Use of magnetic resonance structural imaging to identify disease progression in patients with mild cognitive impairment: A voxel-based morphometry and surface-based morphometry study.

Abstract

Voxel-based morphometry (VBM) and surface-based morphometry (SBM) based on magnetic resonance structural imaging were used to identify disease progression in mild cognitive impairment (MCI) patients. A retrospective analysis was conducted on 154 MCI patients from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database, with 62 patients classified into the progressive MCI (pMCI) group and 92 patients into the stable MCI (sMCI) group. VBM and SBM were employed to identify structural differences between sMCI and pMCI patients, and differential features were extracted for model construction. The logistic regression method was used to establish relevant index models, and the DeLong test was used to compare the diagnostic performance of the different models. Additionally, 51 patients from the National Alzheimer's Coordinating Center (NACC) database were used as an external validation set to further validate the clinical efficacy of the model. Significant structural differences between pMCI and sMCI patients were revealed through VBM and SBM analyses. Volume reductions were observed in the frontal and temporal lobes, and cortical thinning occurred in the left inferior and superior parietal cortices. Reduced gyrification was observed in the bilateral insular gyrus. The structural joint model, which combines volume and cortical indices, demonstrated higher diagnostic accuracy compared to the joint scale index model that combines the Mini-Mental State Examination (MMSE) and Montreal Cognitive Assessment (MOCA) indices. The findings indicate that combined VBM and SBM analysis offers a sensitive and noninvasive approach to detect structural biomarkers of MCI progression, providing a practical tool for early risk stratification and personalized clinical management.