Past events: Anastasia Yendiki. Automated reconstruction of white-matter pathways for cross-sectional and longitudinal group studies.

What: Automated reconstruction of white-matter pathways for cross-sectional and longitudinal group studies

Where: BCBL Auditorium

Who: Anastasia Yendiki, PhD., Martinos Faculty, Laboratory for Computational Neuroimaging, Athinoula A. Martinos Center for Biomedical Imaging, Charlestown, USA.

When: 12 PM, noon.

In this talk I will discuss current development for our automated diffusion MRI tractography tool, TRACULA (TRActs Constrained by UnderLying Anatomy). TRACULA reconstructs major white-matter pathways by using prior information on the anatomical neighborhood of each pathway, as a function of position along the pathway’s trajectory. The benefit of this approach is that it does not constrain the shape, size, or integrity of the pathways to be the same in the study subjects as in the atlas subjects. It only constrains the position of the pathways relative to other major anatomical structures, and it does so in a probabilistic manner rather than using hard constraints. Recently, we have extended the TRACULA framework to analyze longitudinal diffusion MRI data. Instead of performing tractography at each time point independently, as if it were a cross-sectional data point, we reconstruct the pathways jointly, using the data from all time points at once. This improves both the test-retest reliability and the sensitivity of the method for detecting subtle changes in diffusion MRI measures and localizing them along the trajectory of a pathway.