Zhao Qinghua, Lu Hong, Metmer Hichem, Li Will X Y, Lu Jianfeng
School of Computer Science and Engineering, Nanjing University of Science and Technology, Nanjing 210094, China.
School of Computer Science and Engineering, Nanjing University of Science and Technology, Nanjing 210094, China.
Brain Res. 2018 Jan 1;1678:262-272. doi: 10.1016/j.brainres.2017.10.025. Epub 2017 Oct 25.
Investigating the early Alzheimer's disease (AD) more emphasizes sensitive and specific biomarkers, which can help the clinicians to monitor the progression and treatments of AD. Among these biomarkers, default mode network (DMN) functional connectivity is gaining more attention as a potential noninvasive biomarker to diagnose incipient Alzheimer's disease. However, besides changed functional connectivity of DMN, other functional networks haven't yet been examined systematically. Recent brain imaging studies reported that a number of reproducible and robust functional networks, which were distributed in distant neuroanatomic areas. Inspired by these works, in this paper, we apply sparse representation to the whole brain signals to identify these reproducible networks and detect partly affected brain regions of Alzheimer's disease, then adopt sparse inverse covariance estimation (SICE) approach to investigate the changed functional connectivity of intrinsic connectivity networks. Our experimental results show that besides DMN, AD is also affected by others large scale functional brain networks and regions, e.g., executive control network (ECN), frontoparietal network (FPN), where in the superior frontal gyrus (SFGmed) and middle frontal gyrus (MFG) of ECN and in the part paracentral Lobule (PCL) of FPN have an increased functional connectivity, as well as in the Superior Parietal Gyrus (SPG) regions of FPN has shown decreased connectivity. The results may suggest AD is associated with larger scale functional networks and causes the functional connectivity change of many different brain regions. It also proves that these networks may sometimes work together to perform tasks, and such changed functional connectivity may provide a useful baseline for early AD diagnosis.
对早期阿尔茨海默病（AD）的研究更加强调敏感且特异的生物标志物，这些标志物有助于临床医生监测AD的进展及治疗情况。在这些生物标志物中，默认模式网络（DMN）功能连接作为一种潜在的非侵入性生物标志物，用于诊断早期阿尔茨海默病正受到越来越多的关注。然而，除了DMN功能连接的改变，其他功能网络尚未得到系统研究。最近的脑成像研究报告称，一些可重复且稳健的功能网络分布在遥远的神经解剖区域。受这些研究的启发，在本文中，我们将稀疏表示应用于全脑信号，以识别这些可重复的网络并检测阿尔茨海默病部分受影响的脑区，然后采用稀疏逆协方差估计（SICE）方法来研究内在连接网络功能连接的变化。我们的实验结果表明，除了DMN，AD还受到其他大规模功能性脑网络和区域的影响，例如执行控制网络（ECN）、额顶网络（FPN），其中ECN的额上回（SFGmed）和额中回（MFG）以及FPN的中央旁小叶（PCL）部分功能连接增加，而FPN的顶上小叶（SPG）区域连接性降低。这些结果可能表明AD与更大规模的功能网络相关，并导致许多不同脑区的功能连接发生变化。这也证明这些网络有时可能协同执行任务，而这种功能连接的变化可能为早期AD诊断提供有用的基线。
