Abstract
Abnormalities in the brain connectivity in patients with neurodegenerative diseases, such as early mild cognitive impairment (EMCI), have been widely reported. Current research shows that the combination of multiple features of the threshold connectivity network can improve the classification accuracy of diseases. However, in the construction of the threshold connectivity network, the selection of the threshold is very important, and an unreasonable setting can seriously affect the final classification results. Recent neuroscience research suggests that the minimum spanning tree (MST) brain functional network is helpful, as it avoids the methodological biases while comparing networks. In this paper, by employing the multikernel method, we propose a framework to integrate the multiple properties of the MST brain functional network for improving the classification performance. Initially, the Kruskal algorithm was used to construct an unbiased MST brain functional network. Subsequently, the vector kernel and graph kernel were used to quantify the two different complementary properties of the network, such as the local connectivity property and the topological property. Finally, the multikernel support vector machine (SVM) was adopted to combine the two different kernels for EMCI classification. We tested the performance of our proposed method for Alzheimer's Disease Neuroimaging Initiative (ANDI) datasets. The results showed that our method achieved a significant performance improvement, with the classification accuracy of 85%. The abnormal brain regions included the right hippocampus, left parahippocampal gyrus, left posterior cingulate gyrus, middle temporal gyrus, and other regions that are known to be important in the EMCI. Our results suggested that, combining the multiple features of the MST brain functional connectivity offered a better classification performance in the EMCI.
Highlights
At present, the brain functional magnetic resonance imaging data is represented as a brain network composed of nodes and edges (Lópezsanz et al, 2017)
Betweenness of 90 nodes was calculated from the training sets and two sample t-test was performed to evaluate its discriminative power for identifying the early mild cognitive impairment (EMCI) from NC
The results indicate that the local property is more important than the topological property of the minimum spanning tree (MST) brain functional network for the classification
Summary
The brain functional magnetic resonance imaging (fMRI) data is represented as a brain network composed of nodes and edges (Lópezsanz et al, 2017). Through the analysis and study of the brain network, the brain functional network of the mild cognitive impairment (MCI) patients exhibit abnormal local properties and topological structures (Supekar et al, 2008; Sanz-Arigita et al, 2010; Petrella et al, 2011; Liu et al, 2012; Wang et al, 2017; Yan et al, 2018). Jie et al (2014b) constructed an undirected functional brain network of NC and MCI, and extracted the topological features to classify the two groups of subjects, where abnormal regions were found in the brain network including those in the hippocampus, amygdala, and the inferior temporal gyrus. The result showed that this method was able to accurately classify three groups (i.e., NC, MCI, and AD) with an accuracy of 88.4%, and it was found that the left posterior central gyrus, the right inferior temporal gyrus, the left lingual gyrus, the right middle frontal gyrus, and the right thalamus were significantly different from the normal elderly. Wee et al (2016) designed a disease identification framework based on the estimated temporal networks, and analyzed the group differences in the level network property. Yu et al (2016) studied the directed functional connectivity using the Granger causality analysis (GCA), and found that the posterior cingulate cortex (PCC) in the Default Mode Network (DMN) showed directional disorders in receiving and transmitting information
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
