| الوصف: |
Electrophysiological signal processing often requires blind source separation (BSS) techniques due to the nature of mixing source signals. However, its complex computational demands make real-time applicability challenging. In this study, we propose a Constrained Online Recursive Source Separation (CORSS) framework for real time electrophysiological signals processing. With a stepwise recursive unmixing matrix learning rule, the algorithm achieves real-time updates with minimal computational overhead. By incorporating prior information of target signals to optimize the cost function, the algorithm converges more readily to ideal sources, yielding more accurate results. Two downstream tasks, real-time surface electromyogram (sEMG) decomposition and real-time respiratory intent monitoring based on diaphragmatic electromyogram (sEMGdi) extraction, were employed to assess the efficacy of our method. The results demonstrated superior performance compared to alternative methods, achieving a matching rate of 96.00% for the sEMG decomposition task and 98.12% for the sEMGdi extraction task. Our method also exhibits minimal time delay during computation, reflecting its streamlined updating rule as well as excellent real-time capabilities, with only 12.5ms delay when the block size is 0.1s, demonstrates strong performance for real-time processing. Our work shows great significance for applications in real-time human-computer interaction and clinical monitoring. |
