Bipolar disorder (BD), also known as manic-depressive disorder, is a severe chronic psychiatric disorder characterized by alternating episodes of depression and mania or hypomania, accompanied by varying degrees of cognitive impairment, self-injury and suicidal tendencies. It is also characterized by a high prevalence, a high relapse rate, and a high suicide rate (Van Meter et al., 2011). According to studies, approximately two-thirds of BD patients have their onset in childhood and adolescence, which is known as early-onset BD (Hughes et al., 2016). Patients with early onset have more frequent emotional episodes, more comorbidities, a higher risk of suicide, and a worse response to medication (Miller and Black, 2020). The earlier onset predicts a worrying outcome that seriously harms the physical health, mental health, and family and social relationships of children and adolescents.

Grey matter and white matter microstructures are altered in BD patients (Padmanabhan et al., 2015; Lu et al., 2011), and aberrant neuronal activity and connection pathway disruption in the cerebral cortex and subcortical regions have also been demonstrated (He et al., 2016). When assessing facial emotions and executive working memory, the functional connectivity (FC) between prefrontal areas and the amygdala is reduced in BD patients (Passarotti et al., 2012a), and their prefrontal-striato-thalamic circuits also have aberrant FC (Zhang et al., 2022). According to these findings, the neurological dysfunction in BD may include aberrant prefrontal-marginal-subcortical connections as an underlying biological trait. The default mode network (DMN) and the salience network (SN) were found to have abnormal connections in a systematic review of patients with early-onset BD. Dysfunction in the cortico-limbic and cortico-striatal circuits, which primarily involve the occipital and frontal lobes, amygdala, hippocampus, insula, thalamus, and striatum, was also reported (Cattarinussi et al., 2022). Pertinently, the brain is a complex and sophisticated network, and abnormalities in specific pathways or regional activity do not yet fully account for functional changes at the level of the entire neural network. The majority of recent neuroimaging studies have focused on examining BD in youths and adults; nonetheless, the aetiology and pathophysiology of BD with early onset are still not completely understood. Little is known about the pattern of brain network dysfunction in those with early-onset BD. As a result, we need to further evaluate the resting-state large-scale brain network in early-onset BD patients.

Independent component analysis (ICA) divides the neural signal activity of the brain over time into functional connectivity networks containing coupled spatial and temporal information. ICA can effectively detect resting-state networks (RSNs), that is, the voxel spatial pattern of synchronous oscillations (Song et al., 2013). Previous research has hypothesized that abnormal connectivity of large-scale neurofunctional networks may underlie BD pathophysiology (Perry et al., 2019). Therefore, in this study, we used the ICA method to construct large-scale functional brain networks, detect changes in FC within RSNs and in functional network connectivity (FNC) between RSNs in early-onset BD patients, and analyse the correlation between the connectivity values of differential brain regions or networks and neuropsychological scale scores to provide objective neuroimaging data of functional network changes in early-onset BD patients.