The brain undergoes dynamic functional changes throughout life, and determining how these changes evolve and their relationship to chronological age is crucial to understanding the aging process, multi-level disparities, and brain disorders such as Alzheimer’s disease spectrum. Brain clocks or brain age models have emerged as transdiagnostic metric tools that assess brain health as affected by a variety of factors, suggesting that they may capture multimodal diversity in brain function.
Populations in Latin America and the Caribbean (LAC) show greater genetic diversity and different physical, social, and internal exposomes that influence brain phenotypes. Furthermore, inequality in income and socioeconomic status, high levels of air pollution, limited access to timely and effective health care, increasing prevalence of communicable and non-communicable diseases, and low educational attainment are key determinants of brain health in these regions.
Therefore, while measuring the brain age gap could improve our understanding of disease risk and its impact on accelerated aging, There is a lack of research on brain age models in these underrepresented populations. who face large socioeconomic and health disparities. In this line, it has been published in Nature Medicine a study that analyzes this problem.
Factors influencing brain changes
Sex and gender differences emerge as key factors influencing brain changes. Studies of atrophy in the Alzheimer’s disease spectrum show that brain atrophy progresses more rapidly in women than in men. Furthermore, gender inequality at the national level is linked to differences in cortical thickness between the sexes. The Structural gender inequality further aggravates brain healthsince adverse environments can affect dendritic branching and synapse formation.
However, to date the spectrum of brain age abnormalities, including the effects of demographic heterogeneity across geographic regions, between sexes, and across the continuum from brain health to disease, has not been investigated. Furthermore, most studies have been conducted with participants from the Northern Hemisphere, limiting the generalizability of results to the underrepresented populations in the southincluding Africa and Latin America.
On the other hand, the multimodal machine learning studies Machine learning methods show great potential for the analysis of brain aging; however, most focus on structural magnetic resonance imaging (MRI), which misses brain network dynamics. Complex spatiotemporal dimensions can be tracked with spatial precision using functional magnetic resonance imaging (fMRI) and with millisecond precision using electroencephalography (EEG). Furthermore, standard machine learning methods have lower generalization ability compared to deep learning methods.
Los Brain age indices have been limited The predominant use of MRI or positron emission tomography, which are less accessible and affordable in Africa and Latin America, leading to selection bias, offers a solution due to its cost-effectiveness, portability and ease of implementation in the study of aging and dementia. However, few studies have combined accessible techniques with deep learning to develop scalable markers of brain age.
Delving deeper into brain analysis
Against this backdrop, the researchers used fMRI and resting EEG signals separately to assess whether a deep learning computational sequence can capture differences in brain aging in heterogeneous populationsbased on a total of 5,306 datasets. We included fMRI data from 2,953 participants from Argentina, Chile, Colombia, Mexico, and Peru, as well as the USA, China, and Japan. The EEG dataset included 2,353 participants from Argentina, Brazil, Chile, Colombia, and Cuba, as well as Greece, Ireland, Italy, Turkey, and the UK. We focused on Alzheimer’s disease and frontal-temporal dementia (bvFTD) because these conditions are the most common causes of late- and early-onset dementia, respectively.
Models trained and evaluated on non-LAC datasets showed greater agreement with chronological age, whereas models applied to the LAC datasets revealed greater discrepancies in brain age, suggesting accelerated aging. Furthermore, the researchers observed an increase in the brain age gap from controls to mild cognitive impairment (MCI) and Alzheimer’s disease and sex differences indicated a wider brain age gap in women in both the control and Alzheimer’s disease groups.
Most brain clock patterns were independently confirmed and replicated in fMRI and EEG. Macrosocial factors At the aggregate level, such as socioeconomic inequality, pollution, and the burden of communicable and non-communicable diseases, influenced the brain age gap, especially in LAC. These findings provide a framework that captures the multimodal diversity associated with accelerated aging in different global contexts.
Accelerated brain aging
The results of the study suggest that belonging to a region of Latin America and the Caribbean are associated with accelerated brain agingDiversity-related factors, including different exposures and disease outcomes, may influence brain age gaps both in LAC and outside of these regions. Furthermore, they point out that neurocognitive disorders played a crucial role in these gaps.
However, the structural socioeconomic inequalityas well as increasing levels of air pollution and the load of communicable and non-communicable diseasesare also important factors affecting the brain age gap. The fact that these effects are more pronounced in LAC suggests, according to the study, that underlying inequalities and adverse environmental and health conditions have a macrosocial and structural impact on the observed regional differences. Furthermore, immigration could influence brain age through social determinants of health and genetic diversity. In Latin America and the Caribbean, tricontinental admixtures result in remarkable ancestral diversity both within and between countries, affecting dementia prevalence and brain phenotypes. “Future studies should take into account these potential effects on brain age gaps,” the researchers highlight.
You may also like…
