AUTHOR=Jiao Zhuqing , Ji Yixin , Zhang Jiahao , Shi Haifeng , Wang Chuang 

TITLE=Constructing Dynamic Functional Networks via Weighted Regularization and Tensor Low-Rank Approximation for Early Mild Cognitive Impairment Classification

JOURNAL=Frontiers in Cell and Developmental Biology

VOLUME=Volume 8 - 2020

YEAR=2021

URL=https://www.frontiersin.org/journals/cell-and-developmental-biology/articles/10.3389/fcell.2020.610569

DOI=10.3389/fcell.2020.610569

ISSN=2296-634X

ABSTRACT=Brain functional networks constructed via regularization has been used widely in eMCI classification. However, most methods rarely consider the inter-group information from the brain functional networks of different subjects, and they cannot take into consideration some discriminative topology properties at the same time. To overcome these limitations, we propose a novel method to construct dynamic functional networks (DFN) based on weighted regularization (WR) and tensor low-rank approximation (TLA), and apply it to identify early mild cognitive impairment (eMCI) subjects from normal subjects. First, we introduce the weighted regularization term into the DFN construction and obtain WR-based DFNs (WRDFN). Then, we combine the WRDFNs of all subjects into a third-order tensor for TLA processing, and obtain the DFN based on WR and TLA (WRTDFN) of each subject in the tensor. We calculate the weighted-graph local clustering coefficient of each region in each WRTDFN as the effective feature, and use the two-sample t-test for feature selection. Finally, we train the linear SVM classifier to classify the WRTDFNs of all subjects. Experimental results demonstrate that WRTLA can obtain DFN with the scale-free property, and the classification accuracy (ACC), the sensitivity (SEN), the specificity (SPE) and the area under curve (AUC) reach 87.0098%, 83.3824%, 90.6373% and 0.9421, respectively. We also achieve the best classification results compared with other comparable methods. This work can effectively improve the classification performance of DFN constructed by existing methods for eMCI, and has certain reference value for the early diagnosis of Alzheimer's disease (AD).