Increased functional connectivity patterns in mild Alzheimer's disease: A rsfMRI study

Abstract

Background: Alzheimer's disease (AD) is the most common age-related neurodegenerative disorder. In view of our rapidly aging population, there is an urgent need to identify Alzheimer's disease (AD) at an early stage. A potential way to do so is by assessing the functional connectivity (FC), i.e., the statistical dependency between two or more brain regions, through novel analysis techniques.

Methods: In the present study, we assessed the static and dynamic FC using different approaches. A resting state (rs)fMRI dataset from the Alzheimer's disease neuroimaging initiative (ADNI) was used (n = 128). The blood-oxygen-level-dependent (BOLD) signals from 116 regions of 4 groups of participants, i.e., healthy controls (HC; n = 35), early mild cognitive impairment (EMCI; n = 29), late mild cognitive impairment (LMCI; n = 30), and Alzheimer's disease (AD; n = 34) were extracted and analyzed. FC and dynamic FC were extracted using Pearson's correlation, sliding-windows correlation analysis (SWA), and the point process analysis (PPA). Additionally, graph theory measures to explore network segregation and integration were computed.

Results: Our results showed a longer characteristic path length and a decreased degree of EMCI in comparison to the other groups. Additionally, an increased FC in several regions in LMCI and AD in contrast to HC and EMCI was detected. These results suggest a maladaptive short-term mechanism to maintain cognition.

Conclusion: The increased pattern of FC in several regions in LMCI and AD is observable in all the analyses; however, the PPA enabled us to reduce the computational demands and offered new specific dynamic FC findings.

Keywords: Alzheimer’s disease; dynamic functional connectivity; functional connectivity; mild cognitive impairment; point process analysis; resting state fMRI.

Figure 1

displays the methods used in the study. First, the fMRI and T1 image acquisitions were downloaded from the Alzheimer’s disease neuroimaging initiative (ADNI) database. Then the images were preprocessed using the DPARSF pipelines. Timeseries from 116 regions of interest (ROIs) were extracted using the automated anatomical labeling (AAL) atlas. These 116 time series were used to perform three main analyses displayed on the right side of the figure namely, functional connectivity (FC), sliding window analysis (SWA), and point process analysis (PPA). The top right side of the figure shows the FC where the Pearson’s correlation coefficient of each pair of regions was computed using the mean whole signal of each time series; the middle right side of the figure shows the SWA, that consisted of dividing the time series into non-overlapping windows, computing the FC for each window and determining the variability in FC across windows; the third plot on the right side of the image shows the PPA, this is a single frame analysis where points that surpass the threshold of 1 SD of each time series were selected, coincident points between pair of regions where summed and displayed in a matrix of addition. The fourth plot on the right side of the image displays the graph measures conducted in the study. To test the statistical significance of each analysis, a one-way-ANOVA with multiple comparison tests was conducted, corrected with Bonferroni at p < 0.05.

Figure 2

The figure represents the window size analysis before conducting the sliding window correlation analysis. (A) Global FC as a function of window size. The Y-axis represents the mean FC or correlation of all the ROI to ROI correlations windows of a participant with AD. The X-axis shows 140 time points of the whole time series. Each time point represents 3 s (TR = 3 s). The mean correlation varies as a function of window size. Shorter windows present a lower mean correlation while windows from 30 or above present a higher correlation. (B) The Y-axis represents the mean standard deviation (SD) of the SD of all the ROI to ROI correlations across windows of a participant with AD. The X-axis shows that this mean SD varies as a function of window size. Shorter windows present a higher variability while windows of 30 time points or above present a lower SD. (C) The plot shows the difference between the mean correlation of all ROI to ROIs using a certain window size and the global static (D) The difference in the variability among windows using window sizes higher than 30 fluctuates around 0, meaning that results in SWA are almost the same, no matter the window length chosen.

Figure 3

shows no significant differences between groups that were found globally in FC (subplot on the left), SWA (subplot in the middle), and PPA (subplot on the right).

Figure 4

shows between-group significant differences between specific brain networks in static functional connectivity (sFC), point process analysis (PPA), and variability in functional correlation across windows applying sliding window correlation analysis (SWA). Dorsal, dorsal network; DMN, Default Mode Network; SM, Somatosensory network; VLPC, Ventrolateral prefrontal cortex; BG, Basal Ganglia; Thala, thalamus; VI, Visual network I; VII, Visual network II; Cereb, Cerebellum; CEN, Central Executive Network; AN, Auditory Network.