Threshold-free estimation of entropy from a Pearson matrix

Felippe, H.; Viol, A.; de Araujo, D. B.; da Luz, M. G. E.; Palhano-Fontes, F.; Onias, H.; Raposo, E. P.; Viswanathan, G. M.

doi:10.48550/arxiv.2106.05379

Cited by 2 publications

(1 citation statement)

References 55 publications

(134 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…where N is the number of ROIs in the correlation matrix (N=33 or 7) and 𝜆 𝑗 are the eigen values of the 𝜌 matrix (Felippe et al, 2021). The entropy takes minimum value when all the correlation coefficients are unity and reaches the maximum value (log N) if the all the correlations are zero.…”

Section: Whole Brain Variability Properties Of Structural Covariance ...mentioning

confidence: 99%

Brain structural covariances in the ageing brain in the UK Biobank

Dong¹,

Thalamuthu²,

Jiang³

et al. 2022

Preprint

View full text Add to dashboard Cite

Brain structural covariances or pairwise correlations describe how morphologic properties of brain regions are related to one another across individuals. Although it is reported that brain structural covariance changes during brain development, it is not clear how structural covariance relates to the ageing process. Here we investigated the human brain structural covariances of cortical thickness and subcortical volumes in the ageing brain and their associations with age, cognition, and longevity polygenic risk score (longevity-PRS) by using cross-sectional data from the UK Biobank (N = 42075, aged 45-83 years, M/F=19752 /22323). The sample of participants was divided into 84 non-overlapping groups based on their age. The older the age group, the greater the variability in the whole brain structural covariance. However, there was a differential rate of age-related increase of variance between males and females. The variance of females started lower than those of males and then increased with age with a greater gradient than that of males. There was a consistent and significant enrichment of pairwise correlations within the occipital lobe in ageing process. The cortical thickness and subcortical covariances in older groups were significantly different from those in the youngest group. Sixty-two of the total 528 pairs of cortical thickness correlations and 10 of the total 21 pairs of subcortical volume correlations were significantly associated with age after Bonferroni correction. Specifically, with an increasing age, most decreased cortical thickness correlations were found between the regions within the frontal lobe as well as between the frontal lobe regions and regions in other lobes, while pairwise correlations within occipital lobe regions were all strengthening. Most of these correlations were also associated with global cognition and weakly associated with longevity-PRS. These findings revealed that the structural covariance was not stable during ageing. Given the thinning of the cortex and the volumetric reduction of subcortical structures seen in the ageing process, an increased pairwise correlation between the brain regions in the older brain suggested a strengthened coordinated decline between the brain regions involved. However, some of the brain regions demonstrated a differentiated rate of decline which was shown as the inversed or reduced pairwise correlations between these regions.

show abstract

Section: Whole Brain Variability Properties Of Structural Covariance ...mentioning

confidence: 99%

Brain structural covariances in the ageing brain in the UK Biobank

Dong¹,

Thalamuthu²,

Jiang³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

Brain structural covariances in the ageing brain in the UK Biobank

Dong,

Thalamuthu,

Jiang

et al. 2024

Brain Struct Funct

View full text Add to dashboard Cite

The morphologic properties of brain regions co-vary or correlate with each other. Here we investigated the structural covariances of cortical thickness and subcortical volumes in the ageing brain, along with their associations with age and cognition, using cross-sectional data from the UK Biobank (N = 42,075, aged 45–83 years, 53% female). As the structural covariance should be estimated in a group of participants, all participants were divided into 84 non-overlapping, equal-sized age groups ranging from the youngest to the oldest. We examined 84 cortical thickness covariances and subcortical covariances. Our findings include: (1) there were significant differences in the variability of structural covariance in the ageing process, including an increased variance, and a decreased entropy. (2) significant enrichment in pairwise correlations between brain regions within the occipital lobe was observed in all age groups; (3) structural covariance in older age, especially after the age of around 64, was significantly different from that in the youngest group (median age 48 years); (4) sixty-two of the total 528 pairs of cortical thickness correlations and 10 of the total 21 pairs of subcortical volume correlations showed significant associations with age. These trends varied, with some correlations strengthening, some weakening, and some reversing in direction with advancing age. Additionally, as ageing was associated with cognitive decline, most of the correlations with cognition displayed an opposite trend compared to age associated patterns of correlations.

show abstract

Threshold-free estimation of entropy from a Pearson matrix

Cited by 2 publications

References 55 publications

Brain structural covariances in the ageing brain in the UK Biobank

Brain structural covariances in the ageing brain in the UK Biobank

Brain structural covariances in the ageing brain in the UK Biobank

Contact Info

Product

Resources

About