Morphometric Measurement of Mean Cortical Curvature: Analysis of Alterations in Cognitive Impairment

Abstract

Background and Objectives: Cognitive impairment, including mild cognitive impairment (MCI) and Alzheimer's disease (AD), is a growing public health concern. Early detection and an understanding of structural changes are crucial for accurate diagnosis and timely intervention. Cortical curvature, a morphometric measure derived from structural magnetic resonance imaging (MRI), has emerged as a potential biomarker for neurodegenerative processes. This study investigates the relationship between mean cortical curvature and cognitive impairment. Materials and Methods: A cross-sectional study was conducted with 58 participants, categorized into, first, cognitively impaired (CI) and non-cognitively impaired (NC) groups and, second, a normal cognitive group (NC), a mild cognitive performance group (MPG), and a low cognitive performance group (LPG) based on the Montreal Cognitive Assessment (MoCA) score. MRI data were acquired using a 3.0 Tesla scanner, and cortical reconstruction was performed using FreeSurfer 7.2.0. Mean cortical curvature values were extracted for 34 brain regions per hemisphere. Results: Significant differences in mean cortical curvature were found between the CI and NC groups. In the right hemisphere, statistically significant changes in mean curvature were observed in the isthmus cingulate (U = 188.5, p = 0.006), lingual (U = 202.5, p = 0.013), pars orbitalis (U = 221.5, p = 0.031), and posterior cingulate regions (U = 224.5, p = 0.035). In the left hemisphere, significant differences were detected in the cuneus (U = 226.5, p = 0.038) and posterior cingulate (U = 231.5, p = 0.046) regions. Analysis across three cognitive performance groups (NC, MPG, and LPG) showed significant curvature differences in the right isthmus cingulate (H(2) = 7.492, p = 0.024) and lingual regions (H(2) = 6.250, p = 0.044). Conclusions: Decreased mean cortical curvature in brain regions associated with cognitive function could be indicative of cognitive impairment and may reflect early neurodegenerative changes. These results highlight cortical curvature as a potential structural sign for cognitive impairment, showing the need for further investigation in longitudinal studies.

Keywords: MoCA score; atlas-based segmentation; cognitive impairment; dementia; mean cortical curvature; neuroimaging; structural MRI.

Figure 1

(a) Curvature equation and (b) radius of an osculating circle at a point. (c) Mean curvature equation and (d) principal radius and their orthogonal location in relation to each other [11].

Figure 2

Mean curvature differences in the isthmus cingulate region in the right hemisphere in all three groups: NC, MPG, and LPG.

Figure 3

Anatomical location of the isthmus cingulate region. The highlighted dark purple region represents the area where statistically significant differences in the mean cortical curvature were observed between groups.

Figure 4

Mean curvature differences in the lingual region in the right hemisphere in all three groups: NC, MPG, and LPG.

Figure 5

Anatomical location of the lingual region in the right hemisphere, highlighted in salmon pink. This region showed significant differences in mean cortical curvature across the cognitive performance groups (NC, MPG, and LPG).

Figure 6

Mean curvature differences in pars orbitalis region in the right hemisphere in all three groups: NC, MPG, and LPG.

Figure 7

Anatomical location of the pars orbitalis region, shown in green. This area, part of the inferior frontal gyrus, exhibited significant differences in mean cortical curvature between groups.