Assessment of localized brain regions correlated with MMSE using VBM analysis of structural MRI in a Japanese sample

Abstract

The global aging population has led to a significant rise in dementia and cognitive decline, with Alzheimer's disease as the primary cause. Early detection of mild cognitive impairment (MCI), a prodromal stage of dementia, is critical for timely intervention. The Mini-Mental State Examination (MMSE) is commonly used for cognitive screening, yet its limitations-such as ceiling effects and educational biases-may hinder the early identification of subtle cognitive impairments. This cross-sectional study employed voxel-based morphometry (VBM) analysis of structural magnetic resonance imaging (MRI) to explore brain regions positively correlated with MMSE scores in a cohort of 510 participants. Significant gray matter volume (GMV) reductions were observed in the bilateral lateral frontal lobes, left medial frontal lobe, left hippocampus, left anterior cingulate cortex (ACC), and bilateral inferior temporal gyri in association with lower MMSE scores. Participants were classified into three groups-Normal Cognition (NC), MCI, and Dementia (D)-based on MMSE cutoff values. Compared to the NC group, the MCI group exhibited significant GMV reductions in the left hippocampus, left parahippocampal gyrus, left ACC, and right ventromedial prefrontal cortex (vmPFC). The D group showed further GMV reductions in the bilateral hippocampus and left inferior temporal gyrus compared to the MCI group. These findings highlight the clinical utility of VBM-based structural MRI in assessing localized brain atrophy associated with cognitive decline, supporting its potential role in early diagnosis and intervention for MCI. Further research integrating longitudinal studies and multimodal biomarkers is warranted to enhance diagnostic accuracy.

Keywords: Dementia; Gray matter volume; Mild cognitive impairment; Mini-mental state examination; Normal cognition; Structural magnetic resonance image; Voxel-based morphometry.

Fig. 1

Overview of the VBM preprocessing procedure using BAAD in this study. Caption: VBM: Voxel-based morphometry, BAAD: Brain Anatomical Analysis using DARTEL, 3D-T1WI: 3D-T1-weighted images, FLAIR: fluid-attenuated inversion recovery, AC-PC: anterior commissure-posterior commissure, INU: intensity non-uniformity, CAT: computational anatomy toolbox, SPM: statistical parametric mapping, MNI: Montreal Neurological Institute. More information about VBM preprocessing using BAAD can be found elsewhere (Shiino et al., 2021; Syaifullah et al., 2021).

Fig. 2

Localized brain regions indicating a positive correlation with MMSE total score. Caption: FWE-corrected p < 0.05 (k = 40), with age, gender, educational level, and total intracranial volume as covariates. L = left, R = right.

Fig. 3

Changes in gray matter volume from Normal Cognition group to Dementia group. Caption: Red areas in the figure indicate regions where gray matter volume is reduced in the MCI group (MMSE 27–24) compared to the NC group (MMSE 30–28). Green areas show regions where gray matter volume is reduced in the D group (MMSE ≤23) compared to the MCI group (MMSE 27–24). The numbers in the curly brackets in the figure represent Montreal Neurological Institute (MNI) coordinates. NC: Normal cognition, MCI: Mild cognitive impairment, D: Dementia. L = left, R = right, ACC: anterior cingulate cortex, vmPFC: ventromedial prefrontal cortex, HIPP: hippocampus, pHIPg: parahippocampal gyrus, ITG: inferior temporal gyrus.