基于小波分解的颅脑动态电阻抗监测中体动干扰实时处理方法研究

Real-time processing method of motion interferences in dynamic brain electrical impedance monitoring based on wavelet decomposition

目的:为抑制临床颅脑电阻抗动态监测中体动干扰导致的图像伪影,提出一种基于小波分解的体动干扰实时处理方法。方法:首先,使用混合高斯模型描述颅脑电阻抗信号小波分解系数分布特征,对体动干扰和正常信号进行区分并加以处理;其次,采用垒墙式的计算策略,实现小波分解的实时化处理;最后,开展仿真实验与人体实测数据实验验证提出方法的有效性。结果:实验结果表明,该方法可以有效抑制尖峰类型的体动干扰,恢复信号的连续性,并显著减少图像上的重构伪影,恢复正常的图像监测。结论:基于小波分解的体动干扰实时处理方法能够有效抑制临床颅脑电阻抗动态监测中的体动干扰,特别适用于多通道数据采集的应用场景。

Objective To propose a wavelet decomposition-based real-time processing method for motion interferences to suppress the motion interference-induced image artifacts during dynamic brain electrical impedance monitoring.Methods Motion interferences were differentiated from normal signals by using the mixed model Gaussian model to characterize the distribution of wavelet decomposition coefficients of the cranio-cerebral electrical impedance signals;a bricklaying computa-tion strategy was used to realize real-time wavelet decomposition;simulation and human experiments were conducted to verify the validity of the proposed method.Results The experimental results showed that the method could effectively suppress the spike-type motion interferences,restore the signal continuity,and significantly reduce the reconstruction artifacts on the image.Conclusion The wavelet decomposition-based real-time processing method for motion interferences behaves well during dynamic brain electrical impedance monitoring,which is especially suitable for the application scenario of multi-channel data acquisition.