Reproducibility of the Structural Brain Connectome Derived from Diffusion Tensor Imaging

Abstract

Rationale: Disruptions of brain anatomical connectivity are believed to play a central role in several neurological and psychiatric illnesses. The structural brain connectome is typically derived from diffusion tensor imaging (DTI), which may be influenced by methodological factors related to signal processing, MRI scanners and biophysical properties of neuroanatomical regions. In this study, we evaluated how these variables affect the reproducibility of the structural connectome.

Methods: Twenty healthy adults underwent 3 MRI scanning sessions (twice in the same MRI scanner and a third time in a different scanner unit) within a short period of time. The scanning sessions included similar T1 weighted and DTI sequences. Deterministic or probabilistic tractography was performed to assess link weight based on the number of fibers connecting gray matter regions of interest (ROI). Link weight and graph theory network measures were calculated and reproducibility was assessed through intra-class correlation coefficients, assuming each scanning session as a rater.

Results: Connectome reproducibility was higher with data from the same scanner. The probabilistic approach yielded larger reproducibility, while the individual variation in the number of tracked fibers from deterministic tractography was negatively associated with reproducibility. Links connecting larger and anatomically closer ROIs demonstrated higher reproducibility. In general, graph theory measures demonstrated high reproducibility across scanning sessions.

Discussion: Anatomical factors and tractography approaches can influence the reproducibility of the structural connectome and should be factored in the interpretation of future studies. Our results demonstrate that connectome mapping is a largely reproducible technique, particularly as it relates to the geometry of network architecture measured by graph theory methods.

Fig 1. Average connectivity matrices from all subjects for each scanning session (Time 1 and 2 in Scanner A represent measures from the same MRI scanner in different time points, and Time 1 in Scanner B represents a third measure from a different MRI scanner).

Each matrix element represents the weighted connectivity between the ROIs indicated by the column and by the row. The color bars indicate log(weighted connectivity).

Fig 2. Link-wise ICCs.

Each matrix entry represents the ICC observed for the white matter link between the gray matter ROI in the row and the gray matter ROI in the column.

Fig 3. This figure demonstrates each connectome link represented by a line corresponding to the center of mass of the bundle of fibers associated with that link (estimated from deterministic tractography).

Each link is color-coded based on its reproducibility per tractography approach and scanner usage. The colorbars indicate the link-wise ICC.

Fig 4. This figure demonstrates the association between the anatomical properties of each link and link-wise reproducibility.

The anatomical properties are Euclidian distance between connected gray matter ROIs, sum of the volume of the connected ROIs and the ratio of the volumes between the connected ROIs. Link-wise reproducibility is determined by ICCs. The scatter plots demonstrate each anatomical property in the y-axis, and the ICCs in the x-axis. The statistical relationship between these measures is defined by a correlation coefficient, whose details are displayed below each scatter plot.

Fig 5. The scatter plots demonstrate the relationship between link-wise graph theory metrics obtained from connectomes calculated from scanning session in time 1 (x-axis) and in time 2 (y-axis) within the same MRI scanner.

The scale set for the x-axis is the same as for the y-axis for all graphs. The ICC between each pair of measurements is displayed below each scatter plot. Of note, the relationship between degrees was not assessed for probabilistic tractography given the low sparsity of networks generated from probabilistic methods, therefore leading to a ceiling degree effect.

Fig 6. The scatter plots demonstrate the relationship between link-wise graph theory metrics across different scanners (Time 1, Scanner in x-axis and Time 1 Scanner B in y-axis).

The ICC between each pair of measurements is displayed below each scatter plot. Similarly, the relationship between degrees was not assessed for probabilistic tractography given the low sparsity of networks generated from probabilistic methods, therefore leading to a ceiling degree effect.