RBF-SMC-based closed-loop DBS control for a Parkinsonian BG-TH neuronal network
IET Conference Proceedings · 01.11.2026 · 2026 (22) · 436-441
DOI: 10.1049/icp.2026.4488 ↗Kurz gefasst
Fachartikel in der Zeitschrift IET Conference Proceedings, erschienen am 01.11.2026. Verfasst von Wenyu Zhang und 2 weiteren Autor:innen.
Abstract (Originalsprache)
Parkinson's disease (PD) motor symptoms are closely related to abnormal discharge patterns in the basal ganglia-thalamic (BG-TH) circuit. To improve closed-loop deep brain stimulation (DBS) regulation under model uncertainty and nonlinear disturbances, this study develops an RBF-SMC-based control strategy for a Parkinsonian BG-TH neuronal network. Based on the Rubin model, TH, STN, GPe, and GPi neuronal populations are established under healthy and PD conditions. The healthy TH membrane potential is selected as the reference trajectory. A radial basis function neural network is used to approximate uncertain nonlinear terms online, while sliding mode control enhances robustness and finite-time convergence. Simulation results show that the PD state increases STN and GPi activity, decreases GPe activity, and strengthens pathological beta-band oscillations. Compared with SMC and ATSMC, the proposed RBF-SMC controller achieves more accurate TH membrane-potential tracking and lower RMSE and control-efficiency indices, indicating improved DBS regulation performance without substantially increasing stimulation cost.
Zitieren
APA
Zhang, W., Li, X., & Yuan, C. (2026). RBF-SMC-based closed-loop DBS control for a Parkinsonian BG-TH neuronal network. IET Conference Proceedings, 2026(22), 436-441. https://doi.org/10.1049/icp.2026.4488
Vancouver
Zhang W, Li X, Yuan C. RBF-SMC-based closed-loop DBS control for a Parkinsonian BG-TH neuronal network. IET Conference Proceedings. 2026;2026(22):436-441. doi:10.1049/icp.2026.4488
BibTeX
@article{Zhang2026,
author = {Zhang, Wenyu and Li, Xiangyu and Yuan, Chunhua},
title = {RBF-SMC-based closed-loop DBS control for a Parkinsonian BG-TH neuronal network},
journal = {IET Conference Proceedings},
year = {2026},
volume = {2026},
number = {22},
pages = {436--441},
doi = {10.1049/icp.2026.4488}
}
EUVIMED