Wavelets, fractals and fMRI of brain adaptivity

大脑适应性的小波、分形和功能磁共振成像

• 批准号: 7052793
• 负责人: EDWARD Edward BULLMORE
• 金额: $ 26.37万
• 依托单位: UNIVERSITY OF CAMBRIDGE
• 依托单位国家: 美国
• 财政年份: 2003
• 资助国家: 美国
• 起止时间: 2003-04-01 至 2008-03-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/7052793
• 关键词: aging; biological signal transduction; clinical research; cognition; computational neuroscience; data collection methodology /evaluation; drug interactions; drug screening /evaluation; functional magnetic resonance imaging; human subject; image processing; learning; mathematical model; mathematics; memory; method development; neuropharmacology; neuropsychological tests; neurotransmitter agonist; neurotransmitter antagonist; pharmacokinetics

DESCRIPTION (provided by applicant): This is an integrated neuroinformatics project with two long-term objectives: i) to develop and rigorously to evaluate innovative methods for analysis of functional magnetic resonance imaging (fMRI) experiments using wavelets and data resampling as key mathematical tools; ii) to enhance characterization of brain adaptivity to task repetition and variable task difficulty and to test explicit hypotheses concerning effects of normal aging and transmitter-specific drug treatments on brain adaptivity. The key informatics objectives are i) to refine and apply a new best linear unbiased estimator of regression model parameters in the context of long-memory noise (wavelet-generalized least squares); ii) to consolidate and evolve methods for multiresolutional analysis of spatially extended 2- or 3-dimensional processes using wavelets and data resampling; iii) to implement multiscale image registration in the wavelet domain as a framework in which to assess the impact of variably precise image registration on mixed effects modeling of grouped and factorially designed fMRI datasets; iv) to explore 4-dimensional wavelet shrinkage as a novel adjunct or alternative to our ongoing interest in multivariate methods of connectivity analysis; and v) to develop and cross-validate estimators of spectral exponents, fractal dimensions and other complexity measures in fMRI. These informatics objectives are critical to a comprehensive and powerful analysis of drug and aging effects on brain adaptivity and fMRI noise - which are the focus of three key neuroscientific experiments: i) we will test the assumption that fMRI noise has l/f-like spectral properties, and investigate instrumental and physiological sources for 1/f noise, using phantom and biological "resting" fMRI data; ii) we will develop and validate a set of cognitive activation paradigms focused on adaptivity to repetition and load in several domains of memory and learning; and iii) we will conduct a major (N=48) factorially designed pharmacological MRI study in young (aged 20-30 years) and older (aged 60-70 years) healthy participants each scanned on three occasions in a randomized, placebo-controlled experiment. These data will allow us to test hypotheses concerning the main effects of agonist and antagonist drugs targeted on dopaminergic, cholinergic and GABAergic transmitter systems and thereby to elucidate mechanisms for large-scale brain adaptivity. We will also test hypotheses concerning the main effect of aging and the interactive effect of drugs and age on brain adaptivity.

描述（由申请人提供）：这是一个综合神经信息学项目，有两个长期目标：i)开发并严格评估功能磁共振成像（fMRI）实验分析的创新方法，使用小波和数据重采样作为关键的数学工具；Ii)加强大脑对任务重复和可变任务难度的适应性表征，并检验关于正常衰老和递质特异性药物治疗对大脑适应性影响的明确假设。关键的信息学目标是：1)在长记忆噪声（小波-广义最小二乘）的背景下，改进和应用一种新的最佳线性无偏估计回归模型参数；Ii)利用小波和数据重采样巩固和发展空间扩展的二维或三维过程的多分辨率分析方法；iii)在小波域实现多尺度图像配准作为框架，以评估不同精度的图像配准对分组和因子设计fMRI数据集混合效果建模的影响；Iv)探索四维小波收缩作为一种新的辅助或替代我们对多变量连通性分析方法的持续兴趣；v)开发和交叉验证fMRI中谱指数、分形维数和其他复杂性度量的估计。这些信息学目标对于全面而有力地分析药物和衰老对大脑适应性和fMRI噪声的影响至关重要——这是三个关键神经科学实验的重点：i)我们将测试fMRI噪声具有l/f样光谱特性的假设，并研究1/f噪声的仪器和生理来源，使用幻像和生物“静息”fMRI数据；Ii)我们将开发和验证一套专注于记忆和学习多个领域的重复和负荷适应性的认知激活范式；iii)我们将在年轻（20-30岁）和老年（60-70岁）健康参与者中进行主要（N=48）因子设计的药理学MRI研究，每个参与者在随机、安慰剂对照实验中进行三次扫描。这些数据将使我们能够测试关于针对多巴胺能、胆碱能和gaba能递质系统的激动剂和拮抗剂药物的主要作用的假设，从而阐明大规模脑适应性的机制。我们还将检验关于衰老的主要影响以及药物和年龄对大脑适应性的相互作用的假设。