Effects of healthy ageing on electrophysiological brain networks: a normative model
Poster No:
898
Submission Type:
Abstract Submission
Authors:
Chetan Gohil1, Oliver Kohl1, Jemma Pitt1, Mats Van Es1, Andrew Quinn2, Diego Vidaurre3, Martin Turner1, Anna Nobre4, Mark Woolrich1
Institutions:
1University of Oxford, Oxford, United Kingdom, 2University of Birmingham, Birmingham, United Kingdom, 3Aarhus University, Aarhus, Denmark, 4Yale University, New Haven, United States
First Author:
Co-Author(s):
Introduction:
With an increasing proportion of elderly people globally, there is a pressing need to characterise and differentiate healthy changes in the brain from those that lead to cognitive decline.
Using magnetoencephalogy (MEG) recordings, we characterised how oscillatory networks of brain activity changed with age using a large healthy cohort (N=612, 18-88 years old). These oscillatory networks reflect the underlying neurophysiology of the brain [1]. Consequently, pathological changes in neurophysiology, e.g. synaptic health, may be detectable via alterations to these networks.
An open question in the study of ageing is how some individuals preserve their cognitive health [2]. To explore this, we related the oscillatory networks to cognitive performance to identify network features associated with preserving cognitive health.
Using magnetoencephalogy (MEG) recordings, we characterised how oscillatory networks of brain activity changed with age using a large healthy cohort (N=612, 18-88 years old). These oscillatory networks reflect the underlying neurophysiology of the brain [1]. Consequently, pathological changes in neurophysiology, e.g. synaptic health, may be detectable via alterations to these networks.
An open question in the study of ageing is how some individuals preserve their cognitive health [2]. To explore this, we related the oscillatory networks to cognitive performance to identify network features associated with preserving cognitive health.
Methods:
Dataset: In this work, we studied the MEG recordings of a large cross-sectional cohort (N=612, 18-88 years old) from the Cam-CAN dataset [3,4]. The participants were deemed to be cognitively healthy. Each participant had an 8-minute resting-state scan with eyes closed. Each participant also undertook a battery of cognitive tests across five domains: executive function; language; emotional processing; memory; processing speed. The cognitive test scores were reduced to a single metric summarising cognitive health using principal component analysis (PCA), see Figure 1D.
Source reconstruction: The sensor-level MEG recordings were source reconstructed to the voxel level using a volumetric unit-noise-gain invariant LCMV (linearly constrained minimum variance) beamformer [5] using the osl-ephys toolbox [6]. We parcellated the data to 52 regions of interest using the Kohl atlas [7].
Hidden Markov Modelling (HMM): We adopting the time-delay embedded HMM approach to identify transient networks of oscillatory activity [8]. The HMM segments the time series into categorical states of unique spatio-spectral activity. We identified 10 transients network states of oscillatory activity. This work used the osl-dynamics toolbox [8].
Spectral calculations: We calculated cross power spectra using the parcel data and inferred state time courses with a multitaper (2 s window, 0% overlap, 7 DPSS tapers, 4 Hz time half bandwidth).
Source reconstruction: The sensor-level MEG recordings were source reconstructed to the voxel level using a volumetric unit-noise-gain invariant LCMV (linearly constrained minimum variance) beamformer [5] using the osl-ephys toolbox [6]. We parcellated the data to 52 regions of interest using the Kohl atlas [7].
Hidden Markov Modelling (HMM): We adopting the time-delay embedded HMM approach to identify transient networks of oscillatory activity [8]. The HMM segments the time series into categorical states of unique spatio-spectral activity. We identified 10 transients network states of oscillatory activity. This work used the osl-dynamics toolbox [8].
Spectral calculations: We calculated cross power spectra using the parcel data and inferred state time courses with a multitaper (2 s window, 0% overlap, 7 DPSS tapers, 4 Hz time half bandwidth).
Results:
Figure 1A shows 10 transient networks of unique spatio-spectral patterns exhibited by healthy individuals. These networks have fast dynamics with lifetimes on the level of 100 ms. These networks represent fundamental spatio-temporal structure in brain activity.
Figure 1B shows how age affects the dynamics of the transient networks. Most networks increase in occurrence with age (more activations with longer durations). However, the frontal networks (States 4 and 9) decrease in occurrence. A hypothesis for this observation is that these networks become more efficient with age meaning they do not need to activate as often to perform their function.
Figure 1C shows the correlates of good cognitive performance. We observe the direction of the cognitive performance effect in the frontal networks (States 4 and 9) is in the same direction as the age effect. This is hallmark of the compensatory hypothesis for preserving cognitive health with age, see Figure 1E.
Figure 1B shows how age affects the dynamics of the transient networks. Most networks increase in occurrence with age (more activations with longer durations). However, the frontal networks (States 4 and 9) decrease in occurrence. A hypothesis for this observation is that these networks become more efficient with age meaning they do not need to activate as often to perform their function.
Figure 1C shows the correlates of good cognitive performance. We observe the direction of the cognitive performance effect in the frontal networks (States 4 and 9) is in the same direction as the age effect. This is hallmark of the compensatory hypothesis for preserving cognitive health with age, see Figure 1E.
Conclusions:
Healthy individuals exhibit fast transient networks of oscillatory activity. This work has inferred these networks on a particularly large dataset (largest to date). Consequently, they are highly reproducible in new datasets [8]. This means they are well placed to serve as a canonical basis set electrophysiological data (both MEG and EEG). Furthermore, by characterising how these networks change as a function of age for healthy individuals, these networks can serve as a foundational normative model. Such a resource would be particularly valuable in the study of age-associated disease, such as Alzheimer's or Parkinson's. To this end, we have made the networks and code to apply them publicly available: https://github.com/OHBA-analysis/Gohil2024_AgeEffectsRSNs.
Disorders of the Nervous System:
Neurodegenerative/ Late Life (eg. Parkinson’s, Alzheimer’s)
Lifespan Development:
Aging 1
Modeling and Analysis Methods:
Connectivity (eg. functional, effective, structural)
EEG/MEG Modeling and Analysis 2
Keywords:
Aging
Cognition
Data analysis
Electroencephaolography (EEG)
ELECTROPHYSIOLOGY
Machine Learning
MEG
Modeling
1|2Indicates the priority used for review
Abstract Information
By submitting your proposal, you grant permission for the Organization for Human Brain Mapping (OHBM) to distribute your work in any format, including video, audio print and electronic text through OHBM OnDemand, social media channels, the OHBM website, or other electronic publications and media.
The Open Science Special Interest Group (OSSIG) is introducing a reproducibility challenge for OHBM 2025. This new initiative aims to enhance the reproducibility of scientific results and foster collaborations between labs. Teams will consist of a “source” party and a “reproducing” party, and will be evaluated on the success of their replication, the openness of the source work, and additional deliverables. Click here for more information. Propose your OHBM abstract(s) as source work for future OHBM meetings by selecting one of the following options:
Please indicate below if your study was a "resting state" or "task-activation” study.
Healthy subjects only or patients (note that patient studies may also involve healthy subjects):
Was this research conducted in the United States?
Were any human subjects research approved by the relevant Institutional Review Board or ethics panel? NOTE: Any human subjects studies without IRB approval will be automatically rejected.
Were any animal research approved by the relevant IACUC or other animal research panel? NOTE: Any animal studies without IACUC approval will be automatically rejected.
Please indicate which methods were used in your research:
Provide references using APA citation style.
[2] Cabeza, R., Albert, M., Belleville, S., Craik, F. I., Duarte, A., Grady, C. L., ... & Rajah, M.
N. (2018). Maintenance, reserve and compensation: the cognitive neuroscience of healthy ageing. Nature Reviews Neuroscience, 19(11), 701-710.
[3] Shafto, M. A., Tyler, L. K., Dixon, M., Taylor, J. R., Rowe, J. B., Cusack, R., ... &
Cam-CAN. (2014). The Cambridge Centre for Ageing and Neuroscience (Cam-CAN) study protocol: a cross-sectional, lifespan, multidisciplinary examination of healthy cognitive ageing. BMC neurology, 14, 1-25.
[4] Taylor, J. R., Williams, N., Cusack, R., Auer, T., Shafto, M. A., Dixon, M., ... & Henson, R. N. (2017). The Cambridge Centre for Ageing and Neuroscience (Cam-CAN) data repository: Structural and functional MRI, MEG, and cognitive data from a cross-sectional adult lifespan sample. neuroimage, 144, 262-269.
[5] Van Veen, B. D., & Buckley, K. M. (1988). Beamforming: A versatile approach to spatial filtering. IEEE assp magazine, 5(2), 4-24.
[6] Quinn, A. J., van Es, M. W. J., Gohil, C., & Woolrich, M. W. (2022). Ohba software library in python (osl). Zenodo https://doi. org/10.5281/ZENODO, 6875060.
[7] Kohl, O., Woolrich, M., Nobre, A. C., & Quinn, A. (2023). Glasser52: A parcellation for
MEG-Analysis [Data set]. Zenodo. https://doi.org/10.5281/zenodo.10401793.
[8] Gohil, C., Huang, R., Roberts, E., van Es, M. W., Quinn, A. J., Vidaurre, D., & Woolrich,
M. W. (2024). osl-dynamics, a toolbox for modeling fast dynamic brain activity. Elife, 12,
RP91949.