Voxel-based and surface-based morphometry in the cortical thickness, and cortical and subcortical grey matter volume in patients with mild to moderate Alzheimer's disease

Kaidi Li^1,2Dingling Xie³Zhengyong Zhang⁴Chunyu Fu²Chunyang Li^5,6*

• ¹Department of Neurology, Affiliated Hospital of Inner Mongolia Medical University, Hohhot, Inner Mongolia Autonomous Region, China
• ²The Affiliated Hospital of Inner Mongolia Medical University, Hohhot, Inner Mongolia Autonomous Region, China
• ³Xianning Central Hospital, Xianning, Hubei, China
• ⁴Xinxiang Medical University, Xinxiang, Henan Province, China
• ⁵Mental Health Center of Inner Mongolia, Hohhot, Inner Mongolia Autonomous Region, China
• ⁶Inner Mongolia Brain Hospital, Huhhot, China

Aim: This study aimed to investigate alterations in whole-brain cortical thickness (CT) and both cortical and subcortical grey matter volume (GMV) in patients with Alzheimer's disease (AD) compared to healthy controls (HC) using voxel-based morphometry (VBM) and surfacebased morphometry (SBM). Furthermore, we sought to develop a combined predictive model based on these neuroimaging markers and assess its potential clinical utility for the early detection and diagnosis of AD.Methods: A total of 42 patients diagnosed with mild-to-moderate AD and 49 demographicallymatched HC were recruited for this study. VBM and SBM analyses were performed on threedimensional T1-weighted magnetization-prepared rapid gradient echo (3D T1-MPRAGE) imaging sequences to identify brain regions exhibiting statistically significant differences between the AD and HC groups. Brain regions showing significant group differences were selected as regions of interest (ROIs). Pearson correlation analysis was employed to assess the relationship between the extracted neuroimaging metrics (CT, cortical GMV, subcortical GMV) and cognitive performance. Predictive models were constructed using CT (from SBM), cortical GMV, and subcortical GMV (from VBM) metrics derived from the ROIs, both individually and in combination. Model performance in discriminating between AD patients and HCs was evaluated using receiver operating characteristic (ROC) curve analysis.Results: Compared to HCs, AD patients exhibited significant CT reductions primarily in the transverse temporal gyrus, superior temporal gyrus, supramarginal gyrus, insula, temporal pole, entorhinal cortex, and fusiform gyrus. Significant GMV reductions in AD patients were observed predominantly in the hippocampus, parahippocampal gyrus, posterior temporal lobe, inferior temporal gyrus, middle temporal gyrus, limbic lobe structures, fusiform gyrus, amygdala, and thalamus, as detected by VBM analysis. Extracted CT, cortical GMV, and subcortical GMV measures from the ROIs demonstrated significant positive correlations with both MMSE and MoCA scores.SBM and VBM in brain atrophy with AD 3 Conclusion：In AD patients, VBM and SBM showed overlapping cortical GMV and CT reductions. Volume/thickness reductions correlated with lower MMSE/MoCA scores, confirming functional relevance. ROC analysis revealed that combining CT and GMV improved cognitive impairment prediction over single measures. This integrated approach may enhance clinical AD diagnostics and early risk identification.