REMOVAL OF EOG FROM EEG VIA DIRECT INTERROGATION

该提案的长期目标是开发技术， 从脑活动（EEG）测量中去除伪影， 自动脑电图分析 这种能力对于临床和 非临床的努力。 该项目的具体目标是开发一种技术， EOG的影响，从观察到的EEG，这可以实现在线， 实时的 鉴于所有眼球运动和眨眼都会导致眼电图 电信号，其特征在于幅度、频率和相位， 外部驱动信号施加在眼睛附近，全带宽为 信号特性可以模拟自然EOG。 这个，沿着 观察这种驱动对远处EEG电极的影响，将允许 衰减因子或传递函数的估计， 驱动信号。 然后，该传递函数可以用于校正EEG 通过减去加权的自然EOG。 的可行性和有效性 这一技术将在统计和分析两方面加以说明。

The long term objective of this proposal is to develop technology for removing artifacts from brain activity (EEG) measurements in order to allow automated EEG analysis. This capability will be important for clinical and non-clinical endeavors. The specific aim of this project is to develop a technique for removing the EOG influence form the observed EEG, which can be implemented on-line in real-time. Given that all eye movements and blinks result in EOG electrical signals characterized by amplitude, frequency, and phase, an external drive signal applied near the eye with the full bandwidth of signal characteristics can emulate the natural EOG. This, along with the observation of this drive's influence on distant EEG electrodes, will allow an estimation of the attenuation factors, or transfer funciton, of this drive signal. This transfer function, then, can be used to correct the EEG by subtracting the weighted natural EOG. The feasibility and validity of this technique will be shown both statistically and analytically.