Abstract

Most existing aging studies using functional MRI (fMRI) are based on cross-sectional data but misinterpreted their findings (i.e., age-related differences) as longitudinal outcomes (i.e., aging-related changes). To delineate aging-related changes the of human cerebral cortex, we employed the resting-state fMRI (rsfMRI) data from 24 healthy elders in the PREVENT-AD cohort, obtaining five longitudinal scans per subject. Cortical spontaneous activity is measured globally with three rsfMRI metrics including its amplitude, homogeneity, and homotopy at three different frequency bands (slow-5: 0.02–0.03 Hz, slow-4: 0.03–0.08 Hz, and slow-3 band: 0.08–0.22 Hz). General additive mixed models revealed a universal pattern of the aging-related changes for the global cortical spontaneous activity, indicating increases of these rsfMRI metrics during aging. This aging pattern follows specific frequency and spatial profiles where higher slow bands show more non-linear curves and the amplitude exhibits more extensive and significant aging-related changes than the connectivity. These findings provide strong evidence that cortical spontaneous activity is aging globally, inspiring its clinical utility as neuroimaging markers for neruodegeneration disorders.

Highlights

  • MRI has advanced brain aging research and revealed consistent aging patterns of thinning cortical thickness and shrinking surface area (Elliott, 2020)

  • While early resting-state fMRI (rsfMRI) studies focused on a single frequency (e.g., 0.01–0.1 Hz) (Biswal, 2012), Zuo et al were the first to decompose the rsfMRI signals into multiple frequency intervals according to the neural oscillation law (Zuo et al, 2010a)

  • Combining the previous findings, the author hypothesized that the global cortical spontaneous activity (CSA) will show a universal aging pattern while higher frequency bands have more complex aging curves and more local metrics age more extensively across the frequency bands

Read more

Summary

INTRODUCTION

MRI has advanced brain aging research and revealed consistent aging patterns of thinning cortical thickness and shrinking surface area (Elliott, 2020). Aging has been demonstrated to have significant effects on spontaneous brain activity measured by rsfMRI the direction of these effects (i.e., increase vs decrease) remains inconsistent among the studies (Ferreira and Busatto, 2013; Foo et al, 2020) This might be an indication of the limited reliability of the most common functional connectivity metrics (Noble et al, 2019) and the lack of longitudinal rsfMRI dataset (Zuo et al, 2014) in previous studies. While early rsfMRI studies focused on a single frequency (e.g., 0.01–0.1 Hz) (Biswal, 2012), Zuo et al were the first to decompose the rsfMRI signals into multiple frequency intervals according to the neural oscillation law (Zuo et al, 2010a) They demonstrated the specificity of the frequency band to human basal ganglia’s spontaneous activity by directly comparing the rsfMRI amplitudes between two slow bands. Combining the previous findings, the author hypothesized that the global CSA will show a universal aging pattern while (relatively speaking) higher frequency bands have more complex aging curves and (relatively speaking) more local metrics age more extensively across the frequency bands

Participants and MRI Data
Data Preprocessing
CSA Metric Computation
Aging Curve Modeling
RESULTS
DISCUSSIONS
ETHICS STATEMENT
Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call