CNN-GRU-KAN: A Novel Multi-Branch Framework for Parkinson’s Disease Detection Based on Gait Classification
Bioengineering · 10.08.2026 · 13 (8) · 905
DOI: 10.3390/bioengineering13080905 ↗Kurz gefasst
Fachartikel in der Zeitschrift Bioengineering, online erschienen am 10.08.2026. Verfasst von Xingkai Fu und 2 weiteren Autor:innen. Beteiligte Einrichtung: College of Physics and Electronic Information Engineering, Zhejiang Normal University, Jinhua 321004, China. Die Arbeit nennt 43 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Parkinson’s disease (PD) diagnosis and severity assessment increasingly rely on gait analysis. To capture complex gait dynamics from limited sensor data, we propose a novel multi-branch framework integrating a one-dimensional convolutional neural network (1D-CNN), a bidirectional gated recurrent unit (Bi-GRU), and Kolmogorov–Arnold networks (KANs) (CNN-GRU-KAN) utilizing 18-channel vertical ground reaction force (VGRF) signals. Each module serves a distinct clinical purpose: the 1D-CNN branch with squeeze-and-excitation (SE) attention extracts localized spatial plantar pressure patterns, while the Bi-GRU branch with temporal attention captures long-range rhythm abnormalities. Crucially, the KAN serves as the classification head. By utilizing learnable B-spline functions instead of traditional fixed activations, KAN adaptively models the highly non-linear boundaries between healthy controls and varying PD severities, effectively mitigating overfitting. Under rigorous subject-independent cross-validation, our model achieves 98.43% accuracy for binary PD detection and 93.46% for five-class UPDRS severity grading. These results highlight the framework’s strong potential for low-cost, unobtrusive clinical tracking and home monitoring.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- College of Physics and Electronic Information Engineering, Zhejiang Normal University, Jinhua 321004, China.
- College of Engineering and Information Technology, Emirates International University, Sana'a 16881, Yemen.
ORCID-Profile: Xingkai Fu, Minlan Jiang, Mohammed A A Al-Qaness
Zitieren
APA
Fu, X., Jiang, M., & Al-Qaness, M. A. A. (2026). CNN-GRU-KAN: A Novel Multi-Branch Framework for Parkinson’s Disease Detection Based on Gait Classification. Bioengineering, 13(8), 905. https://doi.org/10.3390/bioengineering13080905
Vancouver
Fu X, Jiang M, Al-Qaness MAA. CNN-GRU-KAN: A Novel Multi-Branch Framework for Parkinson’s Disease Detection Based on Gait Classification. Bioengineering. 2026;13(8):905. doi:10.3390/bioengineering13080905
BibTeX
@article{Fu2026,
author = {Fu, Xingkai and Jiang, Minlan and Al-Qaness, Mohammed A A},
title = {CNN-GRU-KAN: A Novel Multi-Branch Framework for Parkinson’s Disease Detection Based on Gait Classification},
journal = {Bioengineering},
year = {2026},
volume = {13},
number = {8},
pages = {905},
doi = {10.3390/bioengineering13080905}
}
EUVIMED