Attention deficit/hyperactivity disorder (ADHD) is characterized by inattention, hyperactivity and impulsivity. In this study, we investigated group differences in dynamic functional connectivity (dFC) between 113 children with inattentive (46 ADHD_I) and combined (67 ADHD_C) ADHD and 76 typically developing (TD) children using resting-state functional MRI data. For dynamic connectivity analysis, the data were first decomposed into 100 independent components, among which 88 were classified into eight well-known resting-state networks (RSNs). Three discrete FC states were then identified using k-means clustering and used to estimate transition probabilities between states in both patient and control groups using a hidden Markov model. Our results showed state-dependent alterations in intra and inter-network connectivity in both ADHD subtypes in comparison with TD. Spending less time than healthy controls in state 1, both ADHD_I and ADHD_C were characterized with weaker intra-hemispheric connectivity with functional asymmetries. In this state, ADHD_I further showed weaker inter-hemispheric connectivity. The patients spent more time in state 2, exhibiting characteristic abnormalities in corticosubcortical and corticocerebellar connectivity. In state 3, a less frequently state observed across the ADHD and TD children, ADHD_C was differentiated from ADHD_I by significant alterations in FC between bilateral temporal regions and other brain areas in comparison with TD. Across all three states, several strategic brain regions, mostly bilateral, exhibited significant alterations in both static functional connectivity (sFC) and dFC in the ADHD groups compared to TD, including inferior, middle and superior temporal gyri, middle frontal gyri, insula, anterior cingulum cortex, precuneus, calcarine, fusiform, superior motor area, and cerebellum. Our results show distributed abnormalities in sFC and dFC between different large-scale RSNs including cortical and subcortical regions in both ADHD subtypes compared to TD. Our findings show that the dynamic changes in brain FC can better explain the underlying pathophysiology of ADHD such as deficits in visual cognition, attention, memory and emotion processing, and cognitive and motor control.

注意缺陷/多动障碍 (ADHD) 的特征是注意力不集中、多动和冲动。在这项研究中，我们调查了 113 名注意力不集中 (46 ADHD _I ) 和合并 (67 ADHD _C )儿童在动态功能连接 (dFC) 方面的组差异) ADHD 和 76 名典型发育 (TD) 儿童使用静息状态功能 MRI 数据。对于动态连通性分析，首先将数据分解为 100 个独立的组件，其中 88 个被分类为 8 个著名的静止状态网络 (RSN)。然后使用 k 均值聚类识别三个离散的 FC 状态，并使用隐马尔可夫模型估计患者和对照组状态之间的转移概率。我们的结果显示，与 TD 相比，两种 ADHD 亚型的网络内和网络间连接的状态依赖性改变。在状态 1 中花费的时间比健康对照少，ADHD _I和 ADHD _C的特征是半球内连接较弱，功能不对称。在这种状态下，多动症_我进一步显示了较弱的半球间连通性。患者在状态 2 中停留的时间更长，表现出皮质下皮质和皮质小脑连接的特征性异常。在状态 3（在 ADHD 和 TD 儿童中观察到的较少见的状态）中，ADHD _C与 ADHD _I不同与 TD 相比，双侧颞区和其他大脑区域之间 FC 的显着变化。在所有三个状态中，与 TD 相比，ADHD 组的几个战略性大脑区域（主要是双侧）在静态功能连接 (sFC) 和 dFC 方面都表现出显着改变，包括颞下回、中回和颞上回、额中回、岛叶、前回扣带皮层、楔前叶、钙质、梭形、上运动区和小脑。我们的结果显示，与 TD 相比，两种 ADHD 亚型的不同大规模 RSN 之间的 sFC 和 dFC 分布异常，包括皮质和皮质下区域。我们的研究结果表明，大脑 FC 的动态变化可以更好地解释 ADHD 的潜在病理生理学，例如视觉认知、注意力、记忆和情绪处理的缺陷，