Study on the topological features of cortical morphological similarity networks in Alzheimer’s disease of different cognitive states
Year: 2026
Type: article
Abstract: Background: Individual morphological similarity networks (MSNs) were constructed based on cortical morphological metrics extracted from structural magnetic resonance imaging (sMRI) data to investigate topological abnormalities in brain structural networks in Alzheimer's disease (AD) and their correlations with cognitive levels. Methods: Based on T1-weighted magnetic resonance imaging (MRI) images from 83 AD patients and 40 healthy controls (HC), we extracted gray matter volume (GMV), cortical thickness (CT), and surface area (SA). Individual MSNs were constructed by calculating the Pearson correlation coefficients between the three-dimensional morphometric feature vectors of pairs of cortical regions. Graph theoretical analysis was then employed to compute nodal topological metrics, including betweenness centrality (Bc), degree centrality (Dc), nodal efficiency (Ne), and nodal local efficiency (Nle), to identify abnormal alterations in the brain structural networks of AD patients. Furthermore, we explored the correlations between mini-mental state examination (MMSE) scores and these altered nodal topological metrics. Based on their MMSE scores, AD patients were further categorized into high cognitive (high-MMSE) and low cognitive (low-MMSE) subgroups. Finally, topological metrics showing significant differences between the AD and HC groups were compared across the high-MMSE, low-MMSE and HC groups. Results: Significant MSN topological alterations in AD were predominantly localized in the frontal regions. Dc was generally reduced in the frontal regions of AD patients, including the bilateral middle frontal gyrus (P=0.027; P=0.007) and bilateral superior frontal gyrus (P=0.007; P=0.020). MMSE scores correlated positively with Dc in the right transverse frontopolar gyri and sulci (r=0.335, P=0.03) but negatively with Dc (r=-0.352, P=0.03) and Ne (r=-0.324, P=0.03) in the right superior frontal sulcus. The low-MMSE group exhibited lower Dc in the left middle frontal gyrus (P=0.0167) and left superior frontal gyrus (P=0.0008) than HCs, with a trend toward lower values compared with the high-MMSE group. Conversely, the high-MMSE group showed significantly lower Dc (P=0.002) and Ne (P=0.0142) in the right superior frontal sulcus than the low-MMSE group. Conclusions: This study demonstrates that AD patients exhibit structural network alterations predominantly involving frontal regions, with characteristic patterns corresponding to cognitive severity. These MSN topological alterations may provide insights into the network mechanisms underlying cognitive heterogeneity in AD and highlight the potential of network-based measures for characterizing disease-related brain alterations.
Authors: Xinran Zhu, Jiahao He, Jiwei You, Jiangnan Lin
Language: English
Subfield: Cognitive Neuroscience
Field: Neuroscience
Domain: Life Sciences
Open Access status: diamond
OpenAlex ID: W7222546315