OBJECTIVE: Mental health care is an important issue in areas stricken by a natural disaster. Some people develop stress-related mental disorders while others are resilient to the traumatic events. In this study, we evaluated the relationships between resilience and structural neural networks derived from the gray matter MRI scan of the brain by using a novel similarity-based approach. METHOD: Participants were 99 healthy subjects who underwent a 1.5-tesla MRI scan and the Connor-Davidson Resilience Scale (CD-RISC) test approximately 1 year on average after the Great East Japan Earthquake. We computed network metrics such as small world properties, the geometric characters of the whole-brain network and degree, betweenness centrality, and clustering coefficient, which were the regional geometric characters and mapped onto the image. RESULTS: Regarding small world properties, there were no significant correlations between the brain network indices and the CD-RISC total score. However, there was a significant negative correlation between the CD-RISC total score and the betweenness centrality in the right anterior cingulate cortex. No significant relationship was found for the CD-RISC total score with regional degree, clustering coefficient, or gray matter volumes. CONCLUSIONS: Our data suggest that the regional brain network at the right anterior cingulate cortex is associated with resilience in people who are exposed to a great natural disaster. The brain connectome may provide adjunct biological information to understand trauma- and stressor-related disorders. (PsycInfo Database Record (c) 2022 APA, all rights reserved).