Department of Biomedical Engineering, University of California, Irvine, CA 92697, USA.
J Neural Eng. 2012 Oct;9(5):056016. doi: 10.1088/1741-2560/9/5/056016. Epub 2012 Sep 25.
Spinal cord injury (SCI) often leaves affected individuals unable to ambulate. Electroencephalogram (EEG) based brain-computer interface (BCI) controlled lower extremity prostheses may restore intuitive and able-body-like ambulation after SCI. To test its feasibility, the authors developed and tested a novel EEG-based, data-driven BCI system for intuitive and self-paced control of the ambulation of an avatar within a virtual reality environment (VRE).
Eight able-bodied subjects and one with SCI underwent the following 10-min training session: subjects alternated between idling and walking kinaesthetic motor imageries (KMI) while their EEG were recorded and analysed to generate subject-specific decoding models. Subjects then performed a goal-oriented online task, repeated over five sessions, in which they utilized the KMI to control the linear ambulation of an avatar and make ten sequential stops at designated points within the VRE.
The average offline training performance across subjects was 77.2 ± 11.0%, ranging from 64.3% (p = 0.001 76) to 94.5% (p = 6.26 × 10(-23)), with chance performance being 50%. The average online performance was 8.5 ± 1.1 (out of 10) successful stops and 303 ± 53 s completion time (perfect = 211 s). All subjects achieved performances significantly different than those of random walk (p < 0.05) in 44 of the 45 online sessions.
By using a data-driven machine learning approach to decode users' KMI, this BCI-VRE system enabled intuitive and purposeful self-paced control of ambulation after only 10 minutes training. The ability to achieve such BCI control with minimal training indicates that the implementation of future BCI-lower extremity prosthesis systems may be feasible.
脊髓损伤（SCI）常导致患者丧失行走能力。基于脑电图（EEG）的脑机接口（BCI）控制的下肢假肢可以在 SCI 后恢复直观的、类似正常身体的行走能力。为了验证其可行性，作者开发并测试了一种新颖的基于 EEG 的、数据驱动的 BCI 系统，用于在虚拟现实环境（VRE）中直观地、自主控制虚拟化身的行走。
8 名健康受试者和 1 名 SCI 受试者进行了以下 10 分钟的训练：受试者在记录和分析 EEG 以生成特定于个体的解码模型的同时，交替进行静息和行走运动想象（KMI）。然后，受试者在五个会话中重复进行一个目标导向的在线任务，他们利用 KMI 控制虚拟化身的线性行走，并在 VRE 内的指定点进行十次连续的停顿。
平均而言，所有受试者的离线训练表现为 77.2 ± 11.0%，范围从 64.3%（p = 0.00176）到 94.5%（p = 6.26×10^-23），而随机表现为 50%。平均在线表现为 10 次成功停顿中的 8.5 ± 1.1 次，完成时间为 303 ± 53 秒（完美= 211 秒）。在 45 次在线会话中的 44 次，所有受试者的表现都显著优于随机行走（p < 0.05）。
通过使用数据驱动的机器学习方法对用户的 KMI 进行解码，该 BCI-VRE 系统仅在 10 分钟的训练后即可实现直观的、有目的的自主行走控制。在最小训练量的情况下实现这种 BCI 控制的能力表明，未来 BCI-下肢假肢系统的实施可能是可行的。
