Aim: Despite accelerating increases in nonsuicidal self-injury (NSSI), identifying promising biomarkers for NSSI and their clinical implications still remains challenging. Our study aimed to investigate possible alterations in brain network topologies and their association with emotional processing in individuals engaging in NSSI by leveraging novel computational methods of individualized structural covariance networks (SCNs) and graph theoretical analysis.Methods: Participants engaging in NSSI (n = 61) and age, sex, and handedness matched controls (n = 62) underwent anatomical T1-weighted MRI scanning and completed self-report questionnaires for emotion dysregulation and emotional contagion traits. After constructing individualized SCNs using cortical thickness measures, graph theoretical analysis was applied to examine the topological properties of structural covariance networks. Results: Compared to controls, individuals engaging in NSSI exhibited significantly reduced global clustering, as well as reduced local clustering coefficient in the right insular sulcus, right middle frontal gyrus, right angular gyrus, left inferior temporal gyrus, and left anterior occipital sulcus (FDR-corrected P < .05). Hubs unique to the NSSI group were the right middle posterior cingulate cortex (mPCC), left middle occipital gyrus, left orbital gyrus, right precentral gyrus, and left superior temporal sulcus. The local clustering coefficient of the right insular sulcus was associated with emotional contagion for positive emotions, and the hubness of the mPCC was associated with emotion dysregulation in the NSSI group. Conclusion: Our findings constitute a first step toward identifying the neurocircuitry-based biomarkers of NSSI, which may be characterized by alterations in structural brain networks and their connections to emotional processing.