[Paper Review] Human Sexual Dimorphism of the Relative Cerebral Area Volumes in the Data of the Human Connectome Project

Balázs Szalkai, Vince Grolmusz|arXiv (Cornell University)|Apr 20, 2016

Advanced Neuroimaging Techniques and Applications9 references3 citations

TL;DR

This study analyzes T1-weighted and diffusion MRI data from 511 subjects in the Human Connectome Project to investigate sexual dimorphism in relative cerebral area volumes. Using FreeSurfer-based segmentation and Holm-Bonferroni corrected correlation analysis, it finds that relative gray matter and subcortical volumes are significantly larger in women than men across most brain regions, with notable differences in correlation strength between hemispheres.

ABSTRACT

The average human brain volume of the males is larger than that of the females. Several MRI voxel-based morphometry studies show that the gray matter/white matter ratio is larger in females. Here we have analyzed the recent public release of the Human Connectome Project, and by using the diffusion MRI data of 511 subjects (209 men and 302 women), we have found that the relative volumes of numerous subcortical areas and the gray matter of most cortical areas are significantly larger in women than in men. Additionally, we have discovered differences of the strengths of the sexual correlations between the same structures in different hemispheres.

Motivation & Objective

• To investigate sexual dimorphism in relative cerebral area volumes using high-resolution MRI data from the Human Connectome Project.
• To determine whether relative gray matter and subcortical volumes differ significantly between males and females, independent of total brain volume.
• To examine whether the strength of sexual dimorphism varies between left and right hemispheres for the same brain structures.
• To provide a publicly available dataset of subject-level relative volume estimates for future neuroanatomical research.
• To avoid confounding variables such as age, education, and handedness by not applying statistical control for them in the analysis.

Proposed method

• Acquired T1-weighted and diffusion MRI data from 511 subjects (209 males, 302 females) in the Human Connectome Project's S500 release.
• Calculated cortical gray matter volumes by multiplying FreeSurfer-derived thickness and area values for each region of interest (ROI).
• Computed relative volumes by dividing each ROI's volume by the FreeSurfer Total Brain Mask Volume (FS_Mask_Vol), serving as the whole-brain reference.
• Performed Pearson correlation between sex (coded as 1=male, 2=female) and relative volume for each of 115 ROIs.
• Applied two-tailed t-tests to compute p-values and used the Holm-Bonferroni method to correct for multiple comparisons across all ROIs.
• Provided anonymized subject-level relative volume data and statistical results via a public data repository (http://uratim.com/relvol/tables.zip).

Experimental results

Research questions

• RQ1 Are relative gray matter and subcortical volumes significantly different between males and females when normalized to total brain volume?
• RQ2 Which specific brain regions exhibit the most pronounced sexual dimorphism in relative volume?
• RQ3 Do the strengths of sexual dimorphism differ between the left and right hemispheres for homologous brain structures?
• RQ4 Are there any brain regions where the relative volume is larger in males despite the overall trend favoring larger volumes in females?
• RQ5 How robust are the observed differences after correcting for multiple comparisons across 115 brain regions?

Key findings

• The relative volume of the total gray matter was significantly larger in females (mean = 0.4218) than in males (mean = 0.4131), with a correlation coefficient of 0.2749 (p = 0).
• The subcortical gray matter relative volume was significantly larger in females (mean = 0.03797) than in males (mean = 0.0364), with a correlation of 0.3564 (p = 0).
• The left superior parietal lobe had a significantly higher relative volume in females (mean = 0.00781) than in males (mean = 0.00734), with a correlation of 0.311 (p = 0).
• The right temporal pole had a significantly larger relative volume in females (mean = 0.00101) than in males (mean = 0.00094), with a correlation of 0.2787 (p = 0).
• Several regions, including the left and right caudate, hippocampus, and thalamus, showed significantly larger relative volumes in females, with correlations ranging from 0.2376 to 0.2704.
• The study identified hemisphere-specific differences in correlation strength, such as stronger sexual dimorphism in the left hemisphere for the posterior cingulate and lateral orbitofrontal regions.