Meeting Banner
Abstract #2828

The X-Guided Clustering Method and Its Application to Unbiased Detection of Differences in Functional Connectivity Networks

Gang Chen1, B. Douglas Ward2, Chunming Xie2, Wenjun Li2, Guangyu Chen2, Joseph S. Goveas3, Piero G. Antuono4, Shi-Jiang Li2

1Biophysics, Medical College of Wisconsin, Milwaukee, WI , United States; 2Biophysics, Medical College of Wisconsin, Milwaukee, WI, United States; 3Psychiatry and Behavioral Medicine, Medical College of Wisconsin, Milwaukee, WI, United States; 4Neurology, Medical College of Wisconsin, Milwaukee, WI, United States

In resting-state functional MRI cross-group studies, the available techniques may bias the results in choosing the specified reference. We introduced an x-guided clustering method to solve this problem. The x-guided clustering method incorporates all of the subject groups by producing group difference information and then creating functional clusters. This method was applied to identify functional clusters that were significantly different between the amnestic mild cognitively impaired and cognitively normal subjects.

Keywords

account advance altered application applied approval arranged assistance authors available behavioral belonging bias biased bilateral biophysics blue board brain built call choosing classification cluster clustering clusters coefficient cognitive cognitively college color conducted connectivity consent correction correlated correlation courses cross decreased define delayed description detailed detection disconnected disease distance distribution editorial efficacy employed false field findings frontal function functional generate grant grants guided health hierarchical highlighted homogeneity identified identify impaired implemented included index individual inferior input institutes institutional introduced linkage masks matrices matrix medical memory middle mild moment monitoring national network networks neurology nine pair pairs paragraph participant particular performance positive posterior potential prediction preprocessed produce product psychiatry random recall reduced relationship resembled resting review sample scale scanner score scores selected selection sent seventeen significantly similarity specifically specified springer status studies subject subjects superior support supported task technical temporal thalamus thank therapies threshold thresholded transformation unbiased variants vector vectors ward written yielded