Functional MRI Method Development

功能性 MRI 方法开发

• 批准号: 9589754
• 负责人: Peter Bandettini
• 金额: $ 212.43万
• 依托单位: NATIONAL INSTITUTE OF MENTAL HEALTH
• 依托单位国家: 美国
• 资助国家: 美国
• 起止时间: 至
• 项目状态: 未结题
• 来源: https://reporter.nih.gov/project-details/9589754
• 关键词: Algorithms; Arousal; Attention; Automobile Driving; Back; Behavior; Behavioral; Belief; Blood Volume; Brain; Brain Mapping; Brain region; Characteristics; Classification; Clinical; Collaborations; Complex; Cortical Column; Data; Detection; Development; Diagnosis; Early identification; Electroencephalography; Elements; Epilepsy; Etiology; Evaluation; Event; Excision; Feeling; Functional Imaging; Functional Magnetic Resonance Imaging; Goals; Image; Individual; Individual Differences; Intervention; Link; Location; Magnetic Resonance Imaging; Maps; Mathematics; Measures; Medical Imaging; Mental disorders; Methodology; Methods; Mind; Modeling; Neurologic; Neurons; Neurosciences Research; Noise; Paper; Paranoia; Participant; Patients; Pattern; Performance; Physiological; Population; Protocols documentation; Research; Resolution; Rest; Risk; Scanning; Scheme; Series; Signal Transduction; Slice; Specificity; Standardization; Stimulus; Stress Tests; Surface; Techniques; Testing; Time; Visual; Work; association cortex; base; blood oxygen level dependent; cognitive task; cortex mapping; design; experimental study; flexibility; functional MRI scan; hemodynamics; imaging modality; mental imagery; mental state; method development; movie; neuroimaging; novel; psychologic; response; sensorimotor system; social; spatiotemporal; specific biomarkers; sustained attention; trait; treatment planning; ultra high resolution

Protocol number 93-M-0170, NCT00001360 The Section on Functional Imaging Methods (SFIM) has advanced functional MRI (fMRI) methodology through development of processing and acquisition methods as well as research on the underlying mechanisms behind the fMRI signal. Our work over the past year has focused on methods to extract more information related to tasks, mental state, and individual traits from the fMRI time series. At high resolution we have more accurately mapped cortical layer connectivity and activity. Due to limited space, we decided to highlight here a few projects that are representative of our ongoing work. Multi-Echo Applications Our group has been pioneering the development and use of Multi-echo fMRI (ME fMRI). Because blood oxygen level dependent (BOLD) and non-BOLD signal fluctuations behave differently as a function of echo time, the unique ME fMRI acquisition of at least one echo allows for efficient and robust detection and removal of non-BOLD fluctuations from the time series. Our work developing and utilizing ME fMRI continues. In collaboration with Dr. Caballero-Gaudes, we are working on the detection of individual BOLD events with no information about the task timing or location. Preliminary tests show that the algorithm can reliably detect BOLD events associated with each individual trial in a manner similar to traditional GLM methods, yet without any information on paradigm timing. This new method has important applications including: uncovering hemodynamic events driving dynamic resting functional connectivity, and detecting inter-ictal events in epilepsy patients without the need for concurrent EEG recordings. Dynamic Functional Connectivity Dynamic functional connectivity (FC), understood as patterns of functional brain connectivity that change at the scale of seconds to minutes, has gained great attention in recent years. FC configurations can differ significantly from those obtained from entire time series, and recent studies suggest there is a link between these dynamics and individual behavior and behavioral traits. We previously demonstrated how whole-brain short-term FC patterns could be used to identify cognitive tasks. We are currently investigating how these patterns may provide useful behaviorally-relevant information about individual subjects to be used as a biomarker for specific traits. We are also investigating elements that drive these dynamics and differences in dynamics ranging from changes in arousal and engagement to changes in attention. Lastly, we are exploring the degree to which dynamics of specific networks can predict behavior. Towards the goal of deriving specific individual traits from FC network strengths, we had subjects perform a 2-back task, look for a visual target in a movie, and solve math problems. Performance was recorded on each task. Surprisingly, the sustained attention network (previously found) provided better prediction of performance than the 2-back-specific network. fMRI Connectivity vs. fMRI Magnitude Changes Our work on task-based connectivity change assessment has opened up an entirely new vista in fMRI processing. We have revealed that, for cognitive tasks, task based connectivity changes are more spatially extensive than magnitude changes and more specific or sensitive for a given task. This observation potentially opens a new subfield of brain mapping where connectivity changes are mapped rather than magnitude changes. We are actively exploring the robustness, repeatability, and task specificity of these observations. Naturalistic Tasks Traditional task-based fMRI experiments use tightly controlled paradigms that often lack ecological validity. Resting-state scans, on the other hand, are unconstrained, making it difficult to separate signal from noise. Naturalistic tasks, in which subjects view a movie or listen to a story in the scanner, may provide a happy medium for studying both group-level functional brain organization and individual differences. By imposing a standardized, time-stamped, and engaging stimulus on all subjects, naturalistic tasks evoke rich patterns of brain activity. These patterns lend themselves to flexible, data-driven analyses such as inter-subject correlation (ISC) and inter-subject functional connectivity (ISFC), which are model-free ways to isolate stimulus-dependent brain activity from spontaneous activity and noise. These techniques have several advantages over traditional approaches: 1) they do not requirea priorimodeling of specific task events and/or assumptions about the functional specificity of individual brain regions; 2) there is no need to assume a fixed hemodynamic response function; and 3) they allow for the characterization of the full spatiotemporal richness of both evoked and intrinsic brain activity. In our work, we hypothesize that individual differences in psychological traits - namely suspicion and paranoia - are able to be differentiated by observation of brain activity elicited by a naturalistic task involving a narrative describing a complex social scenario that was deliberately ambiguous regarding characters trustworthiness and intentions. The intent was that some subjects would interpret the story as more suspicious, and others as less so. The narrative did evoke a range of interpretations. We collected data from 23 healthy subjects as they listened to the pre-recorded narrative during fMRI scanning, and characterized their feelings and beliefs about the narrative. We found that pairs of subjects with more similar activity fluctuations in primary visual and visual association cortices--suggesting that they were engaging in more similar mental imagery during listening--were more likely to speak about the story in similar ways afterward. One ultimate goal of this line of work is to develop a neuroimaging-based stress test for use in both clinical and subclinical populations. Such a test would draw out individual differences beyond what can be observed at rest or using more traditional tasks, potentially leading to earlier identification of subjects at risk for mental illness or helping to guide interventions. Continuation of mapping layer-specific brain activity at 7T In the last year, we continued to develop acquisition fMRI methods that can measure brain activity with ultra-high resolution, aiming to distinguish activity across cortical layers. The novel parameter space explored since October 2016 includes: a. The necessity of physiological noise correction with sub-millimeter voxels where thermal noise dominates. b. The capabilities of layer-fMRI at ultra-high field strengths of 9.4T compared to 7T. c. The layer-dependent signal features across different cortical columns in M1 in a participant-specific surface-based signal evaluation scheme. d. The capabilities to image whole slices of the brain beyond the sensory motor system e. The capabilities of alternative contrast mechanisms, including spin-echo BOLD compared to blood-volume sensitive contrast and gradient-echo BOLD fMRI. f. Resting state connectivity from layers to the rest of the brain, in an attempt to arrive at directional input and causality across cortical networks.

协议号 93-M-0170，NCT00001360 功能成像方法部分 (SFIM) 通过开发处理和采集方法以及研究 fMRI 信号背后的潜在机制，拥有先进的功能 MRI (fMRI) 方法。我们过去一年的工作重点是从功能磁共振成像时间序列中提取更多与任务、心理状态和个人特征相关的信息的方法。在高分辨率下，我们可以更准确地映射皮质层连接和活动。由于空间有限，我们决定在此重点介绍一些代表我们正在进行的工作的项目。 多回波应用 我们的小组一直在开发和使用多回波功能磁共振成像（ME fMRI）方面处于领先地位。由于血氧水平相关 (BOLD) 和非 BOLD 信号波动作为回波时间的函数表现不同，因此至少一个回波的独特 ME fMRI 采集可实现高效、稳健的检测，并从时间序列中去除非 BOLD 波动。 我们开发和利用 ME fMRI 的工作仍在继续。我们与 Caballero-Gaudes 博士合作，致力于在没有任务时间或位置信息的情况下检测单个 BOLD 事件。初步测试表明，该算法可以以类似于传统 GLM 方法的方式可靠地检测与每个单独试验相关的 BOLD 事件，但无需任何有关范式计时的信息。这种新方法具有重要的应用，包括：揭示驱动动态静息功能连接的血流动力学事件，以及检测癫痫患者的发作间期事件，而无需同时进行脑电图记录。 动态功能连接 动态功能连接（FC）被理解为以秒到分钟为单位变化的功能性大脑连接模式，近年来引起了极大的关注。 FC 配置可能与从整个时间序列获得的配置有很大不同，最近的研究表明这些动态与个人行为和行为特征之间存在联系。我们之前演示了如何使用全脑短期 FC 模式来识别认知任务。我们目前正在研究这些模式如何提供有关个体受试者的有用的行为相关信息，以用作特定特征的生物标志物。我们还在研究驱动这些动态的因素以及动态差异，从唤醒和参与度的变化到注意力的变化。 最后，我们正在探索特定网络的动态可以预测行为的程度。 为了从 FC 网络优势中得出特定的个体特征，我们让受试者执行 2-back 任务，在电影中寻找视觉目标，并解决数学问题。记录每项任务的表现。令人惊讶的是，持续注意力网络（之前发现的）比 2-back 特定网络提供了更好的性能预测。 fMRI 连接性与 fMRI 幅度变化 我们在基于任务的连接变化评估方面的工作为功能磁共振成像处理开辟了全新的前景。我们发现，对于认知任务，基于任务的连通性变化比幅度变化在空间上更广泛，并且对于给定任务更具体或更敏感。这一观察结果可能开辟了大脑绘图的一个新子领域，其中绘制的是连通性变化而不是幅度变化。我们正在积极探索这些观察结果的稳健性、可重复性和任务特异性。 自然任务 传统的基于任务的功能磁共振成像实验使用严格控制的范式，通常缺乏生态有效性。另一方面，静息态扫描不受约束，因此很难将信号与噪声分开。自然主义任务，即受试者在扫描仪中观看电影或听故事，可能为研究群体水平的功能性大脑组织和个体差异提供一个愉快的媒介。通过对所有受试者施加标准化的、带有时间戳的、引人入胜的刺激，自然主义任务会唤起丰富的大脑活动模式。这些模式适用于灵活的数据驱动分析，例如受试者间相关性 (ISC) 和受试者间功能连接 (ISFC)，这是将刺激依赖性大脑活动与自发活动和噪音隔离开来的无模型方法。与传统方法相比，这些技术有几个优点：1）它们不需要对特定任务事件进行先验建模和/或对单个大脑区域的功能特异性进行假设； 2）无需假设固定的血流动力学响应函数； 3）它们允许表征诱发和内在大脑活动的完整时空丰富性。 在我们的工作中，我们假设心理特征的个体差异（即怀疑和偏执）可以通过观察自然主义任务引起的大脑活动来区分，该自然主义任务涉及描述复杂社会场景的叙述，该场景在人物的可信度和意图方面故意含糊不清。 。目的是让一些受试者认为这个故事更可疑，而另一些受试者则认为不那么可疑。这个叙述确实引起了一系列的解释。我们收集了 23 名健康受试者在功能磁共振成像扫描过程中聆听预先录制的叙述时的数据，并描述了他们对叙述的感受和信念。我们发现，在初级视觉和视觉关联皮层中具有更相似活动波动的一对受试者（这表明他们在聆听过程中参与了更相似的心理意象）更有可能在之后以类似的方式谈论这个故事。这一工作的最终目标是开发一种基于神经影像学的压力测试，用于临床和亚临床人群。这样的测试将揭示超出休息时或使用更传统的任务所能观察到的个体差异，从而可能导致更早识别有精神疾病风险的受试者或帮助指导干预措施。 继续绘制 7T 层特定大脑活动图 去年，我们继续开发采集功能磁共振成像方法，能够以超高分辨率测量大脑活动，旨在区分皮层各层的活动。自 2016 年 10 月以来探索的新颖参数空间包括：在热噪声占主导地位的亚毫米体素中进行生理噪声校正的必要性。 b.与 7T 相比，层 fMRI 在 9.4T 超高场强下的能力。 c.在特定于参与者的基于表面的信号评估方案中，M1 中不同皮质柱的层相关信号特征。 d.对感觉运动系统之外的整个大脑切片进行成像的能力替代对比机制的功能，包括与血容量敏感对比相比的自旋回波 BOLD 和梯度回波 BOLD fMRI。 f.从各层到大脑其他部分的静息状态连接，试图跨皮质网络达成定向输入和因果关系。