Integrative learning for population of dynamic networks with covariates.
Integrative learning for population of dynamic networks with covariates.
复制标题
DOI：
10.1016/j.neuroimage.2021.118181
复制
发表时间：
2021-08-01
期刊：
影响因子：
5.7
通讯作者：
McGregor KM
中科院分区：
文献类型：
--
作者：
Kundu S;Ming J;Nocera J;McGregor KM
尽管关于动态连接方法的文献正在迅速增长，但主要的焦点是对每个个体进行单独的网络估计，这无法利用信息的共同模式。我们提出了新的图论方法来估计动态网络的总体，这些网络能够以无监督的方式在多个异构样本中借用信息，并由协变量信息指导。具体来说，我们开发了一个贝叶斯产品混合模型，该模型在每次扫描时施加独立的混合先验，并使用协变量对混合权重进行建模，从而产生旨在汇集信息的时变样本簇。计算采用了一种高效的期望最大化算法。大量的仿真研究表明，与现有的动态连接方法相比，在恢复真实动态网络方面取得了巨大的进步。对行为干预的fMRI块任务数据的分析揭示了具有相似动态连通性的个体亚群，并确定了集中在生物学可解释的大脑区域的干预相关动态网络变化。相比之下，现有的动态连接方法能够检测到随着时间的推移连通性的最小变化或没有变化，这在生物学上似乎是不现实的，并且突出了无法系统地跨样本借用信息所带来的挑战。
Although there is a rapidly growing literature on dynamic connectivity methods, the primary focus has been on separate network estimation for each individual, which fails to leverage common patterns of information. We propose novel graph-theoretic approaches for estimating a population of dynamic networks that are able to borrow information across multiple heterogeneous samples in an unsupervised manner and guided by covariate information. Specifically, we develop a Bayesian product mixture model that imposes independent mixture priors at each time scan and uses covariates to model the mixture weights, which results in time-varying clusters of samples designed to pool information. The computation is carried out using an effcient Expectation-Maximization algorithm. Extensive simulation studies illustrate sharp gains in recovering the true dynamic network over existing dynamic connectivity methods. An analysis of fMRI block task data with behavioral interventions reveal subgroups of individuals having similar dynamic connectivity, and identifies intervention-related dynamic network changes that are concentrated in biologically interpretable brain regions. In contrast, existing dynamic connectivity approaches are able to detect minimal or no changes in connectivity over time, which seems biologically unrealistic and highlights the challenges resulting from the inability to systematically borrow information across samples.
登录
查看更多内容
影响因子：
5.7
作者：
Lindquist MA;Xu Y;Nebel MB;Caffo BS
通讯作者：
Caffo BS
影响因子：
5.7
作者：
Monti, Ricardo Pio;Hellyer, Peter;Montana, Giovanni
通讯作者：
Montana, Giovanni
影响因子：
5.7
作者：
Hutchison RM;Womelsdorf T;Allen EA;Bandettini PA;Calhoun VD;Corbetta M;Della Penna S;Duyn JH;Glover GH;Gonzalez-Castillo J;Handwerker DA;Keilholz S;Kiviniemi V;Leopold DA;de Pasquale F;Sporns O;Walter M;Chang C
通讯作者：
Chang C
影响因子：
5.7
作者：
Davey J;Thompson HE;Hallam G;Karapanagiotidis T;Murphy C;De Caso I;Krieger-Redwood K;Bernhardt BC;Smallwood J;Jefferies E
通讯作者：
Jefferies E
影响因子：
3.7
作者：
Balsters, Joshua H.;Whelan, Christopher D.;Ramnani, Narender
通讯作者：
Ramnani, Narender