Characterizing the Microstructural Transition at the Gray Matter-White Matter Interface: Implementation and Demonstration of Age-Associated Differences

The cortical gray matter-white matter interface (GWI) is a natural transition zone where the composition of brain tissue abruptly changes and is a location for pathologic change in brain disorders. While diffusion magnetic resonance imaging (dMRI) is a reliable and well-established technique to characterize brain microstructure, the GWI is difficult to assess with dMRI due to partial volume effects and is normally excluded from such studies.BACKGROUNDThe cortical gray matter-white matter interface (GWI) is a natural transition zone where the composition of brain tissue abruptly changes and is a location for pathologic change in brain disorders. While diffusion magnetic resonance imaging (dMRI) is a reliable and well-established technique to characterize brain microstructure, the GWI is difficult to assess with dMRI due to partial volume effects and is normally excluded from such studies.In this study, we introduce an approach to characterize the dMRI microstructural profile across the GWI and to assess the sharpness of the microstructural transition from cortical gray matter (GM) to white matter (WM). This analysis includes cross-sectional data from a total of 146 participants (18-91 years; mean age: 52.4 (SD 21.4); 83 (57%) female) enrolled in two normative lifespan cohorts at Albert Einstein College of Medicine from 2019-2023. We compute the aggregate GWI slope for each parameter, across each of 6 brain regions (cingulate, frontal, occipital, orbitofrontal, parietal, and temporal) for each participant. The association of GWI slope in each region with age was assessed using a linear model, with biological sex as a covariate.METHODSIn this study, we introduce an approach to characterize the dMRI microstructural profile across the GWI and to assess the sharpness of the microstructural transition from cortical gray matter (GM) to white matter (WM). This analysis includes cross-sectional data from a total of 146 participants (18-91 years; mean age: 52.4 (SD 21.4); 83 (57%) female) enrolled in two normative lifespan cohorts at Albert Einstein College of Medicine from 2019-2023. We compute the aggregate GWI slope for each parameter, across each of 6 brain regions (cingulate, frontal, occipital, orbitofrontal, parietal, and temporal) for each participant. The association of GWI slope in each region with age was assessed using a linear model, with biological sex as a covariate.We demonstrate this method captures an inherent change in fractional anisotropy (FA), a