ¨otz, “Deep, convolutional, and recurrent models for human activity recognition using wearables,” arXiv preprint arXiv:1604.08880, 2016

    2016 arXiv

  28. [36]

    Prediction of freezing of gait in parkinson’s disease using statistical inference and lower–limb acceleration data,

    N. Naghavi and E. Wade, “Prediction of freezing of gait in parkinson’s disease using statistical inference and lower–limb acceleration data,” IEEE transactions on neural systems and rehabilitation engineering, vol. 27, no. 5, pp. 947–955, 2019

    2019

  29. [37]

    Autoencoder with gramian angular summation field for anomaly detection in multivariate time series data,

    U. Yokkampon, A. Mowshowitz, S. Chumkamon, and E. Hayashi, “Autoencoder with gramian angular summation field for anomaly detection in multivariate time series data,” Journal of Advances in Artificial Life Robotics, vol. 2, no. 4, pp. 206–210, 2022

    2022

  30. [38]

    A novel approach and application of time series to image transformation methods on classification of underwater objects,

    A. C. Buz, M. U. Demirezen, and U. Yavano ˘glu, “A novel approach and application of time series to image transformation methods on classification of underwater objects,” Gazi M¨uhendislik Bilimleri Dergisi, vol. 7, no. 1, pp. 1–11, 2020

    2020

  31. [39]

    Real-time patient adaptivity for freezing of gait classification through semi-supervised neural networks,

    V . Mikos, C.-H. Heng, A. Tay, N. S. Y . Chia, K. M. L. Koh, D. M. L. Tan, and W. L. Au, “Real-time patient adaptivity for freezing of gait classification through semi-supervised neural networks,” in 2017 16th IEEE International Conference on Machine Learning and Applications ...

    2017

  32. [40]

    Real-time detection of freezing of gait in parkinson’s disease using multi-head convolutional neural networks and a single inertial sensor,

    L. Borz `ı, L. Sigcha, D. Rodr ´ıguez-Mart´ın, and G. Olmo, “Real-time detection of freezing of gait in parkinson’s disease using multi-head convolutional neural networks and a single inertial sensor,” Artif. Intell. Med., vol. 135, jan 2023

    2023

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.