Structure-function relationship in the human brain

大脑皮层的神经元网络在有目的的行为以及休息和睡眠期间持续活跃。活动水平在时间上受到每个神经元群体的内部动态以及从神经系统的皮质、皮质下和外周中心接收的信号的调节。近年来，人们对功能磁共振成像(FMRI)测量的相关活动的空间模式产生了浓厚的兴趣，因为这些模式被认为反映了神经元群体之间相互作用的空间分布。当参与者被要求执行任务时，通常会检测到一些功能网络，而当不需要参与时(休眠状态)，则会观察到其他网络。最近有人提出，这些功能网络的空间模式可以由大脑的结构性纤维连接来支持。然而，这种结构-功能关系的性质才刚刚开始揭示。近年来，一些作者已经开始研究从结构连通性中推断功能关联模式的可能性，方法是构建结构连通性图(Tractograph)并提取耦合，以用于大脑皮层大规模动力学的计算模型。随后，根据这种动力学，可以模拟血氧水平依赖(BOLD)信号，并将其与来自功能磁共振成像的经验观测进行定量比较。在我的研究中，我将开发大规模的动力学计算模型来研究人脑中的结构-功能关系。我的方法将考虑到这样一种可能性，即大脑皮质不同部分的动力学是不同的，例如，以较低或较高的频率为特征，并且可能因大脑状态或条件的不同而不同。为了评估不同的动力学，我将使用18F-脱氧葡萄糖正电子发射断层扫描测量的代谢活动在休息时或脑电信号自发活动的功率。我假设，通过为不同的皮质区域引入不同的动力学，将在健康清醒的受试者中获得更好的从结构连接到功能的预测，并将在意识状态改变时得到显著改善，例如，睡眠或麻醉引起的改变。对经验功能连接性的良好预测不仅取决于用于模拟的模型，还取决于在测量的BOLD信号中分离出来自其他虚假来源(如心跳或呼吸)的神经元贡献的分析。然后，我将研究不同的统计分解，以测试哪种统计分解可以更有效地分解休息的大脑，更好地分离神经元组件。最后，我将采用最近开发的一项技术，该技术允许我们使用动脉自旋标记(ASL)对比和BOLD信号同时获得可量化的血流灌注图像。ASL已被证明提供了大脑血流量的绝对测量，我假设同时采集将有助于将神经元的归属从BOLD信号的其他可能的生理来源中分离出来。总之，我相信我的研究将有助于更好地理解我们在休息的大脑中观察到的同步模式的基础上的机制。研究结构连接和新陈代谢供应如何支持大脑信息流，最终将使我能够推导出相关功能，使任何网络像大脑一样高效。出于这个原因，我相信神经科学的其他领域，如社会科学或经济学，将充分利用我的成果。

Networks of neurons in the cerebral cortex are continuously active during purposeful behavior, as well as during rest and sleep. Levels of activity are modulated in time by the internal dynamics of each neuronal population and by signals received from cortical, subcortical, and peripheral centres of the nervous system. In recent years, there has been intense interest in the spatial patterns of correlated activity as measured by functional magnetic resonance imaging (fMRI), because these patterns are believed to reflect the spatial distribution of interaction between neuronal populations. While some functional networks are commonly detected when participants are asked to perform a task, other networks are observed when no participation is required (resting state). It has been recently suggested that these functional networks spatial patterns can be supported by the structural fiber wiring of the brain. However, the nature of this structure-function relationship is only just beginning to be revealed. In recent years some authors have begun investigating the possibility of inferring functional correlation patterns from structural connectivity by building structural connectivity maps (tractography) and extracting the couplings to be used in a computational model of the large-scale dynamics of the cerebral cortex. Subsequently, from such dynamics blood oxygen level dependent (BOLD) signals can be simulated and quantitatively compared with empirical observations from fMRI. In my research I will develop large-scale dynamics computational models to investigate the structure-function relationship in the human brain. My approach will take into account the possibility that dynamics are different in different parts of the cortex, for example to be characterized by lower or higher frequency, and could be different for different brain states or conditions. In order to assess different dynamics, I will use metabolic activity as measured by 18F-fluorodeoxyglucose positron emission tomography during rest or the power of the EEG signal spontaneous activity. I hypothesize that by introducing different dynamics for different cortex patches a better prediction of functional from structural connectivity will be obtained in healthy awake subjects and will be substantially improved in altered states of consciousness as for example induced by sleep or anesthesia. A good prediction of the empirical functional connectivity does not depend only on the model used for the simulation but also on the analysis performed to isolate in the measured BOLD signal the neuronal contribution from other spurious sources such heart beating or respiration. I will then investigate different statistical decomposition to test which one can more efficiently break down the resting brain to better isolate the neuronal component. Finally, I will employ a recently developed technique, which permits us to acquire simultaneously quantifiable perfusion images using arterial spin labeling (ASL) contrast and BOLD signals. ASL has been proven to give an absolute measure of cerebral blood flow and I hypothesize that a simultaneous acquisition will help disentangle the neuronal attribution from other possible physiological origins of the BOLD signal. In conclusion I believe my research will importantly contribute to better understand the mechanisms at the base of the synchronization patterns we observe in a resting brain. And studying how the structural wiring together with the metabolic supply can support brain information flow will finally permit me to derive the relevant features that can make any networking efficient as the brain is. For this reason I believe that other fields from neuroscience, like social science or economy, will take full advantage from my results.