Drug addiction can lead to many health-related problems and social concerns. Researchers are interested in the association between long-term drug usage and abnormal functional connectivity. Functional connectivity obtained from functional magnetic resonance imaging data promotes a variety of fundamental understandings in such association. Due to the complex correlation structure and large dimensionality, the modeling and analysis of the functional connectivity from neuroimage are challenging. By proposing a spatio-temporal model for multi-subject neuroimage data, we incorporate voxel-level spatio-temporal dependencies of whole-brain measurements to improve the accuracy of statistical inference. To tackle large-scale spatio-temporal neuroimage data, we develop a computational efficient algorithm to estimate the parameters. Our method is used to first identify functional connectivity, and then detect the effect of cocaine use disorder (CUD) on functional connectivity between different brain regions. The functional connectivity identified by our spatio-temporal model matches existing studies on brain networks, and further indicates that CUD may alter the functional connectivity in the medial orbitofrontal cortex subregions and the supplementary motor areas.

吸毒成瘾会导致许多与健康相关的问题和社会问题。研究人员对长期用药与功能连接异常之间的关联感兴趣。从功能磁共振成像数据获得的功能连接促进了对这种关联的各种基本理解。由于复杂的相关结构和大维度，神经图像功能连接的建模和分析具有挑战性。通过提出多主体神经图像数据的时空模型，我们结合了全脑测量的体素级时空依赖性，以提高统计推断的准确性。为了处理大规模时空神经图像数据，我们开发了一种计算有效的算法来估计参数。我们的方法首先用于识别功能连接，然后检测可卡因使用障碍（CUD）对不同大脑区域之间功能连接的影响。我们的时空模型确定的功能连接与现有的大脑网络研究相匹配，并进一步表明 CUD 可能会改变内侧眶额皮质分区和辅助运动区的功能连接。