Could ratio-based morphometric analysis of subcortical limbic structures assist in Alzheimer's disease diagnosis?

Objectives: Alzheimer’s disease is a neurodegenerative disorder that primarily affects subcortical limbic structures. This study aimed to assess volumetric differences in subcortical limbic structures and to compare the relative volumes of surrounding brain regions - such as the telencephalon, diencephalon, and brainstem subdivisions - between individuals with Alzheimer’s disease and healthy controls. Methods: This study involved 24 patients with Alzheimer’s disease and 16 healthy controls. Subcortical structures were segmented automatically using MRICloud on 3D T1-weighted magnetic resonance imaging scans. To minimize individual anatomical variability, volume ratios relative to neighboring brain regions were also calculated. Results: Significant volume reductions were found in the amygdala (left: P=0.004, right: P=0.005, total: P=0.004), hypothalamus (left: P=0.005, right: P>0.05, total: P=0.007), diencephalon (left: P=0.001, right: P=0.012, total: P>0.05), and mammillary bodies (left: P=0.002, right: P=0.003, total: P=0.003) in the Alzheimer’s disease group compared to healthy controls. Although most volume ratios - particularly those involving the amygdala and mammillary bodies - were higher in the Alzheimer’s disease group, they did not reach statistical significance (P>0.05). Conclusions: This study confirms prominent atrophy in subcortical limbic structures in Alzheimer’s disease. While diencephalon volume was reduced, its ratio to the amygdalae changed minimally, likely reflecting more severe atrophy of the amygdalae. Similarly, the mesencephalon-to-hypothalamus ratio showed no significant difference, suggesting parallel atrophy. These findings support the combined use of abs olute and ratio-based analyses and demonstrate the potential of MRICloud to identify Alzheimer’s disease-related neuroanatomical changes.