Dynamic embedding time series models in functional brain imaging

项目摘要 我们将开发新的大规模网络数据动态嵌入模型，重点是动态连接。 从人体功能磁共振获得的非平稳多变量时间序列的活动矩阵 图像(功能磁共振成像)。我们建议将大脑网络建模为2D曲面，其中表面测地线给出 连接信息。我们的方法将绕过分隔符的使用，并更准确地计算 体素水平上脑功能网络的进化动力学。 我们建议从1206名受试者的数据集中建立动态变化的脑功能网络 人类连接组计划(HCP)数据库包含T1加权磁共振图像(MRI)， 弥散磁共振成像(DMRI)和任务和静息功能磁共振成像(FMRI)。Mri和dmri将结合使用 利用功能磁共振成像建立更多的RefiNed动态连接模型。使用hcp数据库中的243对双胞胎，我们 将决定网络表型与行为、认知及其遗传关联的特殊fic。这项研究将 为研究团体提供大脑网络遗传力图，以及一个多功能的开源 用于对动态变化的大规模脑网络进行建模和可视化的算法工具箱。

Project Summary We will develop new large-scale dynamic embedding models of network data with a focus on dynamic connec- tivity matrices from non-stationary multivariate time series obtained from human functional magnetic resonance images (fMRI). We propose to model brain networks as 2D curved surfaces, where the surface geodesics give connectivity information. Our approach will bypass the use of parcellations and more accurately evaluate the evolutionary dynamics of functional brain networks at the voxel level. We propose to build dynamically changing functional brain networks from a dataset with 1206 subjects from the Human Connectome Project (HCP) database containing T1-weighted magnetic resonance images (MRI), diffusion MRI (dMRI) and task and resting-state functional MRI (fMRI). MRI and dMRI will be used in conjunction with fMRI in building more refined dynamic connectivity models. Using 243 pairs of twins in the HCP database, we will determine network phenotypes specific to behavior, cognition and their genetic associations. This study will provide the research community with the brain network heritability maps and as well as a versatile open-source toolbox of algorithms for modeling and visualizing dynamically changing large-scale brain networks.