Identification of Alzheimer’s Disease Progression Stages Using Topological Measures of Resting-State Functional Connectivity Networks: A Comparative Study

Figure 3

Violin plots of ANCOVA analyses on network-level topological metrics, including global efficiency and characteristic path length. No significant differences were found in rs-FC networks constructed with DTW, in terms of the both metrics. As for Pearson correlation, only MCI and AD groups were significantly identified according to the metric of characteristic path length. In FC networks constructed with GIG-ICA, significant differences between HC, AD, and MCI groups were found in terms of global efficiency.