Tracking functional network connectivity dynamics in the elderly

Abstract

Introduction: Functional magnetic resonance imaging (fMRI) has shown that aging disturbs healthy brain organization and functional connectivity. However, how this age-induced alteration impacts dynamic brain function interaction has not yet been fully investigated. Dynamic function network connectivity (DFNC) analysis can produce a brain representation based on the time-varying network connectivity changes, which can be further used to study the brain aging mechanism for people at different age stages.

Method: This presented investigation examined the dynamic functional connectivity representation and its relationship with brain age for people at an elderly stage as well as in early adulthood. Specifically, the resting-state fMRI data from the University of North Carolina cohort of 34 young adults and 28 elderly participants were fed into a DFNC analysis pipeline. This DFNC pipeline forms an integrated dynamic functional connectivity (FC) analysis framework, which consists of brain functional network parcellation, dynamic FC feature extraction, and FC dynamics examination.

Results: The statistical analysis demonstrates that extensive dynamic connection changes in the elderly concerning the transient brain state and the method of functional interaction in the brain. In addition, various machine learning algorithms have been developed to verify the ability of dynamic FC features to distinguish the age stage. The fraction time of DFNC states has the highest performance, which can achieve a classification accuracy of over 88% by a decision tree.

Discussion: The results proved there are dynamic FC alterations in the elderly, and the alteration was found to be correlated with mnemonic discrimination ability and could have an impact on the balance of functional integration and segregation.

Keywords: aging; dynamic functional network connectivity; functional integration and segregation; graph theory; mnemonic discrimination ability.

Figure 1

The dynamic functional connectivity analysis pipeline. The timeseries signal was extracted from the network regions recognized from the group ICA parcellation method. Then, the regional timeseries were decomposed with a sliding window scheme for a time-varying function network connectivity (FNC) estimation. Those FNC matrixes were fed into a clustering algorithm to obtain different transient brain states by forming a cluster centroid. After that, two types of dynamic features were calculated based on the acquired transient states and temporal signals. Finally, statistical and machine learning methods were applied to verify the extracted dynamic FC features.

Figure 2

Spatial maps of the 60 independent components result from the entire group (28 elderly and 34 younger adults). The coordinates denote the max peak location of functional domains, and different colors pass spatial information. AUD, auditory domain; CC, cognitive control domain; DMN, default mode domain; SC, subcortical; SM, sensorimotor domain; VIS, visual domain.

Figure 3

(A) Static functional network connectivity between 60 independent components resulting in 1770 ( 60 × (60-1)/2 )connectivity pairs for the entire group. Asterisks indicate significant differences between the elderly and younger groups. (B) Circle plot of significant static functional network connectivity differences of 6 domain between the elderly adult and younger group.

Figure 4

(A) Four functional connectivity states as well as their frequencies across all participants using the group-ICA method. (B) Group differences of the 6 selected brain networks between elderly and younger adults in the 4 states. AUD, auditory domain; CC, cognitive control domain; DM, default mode domain; SC, subcortical; SM, sensorimotor domain; VIS, visual domain.

Figure 5

Dynamic connectivity feature analysis for the elderly and younger groups. (A) The fraction of time the occurrence of DFC state 2 and state 3 has significant between group difference. The elder prefer state 2 (P < 0.05) and state 3 (P < 0.05). (B) The dwell time. Once again, the senior group is more like to stay within state 2 and state 3. (C) The number of transition between states. There is no significant difference in the number of state transition between two groups. (D) State Transition Probability matrix. Comparing with younger adults, the older people more inclined to switch to state 3 when they are in state 1, 2, or 4. However, they are also more likely to transfer to other state when they are entering state 3 than younger people. *p < 0.05, ***p < 0.001, and ****p < 0.001 FDR corrected.

Figure 6

Time course of multiple dynamic measures for the different age groups.

Figure 7

Time varying curve of the three dynamic measures of DMN network: local efficiency, vulnerable, nodal betweenness in age-different group, where no matter which metric elder people are lowest.