Altered static and dynamic functional network connectivity and combined Machine learning in asthma

Abstract

Asthma is a reversible disease characterized by airflow limitation and chronic airway inflammation. Previous neuroimaging studies have shown structural and functional abnormalities in the brains of individuals with asthma. However, earlier research has primarily focused on static changes in brain activity, neglecting the effects of asthma on the dynamic characteristics of functional brain networks. This study included 31 asthma patients and 31 healthy controls (HCs). Independent component analysis (ICA) was employed to extract changes in static functional network connectivity (sFNC) and dynamic functional network connectivity (dFNC) from the acquired data. Compared to the HC group, the overall functional connectivity (FC) within the visual network (VN) in asthma patients declined, whereas the FC in the auditory network (AN) and cerebellar network (CN) increased. Additionally, functional network connectivity (FNC) analysis revealed enhanced connectivity between the VN and AN, as well as between the VN and executive control network (ECN), while AN-AN functional connectivity was reduced. The dFNC was primarily characterized by abnormal connections among the default mode network (DMN), AN, and other brain regions. The support vector machine (SVM) model based on FC and FNC demonstrates excellent performance in distinguishing asthma patients from HCs. Our findings highlight significant alterations in functional connectivity within the sFNC and dFNC of asthma patients. These results enhance our understanding of the potential neurobiological mechanisms underlying emotional deficits and cognitive impairments in asthma patients. Furthermore, they provide additional neuroimaging evidence that may be helpful for researchers in identifying potential neurobiological markers to differentiate asthma patients from HCs.