Healthy aging: an automatic analysis of global and regional morphological alterations of human brain

Abstract

Rationale and objectives: Morphologic changes of the human brain during healthy aging provide useful reference knowledge for age-related brain disorders. The aim of this study was to explore age-related global and regional morphological changes of healthy adult brains.

Materials and methods: T1-weighted magnetic resonance images covering the entire brain were acquired for 314 subjects. Image processing of registration, segmentation, and surface construction were performed to calculate the volumes of the cerebrum, cerebellum, brain stem, lateral ventricle, and subcortical nuclei, as well as the surface area, mean curvature index, cortical thickness of the cerebral cortex, and subjacent white matter volume using FreeSurfer software. Mean values of each morphologic index were calculated and plotted against age group for sectional analysis. Regression analysis was conducted using SPSS to investigate the age effects on global and regional volumes of human brain.

Results: Overall global and regional volume loss was observed for the entire brain during healthy aging. Moderate atrophy was observed in subcortical gray matter structures, including the thalamus (R(2) = 0.476, P < .001), nucleus accumbens (R(2) = 0.525, P < .001), pallidum (R(2) = 0.461, P < .001), and putamen (R(2) = 0.533, P < .001). The volume of hippocampus showed a slight increase by 40 years of age, followed by a relatively faster decline after the age of 50 years (R(2) = 0.486, P < .001). Surface area and mean curvature were less affected by aging relative to cortical thickness and subjacent white matter volume. Significant cortical thinning was mainly found in the parietal (R(2) = 0.553, P < .001) and insula regions (R(2) = 0.405, P < .001).

Conclusions: Morphologic alterations of human brain manifested regional heterogeneity in the scenario of general volume loss during healthy aging. The age effect on the hippocampus demonstrated a unique evolution. These findings provide informative reference knowledge that may help in identifying and differentiating pathologic aging and other neurologic disorders.