FUNCTIONAL MRI DATA AQUISITION AND ANALYSIS

这个子项目是许多研究子项目中的一个 由NIH/NCRR资助的中心赠款提供的资源。子项目和 研究者（PI）可能从另一个NIH来源获得主要资金， 因此可以在其他CRISP条目中表示。所列机构为 研究中心，而研究中心不一定是研究者所在的机构。 这一资源的总体目标，关于功能磁共振成像，一直是并仍然是提高特异性和 通过改进数据采集和分析，提高人类功能性MRI的灵敏度。在上一次赠款中， 期间，本TRD的目标集中在基于灌注的fMRI采集方法的开发上， 基于独立成分分析（伊卡）的fMRI分析方法。 作为第一个目标的一部分，我们开发了一个 一种使用脑血容量（CBV）依赖对比度进行功能磁共振成像的新方法，被称为“血管空间 功能磁共振成像（fMRI）。这种方法不需要使用造影剂， 与血氧水平依赖（BOLD）fMRI相比，空间特异性增加。第二个目标，我们 开发了新的独立成分分析（伊卡）方法来分析功能磁共振成像数据，并证明， 伊卡可以发现被标准方法忽略的大脑激活，即使在时间上被忽略， 大脑激活的程度并不完全符合研究者的预期。 在这次更新申请中，该TRD将专注于fMRI采集和分析开发， 我们的合作者面临的几个问题（见下页表1）。 例如，我们的儿科合作者 研究各种发育障碍，如多动症，自闭症，阅读障碍，和创伤性 功能缺陷必须处理降低的顺应性，并且想要更快地扫描。类似的情况也适用于 痴呆和精神病患者。此外，这些调查人员和研究人员研究的影响 研究记忆功能和注意力通常包括非常小的信号调制， 信号激活（视觉，运动）。一些研究人员希望更高的空间分辨率，以更好地研究小 皮质区 所有这些问题都可以通过更高的磁场强度（7.0T）来减少，其中 可获得增加的信噪比。 我们的合作者面临的另一个问题是传统的功能磁共振成像数据分析的局限性， 血流动力学反应，这可能会错过重要的潜在大脑活动。我们已经解决了这个问题， 独立成分分析的“数据驱动”方法，该方法揭示了几个额外的 激活组件。然而，这导致了关于这些的意义和起源的根本问题。 标准fMRI数据分析中不存在的“额外”激活成分。因此，需要评估 这种伊卡的fMRI结果的特异性。最后，我们的许多合作者正在进行儿童功能磁共振成像研究， 以及神经退行性疾病患者的情况。 这些数据分析起来可能有问题，因为这样的研究 参与者可能对实验范例表现出较差的依从性。这方面的最终例子可能是 克里斯滕森医生的昏迷病人伊卡允许使用范例进行研究， 遵守，包括所谓的丰富的自然主义行为，如玩视频游戏或看电影， 昏迷的情况下，只是听一个亲戚说话。 因此，我们在未来一段时间的总体目标是提高功能磁共振成像的灵敏度，解决实验问题， 与fMRI-ICA特异性相关，并开发用于基础和临床研究的功能脑映射方法 这降低了对参与者依从性的要求。具体目标是： AIM 1.在7.0特斯拉下优化fMRI数据采集。 我们将在7.0 T下优化fMRI采集，包括并行成像加速因子、匀场、TE TR选择和切片数量，这取决于我们神经科学合作者的个人需求。 AIM 2.表征fMRI数据的独立分量。 为了表征fMRI数据的独立成分，我们将获取额外的图像数据，包括BOLD 在更高的时间和空间分辨率下的fMRI采集，使用不同对比度的fMRI采集，即 VASO和动脉自旋标记（ASL），以及结构成像，包括MP-MRI和MR血管造影。我们将使用 在三种场强（1.5、3.0和7.0特斯拉）下采集的数据，以评估 粗体功能磁共振成像数据。这些数据将使用多种方法进行分析，包括基于特征的联合伊卡。 AIM 3.发展伊卡方法的fMRI数据丰富的自然行为。 我们将开发先进的伊卡方法，用于分析从事丰富自然科学研究的人的fMRI数据。 行为，并与我们的合作者将这些方法应用于他们的研究目标。这将是为 个人和团体的数据。这些方法最终将与TRD 3中的DTI工作相结合 用于连接功能评估。

This subproject is one of many research subprojects utilizing the resources provided by a Center grant funded by NIH/NCRR. The subproject and investigator (PI) may have received primary funding from another NIH source, and thus could be represented in other CRISP entries. The institution listed is for the Center, which is not necessarily the institution for the investigator. The overall goals of this resource, with respect to fMRI, have been and remain to enhance the specificity and sensitivity of human functional MRI through improvements in data acquisition and analysis. In the previous grant period, the aims of this TRD were focused on the development of perfusion-based fMRI acquisition methods and Independent Component Analysis (ICA) based methods for fMRI analysis. As part of the first aim, we developed a novel method for performing fMRI using cerebral blood volume (CBV) dependent contrast, dubbed "Vascular Space Occupancy (VASO)" fMRI. This approach, which does not require the use of a contrast agent, was shown to have increased spatial specificity compared to Blood Oxygenation Level Dependent (BOLD) fMRI. For the second aim, we developed new Independent Component Analysis (ICA) approaches to fMRI data analysis, and demonstrated that ICA can find brain activation overlooked by standard approaches and can yield robust results even when the timing of brain activation does not precisely match that anticipated by the investigator. In this renewal application, this TRD will focus on fMRI acquisition and analysis developments that will address several issues confronting our collaborators (see Table 1, next page). For instance, our pediatric collaborators studying a variety of developmental disorders such as ADHD, autism, reading disability, and trauma-based functional deficits have to deal with reduced compliance and would like to scan faster. A similar situation is true for patients with dementia and psychosis. In addition, the effects studied by these investigators and by researchers studying memory function and attention often consist of very small signal modulations on top of more robust signal activations (visual, motor). Some investigators would like higher spatial resolution to better study small cortical areas. All of these problems can be reduced by going to higher magnetic field strength (7.0T), where increased signal to noise is available. Another issue confronting our collaborators is the limitation of conventional fMRI data analysis to pre-conceived hemodynamic responses, which may miss important underlying brain activities. We have addressed this by going to the "data-driven" methodology of independent component analysis, which has revealed several additional activation components. However, this has led to fundamental questions about the meaning and origin of these "extra" activation components not present in standard fMRI data analyses. This therefore requires assessment of the specificity of such ICA of fMRI results. Finally, many of our collaborators are pursuing fMRI studies in children and in patients with neuro-degenerative disease. These data can be problematic to analyze, as such research participants may show poor compliance with experimental paradigms. The ultimate example of this may be the coma patients of Dr. Christensen. ICA allows studies to be performed with paradigms that reduce demands on compliance, including such so-called rich naturalistic behaviors as playing a video game or watching a movie or, in the case of coma, just listening to a relative talking. Our overall goals in the coming period are therefore to enhance fMRI sensitivity, to address experimental questions related to fMRI-ICA specificity, and to develop approaches to functional brain mapping for basic and clinical research which reduce demands on participant compliance. The specific aims are: AIM 1. Optimize fMRI data acquisition at 7.0 Tesla. We will optimize fMRI acquisitions at 7.0 T with respect to parallel imaging acceleration factor, shimming, TE and TR choice, and slice number, depending on the individual needs for our neuroscience collaborators. AIM 2. Characterize the independent components of fMRI data. To characterize the independent components of fMRI data, we will acquire additional image data, including BOLD fMRI acquisitions at higher temporal and spatial resolution, fMRI acquisitions using different contrasts, namely VASO and arterial spin labeling (ASL), and structural imaging including MP-RAGE and MR angiography. We will use data acquired at three field strengths (1.5, 3.0, and 7.0 Tesla), to assess the independent components found in BOLD fMRI data. These data will be analyzed using multiple approaches including feature-based joint ICA. AIM 3. Develop ICA methods for fMRI data from rich naturalistic behaviors. We will develop advanced ICA methods for analysis of fMRI data from persons engaged in rich naturalistic behaviors, and work with our collaborators to apply these approaches to their research aims. This will be done for data from both individuals and groups. These approaches will ultimately be combined with the DTI efforts in TRD 3 for connectivity-function assessment.