脑功能信号的Granger因果性分析方法

脑是如何选择和调整分布于全脑的神经活动以实现其功能的?为解决这个问题,近年来人们将推测变量间信息因果流向的Granger因果分析引入脑功能研究.首先介绍了Granger因果分析的基本原理及其在处理脑功能信号时的拓展算法——结构向量自回归模型(SVAR);综述了在对脑功能信号进行Granger因果分析时的技术问题;最后,...     

基于Granger因果分析的癫痫发作间歇期EEG在δ频段的功能连接模式

目的 应用频域Granger因果分析方法,研究颞叶癫痫发作间歇期16导脑电图(EEG)在与癫痫发作相关的δ频段的过度放电功能连接特性.方法 实验数据来自颞叶癫痫9例患者(6例左颞叶癫痫,3例右颞叶癫痫),9例正常对照受试者.记录每例颞叶癫痫受试者在发作间歇期的痫样放电、非痫样放电以及正常对照组的共3个状态的16导EEG;每个状态下各记录10个EEG数据段,每个数据段长度为20s,采样频率为200 Hz;应用带通滤波提取EEG的δ分量(1～4 Hz).应用频域Granger因果分析方法,分别计算痫样放电组、非痫样放电组和正常组10次记录的16通道EEGδ频段分量之间的频域因果度量平均值Iδ;分析以上3个组颞叶区(左颞叶癫痫:T3、T5,右颞叶癫痫:T4、T6)与额区(Fp1、Fp2、F3、F4)和顶区(C3、C4)之间EEG在δ频段的功能连接模式.结果 痫样放电组:下颞叶区(左下颞叶区T5,右下颞叶区T6)与额区、顶区之间Iδ在0.1323±0.0329～0.1670±0.0289;非痫样放电组:下颞叶区与额区、顶区之间的Iδ在0.0300±0.0130～0.0420±0.0072;正常对照组:下颞叶区与额区、顶区之间的Iδ在0.0153±0.0028～0.0193±0.0057.统计结果表明:痫样放电组下颞叶区与额区、顶区之间的Iδ值与非痫样放电组相比差异有统计学意义(P＜0.05),与正常对照组相比差异有统计学意义(P＜0.01);非痫样放电组下颞叶区与额叶、顶叶之间的Iδ值和正常对照组相比差异无统计学意义(P＞0.05).结论 颞叶癫痫发作间歇期在痫样放电状态下,EEGδ频段在下颞叶区与额区、顶区之间存在较强连接,过度放电从下颞叶区传递到额区和顶区.非痫样放电组和正常组的EEGδ频段,下颞叶区与额叶、顶叶之间连接弱,下颞叶区不是EEG信号传导的起始区.     

多体素模式分析在fMRI数据分析中的应用

用多体素模式分析（multivariatepatternanalysis,MVPA）方法分析功能磁共振成像（functionalmagneticresonanceimaging,fMRI）数据是近年来认知神经科学领域的研究热点。MVPA可分析大脑的解剖信息、大脑的功能活动以及人们的行为表现三者之间的关系,从而对人们某个认知行为背后的神经机制进行探讨。本文主要介绍以下内容：①如何用MVPA方法对fMRI数据进行处理和分析,主要包括数据预处理、特征提取和样本创建、分类器的选择与测试。②如何对MVPA的分析结果进行解释：结合认知心理学的相关知识对MVPA的分析结果进行解释。通过本文,可以对MVPA的基本理论,应用特点和发展趋势,以及如何用MVPA方法进行fMRI数据处理和分析有一个较为全面的认识。     

Testing the involvement of baroreflex during general anesthesia through Granger causality approach

Baroreflex sensitivity (BRS) is commonly assessed from spontaneous fluctuations of heart period (HP) and systolic arterial pressure (SAP) during general anesthesia. Unfortunately, general anesthesia depresses autonomic function and, consequently, spontaneous SAP variations could not be capable to drive HP changes, thus preventing the use of spontaneous variability to infer BRS. We applied two Granger causality approaches (F-test and Wald test) during two anesthesiological strategies (i.e. sevoflurane plus remifentanil or propofol plus remifentanil). We found a significant Granger-causality from SAP to HP independently of the anesthesiological strategy; thus suggesting that techniques estimating BRS from spontaneous variability can be utilized during general anesthesia.     

一种基于独立成份分析的fMRI时-空模型数据处理方法

独立成份分析(ICA)是信号处理领域中斯近发展起来的一种很有应用前景的方法，而脑功能磁共振(fMRI)信号的有效分离与识别是一个正在研究和试验之中的技术领域。因此，发展基于ICA的fMRI数据处理方法具有明显的理论价值和应用前景。本文首先介绍了ICA原理，分析了现行ICA—fMRI方法采用的信号与噪声的空域分布相互独立的信号模型所存在的明显不足，然后提出了微域中的信号与噪声的时域过程相互独立的fMRI信号模型，从而建立了一种新的fMRI数据处理方法：邻域独立成份相关法。合理的fMRI实验数据处理结果验证了新方法的合理性。     

基于多变量格兰杰因果关系的运动想象因效网络构建

运动想象神经活动规律的探索为脑损伤肢体瘫痪患者康复训练新方法研发等提供理论指导。基于格兰杰因果关系构建的因效网络是分析运动想象神经活动状态的重要工具,但是格兰杰因果关系只能反映两个变量之间的相互作用,而一个简单的运动想象过程也需要多个神经节点参与,针对该问题,本研究引入可反映一个集群中多个变量之间相互作用的多变量格兰杰因果分析,优化运动想象因效网络构建方法。针对4位受试者,利用多变量及传统格兰杰因果关系,分别构建同一人两种不同运动想象模式的因效网络,并提取网络特征进行运动想象模式分类。结果表明,基于多变量格兰杰因效网络进行4位受试者运动想象模式分类的正确率分别为90.4%、88.8%、91.1%、90.3%,基于格兰杰因效网络的正确率为88.5%、89.3%、90.2%、89.7%。与传统格兰杰因果关系相比,基于多变量格兰杰因果关系构建因效网络,能更准确地反映运动想象神经活动特征状态。     

Investigation of the effective connectivity of resting state networks in Alzheimer's disease: a functional MRI study combining independent components analysis and multivariate Granger causality analysis

Recent neuroimaging studies have shown that the cognitive and memory decline in patients with Alzheimer's disease (AD) is coupled with abnormal functions of focal brain regions and disrupted functional connectivity between distinct brain regions, as well as losses in small‐world attributes. However, the causal interactions among the spatially isolated, but functionally related, resting state networks (RSNs) are still largely unexplored. In this study, we first identified eight RSNs by independent components analysis from resting state functional MRI data of 18 patients with AD and 18 age‐matched healthy subjects. We then performed a multivariate Granger causality analysis (mGCA) to evaluate the effective connectivity among the RSNs. We found that patients with AD exhibited decreased causal interactions among the RSNs in both intensity and quantity relative to normal controls. Results from mGCA indicated that the causal interactions involving the default mode network and auditory network were weaker in patients with AD, whereas stronger causal connectivity emerged in relation to the memory network and executive control network. Our findings suggest that the default mode network plays a less important role in patients with AD. Increased causal connectivity of the memory network and self‐referential network may elucidate the dysfunctional and compensatory processes in the brain networks of patients with AD. These preliminary findings may provide a new pathway towards the determination of the neurophysiological mechanisms of AD. Copyright © 2012 John Wiley & Sons, Ltd.     

空间ICA在fMRI数据上的应用与分析

独立成分分析(ICA)技术试图将多维数据分解成若干个相互统计独立的分量。时间ICA和空间ICA都可以用于分析功能核磁共振成像(fMRI)数据。但由于fMRI数据空间维数远远大于时间维数，为计算方便，在分析fMRI数据时。则更多的使用空间ICA方法。本文在单任务激励实验中，利用ICA方法从fMRI数据中分离出若干个与任务相关的独立分量，其中包括与任务相关的恒定分量(CTR)和与任务相关的暂态分量(TTR)；通过将这些独立分量进行空间映射，得到了与任务相关的脑部激活区域。将此结果与SPM的分析比较，得到了一致的结果。在对结果的分析中，我们进一步指出了ICA方法的特点和局限性。     

一种基于fMRI激活实验数据分析的新方法

多种有关用统计学原理来分析mMRI数据这方面的方法已经被提出，并且得到了广泛的应用。这些分析 目的是为了产生一幅图像以用于确定任务相关的显著信号变化区。本文提出了一种控制－参考－频率法，该方法通过对时间序列的谱密度进行非参数估计来测试任务频率和血液动力学响应之间的关系，从而达到激活区定位的目的。之后本文通过分析同一组fMRI数据将这种方法和统计参数映射法和相关法进行了比较。实验结果证明了本文所提出的方法在探测由于运动任务所产生的激活区方面是有效的。该方法非常适合探测刺激任务是以周期序列的形式产生时的激活区。     

Logarithmic transformation for high-field BOLD fMRI data

Parametric statistical analyses of BOLD fMRI data often assume that the data are normally distributed, the variance is independent of the mean, and the effects are additive. We evaluated the fulfilment of these conditions on BOLD fMRI data acquired at 4 T from the whole brain while 15 subjects fixated a spot, looked at a geometrical shape, and copied it using a joystick. We performed a detailed analysis of the data to assess (a) their frequency distribution (i.e. how close it was to a normal distribution), (b) the dependence of the standard deviation (SD) on the mean, and (c) the dependence of the response on the preceding baseline. The data showed a strong departure from normality (being skewed to the right and hyperkurtotic), a strong linear dependence of the SD on the mean, and a proportional response over the baseline. These results suggest the need for a logarithmic transformation. Indeed, the log transformation reduced the skewness and kurtosis of the distribution, stabilized the variance, and made the effect additive, i.e. independent of the baseline. We conclude that high-field BOLD fMRI data need to be log-transformed before parametric statistical analyses are applied.     

Exploring fMRI data for periodic signal components

We use a Bayesian framework to detect periodic components in fMRI data. The resulting detector is sensitive to periodic components with a flexible number of harmonics and with arbitrary amplitude and phases of the harmonics. It is possible to detect the correct number of harmonics in periodic signals even if the fundamental frequency is beyond the Nyquist frequency. We apply the signal detector to locate regions that are highly affected by periodic physiological artifacts, such as cardiac pulsation.     

Evaluation of the effective connectivity of the dominant primary motor cortex during bimanual movement using Granger causality

Granger causality analysis of the whole brain, voxel-by-voxel, was applied to six right-handed subjects performing a classic bimanual movement, to describe the effective connectivity between the activated voxels in the left primary motor cortex (PMC) and other parts of the brain, by choosing the left PMC as a reference region. The results demonstrated that the left and right PMC interact during bimanual movement, and Granger causality mapping implied a possible cause–effect relationship. The supplementary motor area (SMA) and cerebellum were pre-activated during bimanual movement relative to the left PMC, confirming the prior qualitative results concerning the functions of the SMA and cerebellum in hand movements.     

Effective connectivity analysis of fMRI and MEG data collected under identical paradigms

Estimation of effective connectivity, a measure of the influence among brain regions, can potentially reveal valuable information about organization of brain networks. Effective connectivity is usually evaluated from the functional data of a single modality. In this paper we show why that may lead to incorrect conclusions about effective connectivity. In this paper we use Bayesian networks to estimate connectivity on two different modalities. We analyze structures of estimated effective connectivity networks using aggregate statistics from the field of complex networks. Our study is conducted on functional MRI and magnetoencephalography data collected from the same subjects under identical paradigms. Results showed some similarities but also revealed some striking differences in the conclusions one would make on the fMRI data compared with the MEG data and are strongly supportive of the use of multiple modalities in order to gain a more complete picture of how the brain is organized given the limited information one modality is able to provide.     

An improved multi-objective optimization-based CICA method with data-driver temporal reference for group fMRI data analysis

Group independent component analysis (GICA) has been successfully applied to study multi-subject functional magnetic resonance imaging (fMRI) data, and the group independent component (GIC) represents the commonality of all subjects in the group. However, some studies show that the performance of GICA can be improved by incorporating a priori information, which is not always considered when looking for GICs in existing GICA methods. In this paper, we propose an improved multi-objective optimization-based constrained independent component analysis (CICA) method to take advantage of the temporal a priori information extracted from all subjects in the group by incorporating it into the computational process of GICA for group fMRI data analysis. The experimental results of simulated and real data show that the activated regions and the time course detected by the improved CICA method are more accurate in some sense. Moreover, the GIC computed by the improved CICA method has a higher correlation with the corresponding independent component of each subject in the group, which means that the improved CICA method with the temporal a priori information extracted from the group can better reflect the commonality of the subjects. These results demonstrate that the improved CICA method has its own advantages in fMRI data analysis.     

A toolbox for intensity modulated radiation therapy optimization

We have designed a toolbox that provides an environment for testing radiotherapy optimization techniques, objective functions, and constraints. A set of three-dimensional (3D) pencil beam dose distributions have been computed for a cylindrical phantom. The 6 MV pencil beams were computed using a superposition-based dose engine commissioned for an Elekta SL20 linear accelerator. Due to the cylindrical symmetry of the phantom, the pencil beam dose distributions for any arbitrary beam angle can be determined by simply rotating the pencil beam data sets. Thus, the full accuracy is maintained without the need for additional dose calculations or large data storage requirements. In addition to the pencil beam data sets, tools are included for (1) rotating the pencil beams, (2) calculating the beam's eye view, (3) drawing structures, (4) writing the pencil beam dose data out to the optimizer, and (5) visualizing the optimized results. The pencil beam data sets and the corresponding tools are available for download at http://medschool.umaryland.edu/departments/radiationoncology/pencilbeam/. With this toolbox, researchers will have the ability to rapidly test new optimization techniques and formulations for intensity modulated radiation therapy and 3D conformal radiotherapy.     

一种用于功能磁共振的同步器

目的 功能磁共振（functional magnetic resonance imaging,fMRI）在采集图像数据时,有两种同步信号输出方式：其一,每采集一层图像输出一个同步信号;其二,采集一幅脑图输出一个同步信号.由于第一种输出方式中同步信号过于密集,导致刺激计算机无法及时记录同步信号,从而失去同步.为此本文设计出一种基于复杂可编程逻辑器件（complex programmable logic devices,CPLD）的单参数（同步参数）同步器.方法 采集一幅脑图时,该同步器对第一种同步信号的上升沿进行计数,计数期间输出维持高电平,直到最后一个同步信号到来,才把输出拉低为低电平,从而实现第一种同步信号到第二种同步信号的转换.然后用Quartus 9.1对信号转换进行仿真以验证其功能.结果 经过CPLD处理器后,第一种同步信号被处理成第二种同步信号.此外,基于本设计做出的同步器亦成功应用于GE Sigma3.0T.结论 本文设计了具有同步信号转换能力的同步器,成功实现了输出方式的转变.该同步器达到同步信号转换要求,并具有较好的兼容性.     

Theoretical optimization of multi-echo fMRI data acquisition

Simulations are used to optimize multi-echo fMRI data acquisition for detection of BOLD signal changes in this study. Optimal sequence design (echo times and sampling period (receiver bandwidth)) and the variation in sensitivity between tissues with different baseline T*(2) are investigated, taking into account the effects of physiological noise and non-exponential signal decay. In the case of a single echo, for normally distributed, uncorrelated noise, the results indicate that the sampling period should be made as long as possible (so as to produce an acceptable level of image distortion), up to a maximum sampling period of 3T*(2), (i.e. optimum TE = 1.5T*(2)). Combining the signal from multiple echoes using weighted summation improves the contrast-to-noise ratio (CNR), at a reduced optimum echo interval. If the BOLD effect causes a constant change in relaxation rate, DeltaR*(2), independent of the tissue R*(2), then a multi-echo acquisition causes considerable variation in sensitivity to BOLD signal changes with tissue T*(2), so that if the sequence is optimized for a target tissue T*(2) it will be more sensitive to BOLD signal changes in tissues with shorter T*(2) values. Fitting for DeltaR*(2) reduces the CNR, and when using this approach, the shortest echo time interval should be used, down to a limit of about 0.3T*(2), and as many echoes as possible within the constraints of TR or hardware limitations should be collected. It is also shown that the optimal sequence will remain optimum or close to optimum irrespective of whether there are physiological noise contributions.