Abstract
Brain functional connectivity network (BFCN) analysis has been widely used in the diagnosis of mental disorders, such as schizophrenia. In BFCN methods, brain network construction is one of the core tasks due to its great influence on the diagnosis result. Most of the existing BFCN construction methods only consider the first-order relationship existing in each pair of brain regions and ignore the useful high-order information, including multi-region correlation in the whole brain. Some early schizophrenia patients have subtle changes in brain function networks, which cannot be detected in conventional BFCN construction methods. It is well-known that the high-order method is usually more sensitive to the subtle changes in signal than the low-order method. To exploit high-order information among brain regions, we define the triplet correlation among three brain regions, and derive the second-order brain network based on the connectivity difference and ordinal information in each triplet. For making full use of the complementary information in different brain networks, we proposed a hybrid approach to fuse the first- and second-order brain networks. The proposed method is applied to identify the biomarkers of schizophrenia. The experimental results on six schizophrenia datasets (totally including 439 patients and 426 controls) show that the proposed method outperforms the existing brain network methods in the diagnosis of schizophrenia.
Highlights
Resting-state functional magnetic resonance imaging studies indicate that there exists a disorder-related alteration in Brain functional connectivity network (BFCN) (Bluhm et al, 2007; Jafri et al, 2008; Fornito and Bullmore, 2010; Shafiei et al, 2018; Wang et al, 2018)
The brain networks were constructed using the proposed method and a series of comparison methods mentioned below. Since this experiment is designed for comparing the performance of different functional brain networks, the same feature selection and classification algorithm in our method are all performed on these networks to ensure the comparability of the functional brain networks
It can be seen that our BFCN construction method has the best diagnostic results on six schizophrenia datasets among all brain network construction methods
Summary
Resting-state functional magnetic resonance imaging (rs-fMRI) studies indicate that there exists a disorder-related alteration in BFCN (Bluhm et al, 2007; Jafri et al, 2008; Fornito and Bullmore, 2010; Shafiei et al, 2018; Wang et al, 2018). Most of the existing methods first construct the brain functional network by measuring the correlation of brain regions or voxels, and perform feature extraction on the. Brain Network With Second-Order Information produced large-scale brain networks for selecting the significant features. These connectivities or sub-networks having significant alterations in some indicators, e.g., topological metrics (Fei et al, 2014) and the alteration degree (Guo et al, 2014; Zhu et al, 2018), are chosen as the biomarker for the disease. The previous BFCN construction methods mainly focus on revealing the low-order information among brain regions or voxels. Zhou et al (2018) embedded the second-order information among brain regions into brain network and applied it into identifying mild cognitive impairment Considering the correlation between the two brain regions may be affected by other regions, Guo et al (2014) proposed to eliminate this kind influence from other connectivities through the partial correlation (ParC) test, and applied it to the classification of schizophrenia patients and healthy controls. Li et al (2019) and Qiao et al (2016) constructed the low rank and self-weighted brain network by introducing prior knowledge to the model. Zhou et al (2018) embedded the second-order information among brain regions into brain network and applied it into identifying mild cognitive impairment
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
