Differentiating between healthy control participants and those with mild cognitive impairment using volumetric MRI data
MetadataShow full item record
OBJECTIVE: To determine whether volumetric measures of the hippocampus or entorhinal cortex in combination with other cortical measures can differentiate between cognitively normal individuals and participants with amnestic mild cognitive impairment (MCI). METHODS: T1-weighted magnetic resonance imaging (MRI) data acquired from 46 cognitively normal participants and 50 participants with amnestic MCI as part of the Boston University Alzheimer's Disease Center research registry and the Alzheimer's Disease Neuroimaging Initiative were used in this cross-sectional study. Cortical and subcortical volumes, including hippocampal subfield volumes, were automatically generated from each participant’s structural MRI data using FreeSurfer v6.0. Nominal logistic regression models containing these variables were used to evaluate their ability to identify participants with MCI. RESULTS: A model containing 11 regions of interest (insula, superior parietal cortex, rostral middle frontal cortex, middle temporal cortex, pars opercularis, paracentral lobule, whole hippocampus, subiculum, superior temporal cortex, precentral cortex and caudal anterior cingulate cortex) fit the data best (R2 = 0.7710, whole model test chi square = 102.4794, p < 0.0001). CONCLUSIONS: Volumetric measures acquired from MRI were able to correctly identify most healthy control subjects and those with amnestic MCI using measures of selected medial temporal lobe structures in combination with those from other cortical areas yielding an overall classification of 95.83% for this dataset. These findings support the notion that while clinical features of amnestic MCI may reflect medial temporal atrophy, differences that can be used to distinguish between these two populations are present elsewhere in the brain. This finding further affirming that atrophy can be identified before clinical features are expressed. Additional studies are needed to assess how well other imaging modalities, such as resting state functional connectivity, diffusion imaging, and amyloid and tau position emission tomography (PET), perform in classifying participants who are cognitively normal versus those who are amnestic MCI.
RightsAttribution 4.0 International