A hierarchical Bayesian model to find brain-behaviour associations in incomplete data sets.
A hierarchical Bayesian model to find brain-behaviour associations in incomplete data sets.
复制标题
DOI：
10.1016/j.neuroimage.2021.118854
复制
发表时间：
2022-04-01
期刊：
影响因子：
5.7
通讯作者：
Mourao-Miranda J
中科院分区：
文献类型：
--
作者：
Ferreira FS;Mihalik A;Adams RA;Ashburner J;Mourao-Miranda J
典型相关分析（CCA）及其正则化版本已广泛应用于神经影像学社区，以揭示两种数据模式（例如，脑成像和行为）之间的多变量关联。然而，这些方法有固有的局限性：(1)关于关联的统计推断往往不稳健；(2)每个数据模态内的关联没有建模；(3)缺失值需要输入或移除。组因子分析（GFA）是一种分层模型，通过提供贝叶斯推理和建模特定于模态的关联来解决前两个限制。在这里，我们提出了GFA的扩展，以处理丢失的数据，并强调GFA可以用作预测模型。我们将GFA应用于人类连接组计划（HCP）的脑连接和非成像测量的合成和真实数据。在合成数据中，GFA揭示了潜在的共享和特定因素，并正确预测了完整和不完整数据集中未观察到的数据模式。在HCP数据中，我们确定了四个相关的共享因素，捕捉情绪、酒精和药物使用、认知、人口统计学和精神病理学测量与默认模式、额顶叶控制、背侧和腹侧网络和脑岛之间的关联，以及描述大脑连接内部关联的两个因素。此外，GFA预测了一组来自大脑连通性的非成像测量。这些发现在完整和不完整的数据集中是一致的，并重复了先前文献中的发现。GFA是一种很有前途的工具，可用于揭示基准数据集（如HCP）中多个数据模式之间和内部的关联，并且很容易扩展到更复杂的模型，以解决更具挑战性的任务。
Canonical Correlation Analysis (CCA) and its regularised versions have been widely used in the neuroimaging community to uncover multivariate associations between two data modalities (e.g., brain imaging and behaviour). However, these methods have inherent limitations: (1) statistical inferences about the associations are often not robust; (2) the associations within each data modality are not modelled; (3) missing values need to be imputed or removed. Group Factor Analysis (GFA) is a hierarchical model that addresses the first two limitations by providing Bayesian inference and modelling modality-specific associations. Here, we propose an extension of GFA that handles missing data, and highlight that GFA can be used as a predictive model. We applied GFA to synthetic and real data consisting of brain connectivity and non-imaging measures from the Human Connectome Project (HCP). In synthetic data, GFA uncovered the underlying shared and specific factors and predicted correctly the non-observed data modalities in complete and incomplete data sets. In the HCP data, we identified four relevant shared factors, capturing associations between mood, alcohol and drug use, cognition, demographics and psychopathological measures and the default mode, frontoparietal control, dorsal and ventral networks and insula, as well as two factors describing associations within brain connectivity. In addition, GFA predicted a set of non-imaging measures from brain connectivity. These findings were consistent in complete and incomplete data sets, and replicated previous findings in the literature. GFA is a promising tool that can be used to uncover associations between and within multiple data modalities in benchmark datasets (such as, HCP), and easily extended to more complex models to solve more challenging tasks.
登录
查看更多内容
影响因子：
2.7
作者：
Lee WH;Moser DA;Ing A;Doucet GE;Frangou S
通讯作者：
Frangou S
影响因子：
3
作者：
Sui, Jing;Adali, Tuelay;Yu, Qingbao;Chen, Jiayu;Calhoun, Vince D.
通讯作者：
Calhoun, Vince D.
影响因子：
82.9
作者：
Drysdale AT;Grosenick L;Downar J;Dunlop K;Mansouri F;Meng Y;Fetcho RN;Zebley B;Oathes DJ;Etkin A;Schatzberg AF;Sudheimer K;Keller J;Mayberg HS;Gunning FM;Alexopoulos GS;Fox MD;Pascual-Leone A;Voss HU;Casey BJ;Dubin MJ;Liston C
通讯作者：
Liston C
DOI：
10.1109/tnnls.2014.2376974
发表时间：
2015-09-01
影响因子：
10.4
作者：
Klami, Arto;Virtanen, Seppo;Kaski, Samuel
通讯作者：
Kaski, Samuel
影响因子：
4.6
作者：
Mihalik, Agoston;Ferreira, Fabio S.;Mourao-Miranda, Janaina
通讯作者：
Mourao-Miranda, Janaina