New approach for identification pHFO networks to predict epileptogenesis

项目总结 癫痫是最常见的严重神经系统疾病之一，约40%的癫痫患者没有 对现有的治疗方法有反应。临床上，长期难治性癫痫的手术结果为阴性。 通常与分布性癫痫发作有关，而不是局部致痫区域。了解 癫痫作为一种大规模的脑网络异常，使新的治疗选择和治疗方案得以发展 研究方向。目前，大多数与癫痫网络分析相关的研究已经被 重点放在发作期，而很少有人致力于分析癫痫发生的早期阶段 (潜伏期)。研究癫痫发生的大脑网络特性也同样重要，并有助于 开展抗癫痫干预，防治癫痫。在我们实验的早期，我们 发现了病理性高频振荡(PHFO)，这是癫痫发生的可靠生物标志物。 它们是由病理上相互连接的神经元簇(PIN簇)产生的，并反映出 人口激增。最近慢性癫痫动物模型的最新进展证明了这种空间分布 PHFO事件，暗示癫痫发生过程中大规模PIN簇网络的发展。它是 研究PIN簇形成的致痫网络的网络拓扑和特征至关重要 以进一步了解癫痫发生的潜在机制。 为了填补这一空白，目前的研究计划是探索基于PHFO的网络使用红藻氨酸(KA)- 诱导性癫痫持续状态(SE)模型。我们假设SE后癫痫的发生是 依赖于由空间产生量表示的大规模PIN集群网络的形成 和PHFOS的时间耦合。结合生物相容的有机材料神经接口阵列 (Neurogrid)使用多通道硅探针，我们的目标是识别PHFO的空间和时间分布 (目标1)。利用图论分析、香农信息熵等先进的计算算法 (Se)，我们建议研究PHFO致痫的因果关系和特征。 网络(AIM2)。这项研究的结果将评估新的基于网络的记录设计的健壮性 和算法开发。它还将确定PHFO派生的网络参数是否为 癫痫发生的可靠生物标志物。未来的计划是将PHFO网络的概念和 用于癫痫临床研究的计算机工具。这种方法可能会打开一个新的方向 预防癫痫的发展和治疗癫痫。

PROJECT SUMMARY Epilepsy is among the most common serious neurological disorders, and about 40% of epilepsy patients do not respond to existing treatment. Clinically, the prolonged, refractory epilepsy with negative surgical outcomes is often associated with distributed epilepsy onset rather than a local epileptogenic zone. Understanding the epilepsy as a large-scale brain network abnormality enables the development of new treatment options and research directions. At present, the majority of research related to analysis of the epileptic network has been focused on the ictal period, while few have been devoted to the analysis of the earlier stages of epileptogenesis (latent period). Investigating the brain network properties of epileptogenesis is as important and can help develop antiepileptogenic interventions for epilepsy prevention and cure. Early in our experiments, we discovered pathological high-frequency oscillations (pHFOs), which are reliable biomarkers of epileptogenesis. They are generated by clusters of pathologically interconnected neurons (PIN-clusters) and reflect bursts of population spikes. Recent updates in the animal models of chronic epilepsy evidenced the spatially distributed pHFO events, which implies the development of large-scale PIN-cluster networks during epileptogenesis. It is critical to study the network topology and characteristics of PIN-cluster-formed epileptogenic networks in order to further understand the underlying mechanisms of epileptogenesis. To fulfill this gap, the present study plan is to explore pHFO-based networks using the Kainic Acid (KA)- induced status epilepticus (SE) model of epileptogenesis. We hypothesize that epileptogenesis after SE is dependent upon the formation of large-scale PIN-cluster networks that is expressed by the spatial occurrence and temporal coupling of pHFOs. Combining the biocompatible, organic–material based neural interface array (NeuroGrid) with multichannel silicon probes, we aim to identify the spatial and temporal profiles of pHFOs (Aim1). Using the advanced computational algorithms such as graph theory analysis and Shannon Entropy (SE), we propose to investigate the causal relationship and characteristics of the pHFO-based epileptogenic networks (Aim2). The outcome of this study will assess the robustness of novel network-based recording design and algorithm development. It will also determine whether the pHFO-derived network parameters are a reliable biomarker of epileptogenesis. The future plans are to translate the pHFO-network concept and computational tools into the clinical study of epilepsy. This approach may open a new direction to the prevention of epilepsy development and cure epilepsy.