 This study proposes a new approach to improve the accuracy of brain computer interfaces, BCIs, for stroke rehabilitation. It combines two existing methods, temporal-spatial analysis and network analysis, to identify patterns of activity in the brain related to motor tasks. The resulting combination of these two methods has been shown to increase the accuracy of BCI classification for stroke patients, with some patients achieving over 80% accuracy. This suggests that combining these two approaches may be beneficial for developing more effective BCI training programs for stroke rehabilitation. This article was authored by Lei Chao, Wenrong Wang, Chen Xihuang, and others.