Magnetic resonance imaging and modelling of brain function and microvascular physiology

功能磁共振成像(FMRI)提供了一种非侵入性的工具，无需使用电离辐射或外源性造影剂即可动态绘制大脑活动图，从而使人类神经科学发生了革命性的变化。大多数fMRI研究利用血液氧合水平依赖(BOLD)效应，即血液的氧饱和度调节MRI信号水平，以定位与神经元活动变化相关的血液氧合变化。最近，定量功能成像技术利用BOLD效应来估计与基础和临床神经科学相关的生理参数，如血氧饱和度和脑氧代谢率，而另一种“非BOLD”成像技术已被开发用于测量额外的生理参数，如脑血流量。由于大脑是如此严重地依赖微血管系统来持续和适应性地供应营养，磁共振成像为绘制和测量健康和疾病中的大脑功能提供了一个强大的工具。我们现在知道，传统的BOLD功能磁共振成像受到一系列因素的影响，既有生理因素，也有习得因素。这些复杂的影响使得大胆的功能磁共振成像具有挑战性，既涉及潜在的神经元活动，也涉及血管生理学。我的长期愿景是，通过消除或利用功能磁共振信号的生物物理偏差，我们将能够以更高的特异性和敏感度成像神经元活动以及微血管功能和结构。在这个计划中，我们将开发建模工具来改进我们对BOLD信号的解释，我们将开发对特定生理参数敏感的MRI采集技术。主要目标是：1)开发和实验验证血液的真实生物物理模型，以模拟来自血液的MRI信号--BOLD信号中一个重要但通常被忽略的分量。2)提高被称为“校准fMRI”的技术的可及性和适用性，该技术可用于测量氧代谢的变化。3)开发新的图像采集和重建技术，以便在减少图像伪影的情况下进行功能磁共振成像，并以更高的特异性和信噪比成像脑血流。这项研究计划将fMRI从目前受到众多偏见限制的定性脑成像工具提升到提高神经元特异性和准确测量微血管生理学的功能。我们开发的工具将使我们能够更好地了解神经元活动和血流动力学之间的紧密耦合，而几乎所有的功能成像技术都依赖于此。最终，这些发展可以被整合到基础和临床人类神经科学中，在那里可以探索关于神经元活动、新陈代谢和血管功能在健康寿命期间以及在一系列神经和精神障碍中的变化的公开问题。

Functional magnetic resonance imaging (fMRI) has revolutionized human neurosciences by providing a non-invasive tool for dynamically mapping brain activity without the use of ionizing radiation or exogenous contrast agents. Most fMRI studies exploit the Blood Oxygenation Level-Dependent (BOLD) effect, whereby the oxygen saturation of blood modulates the MRI signal level, to localize changes in blood oxygenation that are related to changing neuronal activity. More recently, quantitative functional imaging techniques have used the BOLD effect to estimate physiological parameters of basic and clinical neuroscientific relevance, such as blood oxygen saturation and the cerebral metabolic rate of oxygen, while alternative "non-BOLD" imaging techniques have been developed to measure additional physiological parameters, such as cerebral blood flow. Because the brain is so heavily reliant on the microvasculature for a continuous and adaptive supply of nutrients, MRI provides a powerful tool for mapping and measuring brain function in health and disease. We now know that conventional BOLD fMRI is influenced by a wide range of factors, both physiological and acquisition related. These complex influences make BOLD fMRI challenging to relate to both the underlying neuronal activity and the vascular physiology. My long-term vision is that by removing or exploiting the biophysical biases of fMRI signals, we will be able to image neuronal activity and microvascular function and structure with greater specificity and sensitivity. In this proposal, we will develop modelling tools to improve our interpretation of the BOLD signal and we will develop MRI acquisition techniques that are sensitive to specific physiological parameters. The main objectives are: 1) Develop and experimentally validate a realistic biophysical model of blood to simulate the MRI signal from blood - a significant yet often ignored component of the BOLD signal. 2) Improve the accessibility and applicability of a technique known as "calibrated fMRI", which can be used to measure changes in oxygen metabolism. 3) Develop novel image acquisition and reconstruction techniques for performing fMRI with reduced image artifacts and for imaging cerebral blood flow with improved specificity and signal-to-noise ratio. This research program will elevate fMRI from a qualitative brain mapping tool that is currently limited by numerous biases to mapping function at increased neuronal specificity and accurately measuring the microvascular physiology. The tools that we develop will allow us to better understand the tight coupling between neuronal activity and hemodynamics, which nearly all functional imaging techniques rely on. Ultimately, these developments can be integrated into basic and clinical human neurosciences, where open questions on alterations of neuronal activity, metabolism, and vascular function over the healthy lifespan and in a host of neurological and psychiatric disorders can be probed.