Colpan Mustafa Efkan, Li Yue, Dwyer Jennifer, Mogul David Jeffrey
Pritzker Institute of Biomedical Science & Engineering; Illinois Institute of Technology, 10 West 32nd Street, Chicago, IL 60616, U.S.A.
Epilepsia. 2007 Aug;48(8):1594-603. doi: 10.1111/j.1528-1167.2007.01073.x. Epub 2007 Apr 18.
A responsive electrical brain stimulation system using control feedback was investigated for the treatment of seizures.
A proportional feedback stimulation system was designed. Penicillin-induced episodic seizures were created in rat primary motor cortex. Both intracranial (proximal to seizure focus) and extracranial EEGs were monitored. Current stimulation was applied at the seizure focus by using the intracranial EEG as the current-stimulus template. Different gains (H) for determining feedback stimulus amplitudes were tested.
The effect of feedback stimulation on seizures was initially assessed by measuring change in variance of the amplitude histogram of the intracranial EEG before and during stimulation. Mean reduction in amplitude variance during seizure activity was significant, with variance during stimulation progressively reduced as feedback gain was increased, indicating that overall suppression of seizure amplitude depended on H. Further increases in feedback gain typically produced saturating oscillations, indicating that this level of H resulted in instability. Frequency analysis of seizure and stimulation periods for each of the effective levels of H demonstrated close correlation across a large frequency domain, suggesting that the reduction in EEG seizure amplitude during feedback stimulation was possibly because of shunting of neuronal currents near electrodes as opposed to an alteration of neuronal dynamics. Although the frequency and energy responses during seizures before or during feedback stimulation remained well correlated in the delta band, this correlation progressively decreased across the theta, alpha, and beta bands.
These results demonstrate that proportional feedback stimulation holds the promise of suppressing seizure activity. More-complicated control algorithms for generating feedback stimulation may provide further improvements in seizure suppression.
研究一种使用控制反馈的响应式脑电刺激系统用于治疗癫痫发作。
设计了一种比例反馈刺激系统。在大鼠初级运动皮层诱发青霉素引起的发作性癫痫。监测颅内（靠近癫痫病灶）和颅外脑电图。以颅内脑电图作为电流刺激模板，在癫痫病灶处施加电流刺激。测试了用于确定反馈刺激幅度的不同增益（H）。
反馈刺激对癫痫发作的影响最初通过测量刺激前和刺激期间颅内脑电图幅度直方图方差的变化来评估。癫痫发作活动期间幅度方差的平均降低是显著的，随着反馈增益增加，刺激期间的方差逐渐降低，表明癫痫发作幅度的总体抑制取决于H。反馈增益的进一步增加通常会产生饱和振荡，表明这个H水平会导致不稳定性。对每个有效H水平的癫痫发作期和刺激期进行频率分析，发现在大频率范围内密切相关，这表明反馈刺激期间脑电图癫痫发作幅度的降低可能是由于电极附近神经元电流的分流，而不是神经元动力学的改变。虽然在反馈刺激之前或期间癫痫发作期间的频率和能量响应在δ波段仍保持良好的相关性，但这种相关性在θ、α和β波段逐渐降低。
这些结果表明比例反馈刺激有望抑制癫痫发作活动。用于产生反馈刺激的更复杂控制算法可能会在癫痫抑制方面提供进一步的改善。
