Meeting Banner
Abstract #2883

Comparing Differences in Alzheimer's Disease Classification Between Structural and Resting-State Functional MRI Biomarkers

Guangyu Chen1, Chunming Xie1, Guangyu Chen1, Wenjun Li1, B. Douglas Ward1, Jennifer L. Jones2, Malgorzata Franczak2, Piero G. Antuono2, Shi-Jiang Li1

1Department of Biophysics, Medical College of Wisconsin, Milwaukee, WI, United States; 2Department of Neurology, Medical College of Wisconsin, Milwaukee, WI, United States

Data from 65 Alzheimer's disease and 136 cognitively normal subjects were included in the study. We compared Alzheimer's disease classification techniques, using structural and resting-state functional MRI biomarkers. When the RfMRI classification result agreed with the structural MRI classification result, the subjects were classified with 90% accuracy. The incongruous classification results may predict disease conversion or treatment response.

Keywords

abnormal accuracies accuracy achieved affected agree agrees alone always among assigned assistance authors available bank best better biophysics brain caused certain chance characteristics chosen classification classified classifier classifiers cognitively cohort college color comparing concentration congruous congruously connectivity convert database defined determine determined deviation discussed disease divided editorial either employed evaluate evaluated evaluation every except explain explains five frequency functional future gang grant grants gray health hypotheses hypothesized ideally improve included incongruous incongruously initiative instead institutes intact known latter leave made maximizing medical modalities national negative neurology optimal parts portion power procedure program radiology randomly ranks reduce reduced repeated researchers respond rest resting separation sixty sorted specifically status step steps structural studies subject subjects suffers summarized support supported technical thank trade train training treatments true type variation ward