Search

Real-Time EEG Intention Decoding System Based on Edge Computing: Low-Latency Optimization for VR Neurorehabilitation

Author(s)

Jiayi Shen, Luyan Shen, Sitong Ruan, Zhihui Xu, Jingyi Xu*

Affiliation(s)

Artificial Intelligence College, Zhejiang Dongfang Polytechnic, Wenzhou, Zhejiang, China *Corresponding Author

Abstract

To address the high latency, strong network dependence, and significant privacy and security risks of traditional cloud-based brain-computer interface (BCI) systems in virtual reality (VR) neurorehabilitation applications, this paper proposes a real-time EEG intention decoding system based on edge computing. The system offloads EEG signal preprocessing and feature extraction modules to the local Neural Processing Unit (NPU) chip of the VR headset. By designing a multithreaded asynchronous scheduling algorithm to achieve efficient parallel processing of computing tasks, it successfully compresses the end-to-end EEG→VR feedback latency to 21.7 ± 1.2 ms, which is far below the 250 ms critical threshold required for neural plasticity closed loops. Meanwhile, this paper presents a lightweight hybrid decoding model that integrates temporal-frequency-spatial features. In motor imagery tasks involving 12 healthy subjects and 8 stroke patients, the average decoding accuracies reached 92.3% and 85.7% respectively, significantly alleviating the "intention misinterpretation" problem of traditional systems. Preliminary clinical validation demonstrates that the system can effectively construct a "thought-action-feedback" neural plasticity closed loop, providing an efficient, safe, and convenient new rehabilitation training solution for patients with post-stroke motor dysfunction.

Keywords

Edge Computing; Real-Time EEG Intention Decoding; VR Neurorehabilitation; Low-Latency Optimization; Neural Plasticity; Multithreaded Asynchronous Scheduling

References

[1] Wolpaw J R, Birbaumer N, McFarland D J, et al. Brain-computer interfaces for communication and control. Clinical Neurophysiology, 2002, 113(6): 767-791. [2] Shadmehr R, Smith M A, Krakauer J W. Error correction, sensory prediction, and adaptation in motor control. Annual Review of Neuroscience, 2010, 33: 89-108. [3] Xu G H, Zhang J, Wang J. Advances in the application of brain-computer interface technology in stroke rehabilitation. Chinese Journal of Rehabilitation Medicine, 2023, 38(2): 267-272. [4] Shi W, Cao J, Zhang Q, et al. Edge computing: Vision and challenges. IEEE Internet of Things Journal, 2016, 3(5): 637-646. [5] Yang X K, Li Y Q, Wu W. Advances in edge computing technology for brain-computer interfaces. Acta Automatica Sinica, 2024, 50(3): 567-582. [6] Leeb R, Friedman D, Müller-Putz G R, et al. Towards brain-computer interface driven virtual reality systems: Design and implementation. Presence: Teleoperators and Virtual Environments, 2007, 16(3): 290-307. [7] Ang K K, Chin Z Y, Zhang H, et al. Filter bank common spatial pattern (FBCSP) in brain-computer interface//2008 IEEE International Joint Conference on Neural Networks (IEEE World Congress on Computational Intelligence). IEEE, 2008: 2390-2397. [8] Zhang X, Wang Y, Jin J, et al. A lightweight CNN-based EEG classification model for edge computing. IEEE Transactions on Neural Systems and Rehabilitation Engineering, 2022, 30: 1234-1243. [9] Chen X G, Wang L, Liu C. Application of multithreaded asynchronous scheduling in real-time brain-computer interface systems. Computer Engineering and Applications, 2023, 59(12): 189-195. [10] Li Y, Wang H, Zhang L, et al. EdgeBCI: An edge computing framework for low-latency brain-computer interfaces. IEEE Internet of Things Journal, 2024, 11(8): 13456-13467.