Premium
P3‐360: IMPAIRED BRAIN SPONTANEOUS ACTIVITY OF ALZHEIMER DISEASE REVEALED BY MULTICENTER RESTING FMRI (N=688)
Author(s) -
Li Jiachen,
Jin Dan,
Liu Bing,
Wang Dawei,
Wang Pan,
Wang Qing,
Yu Chunshui,
Zhang Xi,
Zhang Xinqing,
Zhou Yuying,
Liu Yong,
Han Ying
Publication year - 2018
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.2018.06.1722
Subject(s) - resting state fmri , audiology , voxel , dementia , brain activity and meditation , psychology , neuroscience , medicine , disease , electroencephalography , radiology
Background: The brain is a complex and interconnected system; multiple neurodegenerative changes can cumulate to contribute to cognitive decline. The MRI based Brain Atrophy and Lesion Index (BALI) has been created to evaluate structural changes of the whole brain by summarizing deficits in several categories. So far, BALI has been applied to thousands of subjects from multiple independent datasets from around the world. In the present study, we examined the consistency of the BALI assessment in differentiating subjects with Alzheimer’s disease, mild cognitive impairment, and normal cognition, and the associations of BALI with age and cognition. Methods:Data were obtained from studies published between 2010 and 2017, which involved large-scale open-access multi-centre datasets (n1⁄42790), as well as several local datasets (n1⁄4310). Evaluation of the MR images followed the standard BALI assessment schema and carried out by multiple raters trained on the method. Results from the studies were compared and then pooled. Differences in the mean BALI scores across diagnoses were investigated using the Wallis-Kruskal Chi2 test and Cohen effect size. Results: Inter-rater agreement rate was consistently high, ranging from 0.81 to 0.91. In all datasets, there was a difference in the BALI total score and sub-categorical scores between diagnostic groups (e.g. total BALI, Chi>24.0, p<0.001) with variations between studies. Analyses combining the samples suggested a medium to large effect size depending on diagnosis. The associations between the total BALI score and age (p<0.0001) and various measures of cognition (p<0.01) were also consistent. Conclusions:The BALI was applied to a series of datasets for assessment of whole-brain structural changes in ageing-dementia, showing as both robust and generalizable. The method allows us to investigate the impact of aging on brain structural health, and assess how this affects cognitive decline and dementia. Variations in BALI assessment between different studies can be explained by differences in the study samples.