Chong-Yaw Wee1, Pew-Thian Yap1, Kevin Denny2, Lihong Wang2, Dinggang Shen1
1Radiology, University of North Carolina, Chapel Hill, NC, United States; 2Brain Imaging & Analysis Center, Duke University Medical Center, Durham, NC, United States
We introduce an effective network-based multivariate classification algorithm, using multi-spectral connectivity networks derived from resting-state functional MRI, to accurately identify MCI patients from normal controls. Classification accuracy given by our approach is 86.5%, which is at least an 18.9% increment from methods using a single frequency band. The AUC value of our method is 0.863, indicating good diagnostic power. Significant improvements and promising results indicate that the proposed framework can potentially serve as a complementary approach to clinical diagnosis of alteration in brain functions associated with cognitive impairment, especially at early stages.