Brain function changes during our lifespan. So far, only pairwise functional interactions between brain regions have been employed to predict brain age. High-order interactions (HOI) can capture complex non-linear associations between pairwise regions and the rest of the brain, detecting key synergistic and redundant couplings. Our recent study published in Nature Medicine significantly advances our understanding of brain aging across diverse populations. We employed cutting-edge deep learning techniques to analyze HOI from functional MRI (fMRI) and electroencephalography (EEG), encompassing over 5,000 participants from 15 countries.

At the heart of our approach lies graph convolutional networks (GCNs) designed to process graph representations from HOI. This method allows us to capture complex relationships in brain activity, unlike traditional convolutional neural networks, which are more suited for structural imaging. We further innovated by introducing an advanced metric called organizational information (Ω) to quantify HOI between brain regions.

Organizational Information: A Detailed Assessment of Brain Connectivity

The Ω metric provides a more comprehensive view of brain connectivity by assessing the balance between redundancy and synergy in HOI among brain regions. This approach goes beyond traditional pairwise connectivity measures, allowing us to capture the complex interdependencies between two brain regions and the rest of the brain. Positive Ω values indicate redundancy dominance, suggesting shared randomness in brain activity, while negative values point to synergy dominance, implying collective constraints in neural interactions. These HOI patterns are used as input features for the deep learning models, and a brain age prediction is obtained. Our analysis reveals that frontoposterior networks are crucial in brain aging. Both fMRI and EEG models highlighted large-scale frontoposterior HOI as critical predictors of brain age. These findings suggest that the integrity of frontoposterior networks may be a key indicator of brain health and aging.

The HOI to track the Progression of Cognitive Decline

Our brain-age gap models reveal a progressive acceleration of brain aging from healthy individuals to those with neurocognitive disorders. We observe increasing brain-age gaps from healthy controls to individuals with mild cognitive impairment (MCI), and further increases in those with Alzheimer's disease (AD) and frontotemporal dementia (FTD).

We find that individual clinical conditions, country-level socioeconomic factors, gender disparities, and environmental exposomes influence brain-age gaps. Individuals in countries with more significant socioeconomic inequalities, higher pollution levels, and limited healthcare access tend to exhibit older brain ages.

Towards a more Holistic Understanding of Brain Aging

Our research demonstrates the power of combining advanced technical methods, such as the Ω metric and GCNs, with a detailed understanding of social and cultural factors. By focusing on frontoposterior networks and adopting this multimodal diversity approach, we can develop more accurate models of brain aging and create targeted interventions to promote brain health across all populations.

This study advances our understanding of brain aging. It highlights the urgent need for global health policies that address the social and environmental factors contributing to accelerated brain aging, particularly in regions facing greater socioeconomic challenges. Future research should continue to explore how these factors specifically impact frontoposterior network integrity and overall brain health.

For more information: 

Moguilner S, Baez S, Hernandez H, et al. Brain clocks capture diversity and disparities in aging and dementia across geographically diverse populations. Nat Med. Published online August 26, 2024. doi:10.1038/s41591-024-03209-x

Hernandez H, Baez S, Medel V, et al. Brain health in diverse settings: How age, demographics and cognition shape brain function. Neuroimage. 2024;295:120636. doi:10.1016/j.neuroimage.2024.120636

Ibanez A, Slachevsky A. Environmental-genetic interactions in ageing and dementia across Latin America. Nat Rev Neurol. 2024;20(10):571-572. doi:10.1038/s41582-024-00998-0

Evans TE, Vilor-Tejedor N, Operto G, et al. Structural brain differences in the Alzheimer's disease continuum: Insights into the heterogeneity from a large multi-site neuroimaging consortium. Biol Psychiatry Cogn Neurosci Neuroimaging. Published online July 29, 2024. doi:10.1016/j.bpsc.2024.07.019

The poster image was created with Meta AI under supervision.