Non-Invasive Wearable Systems for Parkinsonian Tremor Suppression: A Review of Machine Learning Monitoring and Hardware Modulation Pipelines
Undergraduate Research in Natural and Clinical Science and Technology Journal · 15.07.2026 · 1-16
DOI: 10.26685/urncst.983 ↗Kurz gefasst
Fachartikel in der Zeitschrift Undergraduate Research in Natural and Clinical Science and Technology Journal, online erschienen am 15.07.2026. Verfasst von Elena Osipyan und 5 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Introduction
Parkinson's disease (PD) is a progressive neurodegenerative disease. Tremors are one of the most prevalent and functionally disabling motor symptoms of PD. Non-invasive wearable technologies for tremor monitoring and suppression have renewed interest, as there are limitations of pharmacological therapies and surgical interventions. This literature review examines advances in wearable hardware systems and machine learning (ML) approaches for PD tremor management.
Methods
A structured literature review was conducted using PubMed, IEEE Xplore, ScienceDirect, PMC, Frontiers, and Google Scholar for peer-reviewed studies published between 2017 and 2025. Inclusion criteria targeted non-invasive wearable systems for PD tremor monitoring or suppression. A total of six ML-focused studies and six hardware-focused studies met eligibility criteria. Qualitative synthesis examined trends in software and hardware design.
Results
Six studies met the inclusion criteria for both hardware and ML approaches. Across the ML studies, multimodal and inertial sensing is associated with more reliable performance in PD tremor detection, with reported accuracies above 80% in gait analysis tasks, 96.98% accuracy with high sensitivity (91.06%) and specificity (98.56%) in tremor detection under controlled conditions. The qualitative analysis of hardware-based studies showed that passive and semi-active wearable systems were the most explored intervention types. These devices reduced tremor-related kinematic features by 55-80%. Validation, across both ML and hardware studies, was often limited by small, demographically homogenous PD patient cohorts, simulated tremors, or healthy control populations.
Discussion
For both ML and wearable hardware approaches, validation remains heterogeneous, with small sample sizes and inconsistent outcome measures limiting comparability. Trade-offs between personalisation, device complexity, and wearability were evident across passive, semi-active, and active designs. Inclusion of standardised, clinically meaningful evaluation metrics, both in ML studies and in hardware studies, and implementing clear benchmarks for tremor modulation will increase cross-study comparability and clinical implementation.
Conclusion
Overall, current research includes a wide range of pre-clinical machine learning and hardware approaches for Parkinsonian tremor detection and modulation. Future work should focus on larger, clinically representative cohorts, standardised outcome metrics, and real-world validation to support meaningful integration of wearable hardware and ML for clinical translation.
Zitieren
APA
Osipyan, E., Azam, A., Rehman, Z., Sribalan, S., Duraisingh, J., & Youssef, R. (2026). Non-Invasive Wearable Systems for Parkinsonian Tremor Suppression: A Review of Machine Learning Monitoring and Hardware Modulation Pipelines. Undergraduate Research in Natural and Clinical Science and Technology Journal, 1-16. https://doi.org/10.26685/urncst.983
Vancouver
Osipyan E, Azam A, Rehman Z, Sribalan S, Duraisingh J, Youssef R. Non-Invasive Wearable Systems for Parkinsonian Tremor Suppression: A Review of Machine Learning Monitoring and Hardware Modulation Pipelines. Undergraduate Research in Natural and Clinical Science and Technology Journal. 2026:1-16. doi:10.26685/urncst.983
BibTeX
@article{Osipyan2026,
author = {Osipyan, Elena and Azam, Aliza and Rehman, Zayn and Sribalan, Shrena and Duraisingh, Joseph and Youssef, Robert},
title = {Non-Invasive Wearable Systems for Parkinsonian Tremor Suppression: A Review of Machine Learning Monitoring and Hardware Modulation Pipelines},
journal = {Undergraduate Research in Natural and Clinical Science and Technology Journal},
year = {2026},
pages = {1--16},
doi = {10.26685/urncst.983}
}
EUVIMED