Abstract
Purpose: Fiber tracking with diffusion weighted magnetic resonance imaging has become an essential tool for estimating in vivo brain white matter architecture. Fiber tracking results are sensitive to the choice of processing method and tracking criteria. Phantom studies provide concrete quantitative comparisons of methods relative to absolute ground truths, yet do not capture variabilities because of in vivo physiological factors. Methods: To date, a large-scale reproducibility analysis has not been performed for the assessment of the newest generation of tractography algorithms with in vivo data. Reproducibility does not assess the validity of a brain connection however it is still of critical importance because it describes the variability for an algorithm in group studies. The ISMRM 2017 TraCED challenge was created to fulfill the gap. The TraCED dataset consists of a single healthy volunteer scanned on two different scanners of the same manufacturer. The multi-shell acquisition included b-values of 1000, 2000 and 3000 s/mm2 with 20, 45 and 64 diffusion gradient directions per shell, respectively. Results: Nine international groups submitted 46 tractography algorithm entries. The top five submissions had high ICC > 0.88. Reproducibility is high within these top 5 submissions when assessed across sessions or across scanners. However, it can be directly attributed to containment of smaller volume tracts in larger volume tracts. This holds true for the top five submissions where they are contained in a specific order. While most algorithms are contained in an ordering there are some outliers. Conclusion: The different methods clearly result in fundamentally different tract structures at the more conservative specificity choices (i.e., volumetrically smaller tractograms). The data and challenge infrastructure remain available for continued analysis and provide a platform for comparison.
