Morphometric Similarity Networks, or MSNs, represent the brain as a network where nodes are distinct brain regions, and edges represent statistical similarity in morphometric properties across a population. Strong edges indicate that two regions tend to have similar structural measures, like cortical thickness or volume, across different individuals.
Compared to functional and structural connectivity, MSNs have unique advantages and disadvantages. They reflect slower biological processes like genetic influences and developmental trajectories, are less susceptible to artifacts, and use standard MRI data. However, they provide indirect relationships and are population-dependent, requiring groups of subjects rather than individual networks.
The key assumption of MSNs is that brain regions showing similar patterns of morphometric variation across a population are biologically related through shared genetic factors, developmental processes, or common vulnerability. Unlike functional connectivity which uses time series, or structural connectivity which uses tract properties, MSNs use morphometric measures like cortical thickness, surface area, and gray matter volume.
MSNs have been widely applied to study neurological and neurodegenerative diseases. In Alzheimer's disease, they reveal altered network topology and weakened covariance in regions like the medial temporal lobe. Parkinson's disease shows changes in basal ganglia networks, while schizophrenia displays disrupted neurodevelopmental patterns. Multiple sclerosis demonstrates how white matter damage affects gray matter covariance.
Research findings show that MSNs reveal network topology alterations that reflect disease progression, with specific connection changes correlating with clinical symptoms. They serve as potential biomarkers for early disease detection and help track atrophy spread patterns. MSNs provide a complementary perspective to functional and structural connectivity, offering comprehensive insights into brain organization and disease mechanisms.