Leng Jiancai, Yu Xin, Wang Chongfeng, Zhao Jinzhao, Zhu Jianqun, Chen Xinyi, Zhu Zhaoxin, Jiang Xiuquan, Zhao Jiaqi, Feng Chao, Yang Qingbo, Li Jianfei, Jiang Lin, Xu Fangzhou, Zhang Yang
International School for Optoelectronic Engineering, Qilu University of Technology (Shandong Academy of Sciences), No.3501, Daxue Road Changqing District, Jinan, 250353 Shandong China.
School of Mathematics and Statistics, Qilu University of Technology (Shandong Academy of Sciences), No.3501, Daxue Road Changqing District, Jinan, 250353 Shandong China.
Cogn Neurodyn. 2024 Oct;18(5):3015-3029. doi: 10.1007/s11571-024-10136-7. Epub 2024 Jun 10.
Spinal cord injury (SCI), which is the injury of the spinal cord site resulting in motor dysfunction, has prompted the use of motor imagery (MI)-based brain computer interface (BCI) systems for motor function reconstruction. However, analyzing electroencephalogram signals and brain function mechanisms for SCI patients is challenging. This is due to their low signal-to-noise ratio and high variability. We propose using the phase locking value (PLV) to construct the brain network in α and β rhythms for both SCI patients and healthy individuals. This approach aims to analyze the changes in brain network connectivity and brain function mechanisms following SCI. The results show that the connection strength of the α rhythm in the healthy control (HC) group is stronger than that in the SCI group, and the connection strength in the β rhythm of the SCI group is stronger than that in the HC group. Moreover, we extract the PLV with common spatial pattern (PLV-CSP) feature from the MI data of the SCI group. The experimental results for 12 SCI patients include that the peak classification accuracy is 100%, and the average accuracy of the ten-fold cross-verification is 95.6%. Our proposed approach can be used as a potential valuable method for SCI pathological studies and MI-based BCI rehabilitation systems.
脊髓损伤（SCI）是指脊髓部位的损伤导致运动功能障碍，这促使人们使用基于运动想象（MI）的脑机接口（BCI）系统来进行运动功能重建。然而，分析脊髓损伤患者的脑电图信号和脑功能机制具有挑战性。这是因为他们的信号噪声比低且变异性高。我们建议使用锁相值（PLV）为脊髓损伤患者和健康个体构建α和β节律的脑网络。这种方法旨在分析脊髓损伤后脑网络连接性和脑功能机制的变化。结果表明，健康对照组（HC）的α节律连接强度强于脊髓损伤组，而脊髓损伤组的β节律连接强度强于健康对照组。此外，我们从脊髓损伤组的运动想象数据中提取了具有共同空间模式的锁相值（PLV-CSP）特征。12名脊髓损伤患者的实验结果显示，峰值分类准确率为100%，十折交叉验证的平均准确率为95.6%。我们提出的方法可作为脊髓损伤病理研究和基于运动想象的脑机接口康复系统的一种潜在有价值的方法。
