Premium
P2‐448: Early treatment with pioglitazone in APP transgenic mice shows focal hippocampal volume increase related to improved cognitive performance
Author(s) -
Badhwar Amanpreet,
Lerch Jason P.,
Sled John G.,
Hamel Edith
Publication year - 2010
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1016/j.jalz.2010.08.022
Subject(s) - morris water navigation task , pioglitazone , hippocampal formation , hippocampus , water maze , medicine , striatum , effects of sleep deprivation on cognitive performance , neuroscience , cognition , psychology , endocrinology , dopamine , type 2 diabetes , diabetes mellitus
the course and treatment of dementia and other neurodegenerative diseases. In this study, we employed resting-state MRI connectivity methods and the large-scale network (LSN) analyses to discriminate Alzheimer’s disease (AD), mild cognitive impairment (MCI) and age-matched cognitively normal (CN) subjects. Methods: Resting-state functional magnetic resonance imaging was employed to acquire the voxelwised time series in the whole brains of 55 human subjects (20AD,15MCI, and20CNsubjects). Thebrains were parcellated into 116 regions of interest (ROIs). ADecreasedConnectivity Index (DCI) and an Increased Connectivity Index (ICI) were obtained from empirically selected pairwised ROIs. The DCI and ICI were employed to classify the subjects. Error estimation of the classifications was performed with the leave-one-out (LOO)method.Results:Tables 1-3 show the result of the LOOestimate in a two by two analysis. The first row shows the number of subjects within each group classified by clinical assessment as the “golden standard”, the second and the third rows show the percentage and the absolute number of the subjects (in parenthesis) that were classified with the LSN method. And the last row shows the overall accuracy of the classifier. Conclusions:We employed the large-scale network analysis to discriminate between control, AD and MCI subjects. MCIs can be discriminated from controls and ADs with an accuracy of 86% and 80%, respectively. The estimated false positive and negative cases may be largely related to the error of clinical assessment and the possible discrepancy between clinical assessment and network test conducted in this study may provide physicians a reference for better diagnosis. Clearly, further longitudinal study will be needed to validate this method. In summary, this non-invasive network approach not only can distinguish between control, MCI and AD subjects, but also provide different neural network patterns between different groups, whichwill lead us to deeper understanding the mechanisms of dementia. Table 1 Classify between CN and AD subjects.