Investigating Accuracy Versus Gait Strides Tradeoff in Parkinson's Detection
IEEE Instrumentation & Measurement Magazine · 2026-11 · 29 (8) · 43-48
DOI: 10.1109/mim.2026.11224907 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Instrumentation & Measurement Magazine, 2026-11. Verfasst von Devashish Thapliyal und 2 weiteren Autor:innen. Die Arbeit nennt 10 Literaturangaben.
Abstract (Originalsprache)
Gait analysis helps detect Parkinson's disease (PD); however, long walking tests lack standardization. This study determines the minimum steps needed for accurate PD classification. We tested a Tensor Decomposition-based support vector machine model on sequences of strides ranging from 10 to 100. Statistical testing confirmed that just 32 strides of straightline walking achieved 88.11% ± 1.59% accuracy. This is statistically non-inferior to using a full100-stride baseline, which attained an accuracy of 89.97% ± 1.91%.. These findings prove that highly accurate PD detection is possible with a brief straightline walking test, significantly streamlining clinical data collection without sacrificing performance.
Zitieren
APA
Thapliyal, D., Pandey, B., & Joshi, D. (2026). Investigating Accuracy Versus Gait Strides Tradeoff in Parkinson's Detection. IEEE Instrumentation & Measurement Magazine, 29(8), 43-48. https://doi.org/10.1109/mim.2026.11224907
Vancouver
Thapliyal D, Pandey B, Joshi D. Investigating Accuracy Versus Gait Strides Tradeoff in Parkinson's Detection. IEEE Instrumentation & Measurement Magazine. 2026;29(8):43-48. doi:10.1109/mim.2026.11224907
BibTeX
@article{Thapliyal2026,
author = {Thapliyal, Devashish and Pandey, Bhaskar and Joshi, Deepak},
title = {Investigating Accuracy Versus Gait Strides Tradeoff in Parkinson's Detection},
journal = {IEEE Instrumentation & Measurement Magazine},
year = {2026},
volume = {29},
number = {8},
pages = {43--48},
doi = {10.1109/mim.2026.11224907}
}
EUVIMED