Abnormal topological organization of T1 and quantitative susceptibility mapping covariance networks in Alzheimer’s disease

Motivation: T1 mapping and QSM are effective in detecting Alzheimer's disease, but their covariance networks have not been fully explored. Goal(s): To determine whether network-based metrics can aid in early diagnosis and intervention for AD. Approach: Construct covariance networks using QSM and T1 maps. Topological metrics —clustering coefficient, path length, and small-world index —were calculated to assess group differences. Correlation analysis examined the relationship between these metrics and cognitive function. Results: There are significant differences among the three groups. The AD and MCI groups exhibited higher network parameters, with altered hub distributions. Network parameters are negatively correlated with cognition. Impact: These findings provide new insights into AD pathology by revealing disrupted network organization, offering potential biomarkers for early diagnosis, intervention, and disease monitoring.