Diffusion magnetic resonance imaging provides a useful tool to examine structural white matter changes in health and disease. Tract-based spatial statistics (TBSS)¹ is a popular tool for the evaluation of diffusion data in group studies. As a quantitative measure, TBSS typically analyses the fractional anisotropy (FA) to find statistically significant group differences, whereby tractography is not explicitly performed. Calamante proposed a method to create super-resolution track-density (TD) maps² based on diffusion fiber tractograms. Nevertheless, the resulting super resolution TD images seem unfeasible as a quantitative marker due to biases in the tractograms, introduced by seeding strategies or intrinsic tracking algorithm properties. Daducci introduced a Convex Optimization framework called COMMIT³, whereby a local forward model and a global optimization algorithm improve the fit of the tractogram with regard to the measured diffusion signal. This optimization leads to reduced false positive fibers and algorithm specific tracking biases in the tractogram. In this study, we compare optimized TD evaluated using TBSS and tract profiles derived from automating fiber-tract quantification (AFQ) analysis⁴.Datasets of 29 unmedicated adult patients suffering from major depressive disorder and 37 healthy controls (age and gender matched) were acquired on a Philips Achieva 3T system. The following scan parameters were used for the diffusion acquisition: TR: 13.4s, TE: 55ms, FOV: 240x240mm2, with 72 contiguous slices, slice thickness: 2mm, acquisition matrix of 120x120, SENSE factor of 2. Diffusion-weighted images were acquired along 32 directions distributed uniformly on a half-sphere with a b-value of 1000 s/mm2 in addition to a b = 0 scan. Additionally, 1mm isotropic T1-weighted structural images were recorded. For each dataset, the diffusion data was corrected for eddy-currents and subject motion using FSL⁵ and susceptibility induced distortions were reduced by applying the recently proposed INVERSION method⁶. Whole brain fiber tracking was performed using the constrained spherical deconvolution approach and the iFOD2 algorithm implemented in MRtrix package⁷. Tissue priors were estimated from T1 images in FSL and used for the anatomically constrained tractography with SIFT inspired dynamic seeding. A total of 2.5 million fibers were generated per subject. Furthermore, each tractogram was optimized with the COMMIT framework using an intra-cellular stick model and two isotropic compartments (CSF, gray matter). The derived fiber weights from COMMIT were used to calculate an optimized high-resolution (1.25mm isotropic) track density for each. TBSS analysis was performed with the FA and optimized track density (optTD) maps. For the optTD map, the skeleton and transformation parameters from the TBSS analysis of the FA images were applied. Results were corrected for multiple comparison (P<.05). Additionally, the optTD was evaluated along the 20 major fiber bundles using the AFQ framework. To check the optimization quality of the COMMIT framework, fit errors were also evaluated using AFQ.The TBSS results of the FA did not reveal any significant group differences, whereas the optTD showed multiple significant clusters, predominantly in the left hemisphere, where the patient group exhibits a larger optTD compared to the control group. Overlapping results from TBSS and AFQ analyses are shown in Figure 1 and 2. Fit errors from the COMMIT optimization did not show any significant group differences in the AFQ analysis.The FA maps did not show any significant changes between the two groups, whereas the optTD was sensitive to group differences in TBSS and AFQ analyses. Both evaluation methods have limitations, such as normalization errors and back-projection step for TBSS and fiber segmentation accuracy for AFQ. Therefore, the overlapping significant clusters hint to a rather robust group difference, compared to results derived from a single analysis. The optTD is reported as absolute values but is highly dependent on the acquisition protocol and the used fiber-forward model in the optimization.We showed a high sensitivity in the newly introduced optTD compared to the FA in this particular example. Significant group differences were found using TBSS and AFQ analyses indicating robust findings. High resolution maps derived from tractograms provide promising novel contrasts to investigate white matter abnormalities in mental disorders.