Study on Identification of Brain Regions to Easily Assess for Mild Cognitive Impairment Using MR-phase Information [Presidential Award Proceedings]

Recently, we developed a novel method to quantitatively evaluate amyloid plaques in the brain using magnetic resonance imaging (MRI) phase information, called double gaussian analysis (DGA). DGA nonlinearly evaluates the phase distribution of iron because of aging (aging iron) and iron in the amyloid plaque (amyloid iron), which is always lower than that of aging ; therefore, it sometimes provides an erroneous solution of the phase distribution of amyloid iron. Contrastingly, DGA provides the phase distribution of aging iron as a stable solution. Here, we used the phase information of aging iron to define brain regions representing cognitive brain function in participants with mild cognitive impairment. MRI were performed using 3T-MRI and phase images were derived in the DICOM format. We segmented the brains of participants into brain regions using the AAL atlas and integrated left and right brain regions into 85 regions and carefully selected as regions of interest (ROI) based on the following selection criteria : the average number of voxels is ＞3000, the region is related to cognitive function, and exceeded iron loading is previously known (exclusion criterion). We selected 17 ROIs for evaluation in this study and set the threshold between the cognitive decline and maintained groups as 27/26 in mini-mental state examination and derived the quantitative value as the change rate in TE after applying DGA. We found that the inferior parietal gyrus, excluding the angular gyrus and supramarginal gyrus, could discriminate only between the cognitive maintained and decline groups (p＜0.05). Quantitative values of the fusiform gyrus, insula, inferior temporal gyrus, and precuneus subtracted from the inferior parietal gyrus showed significant differences between cognitive decline and maintained groups (p＜0.05).