Mini-Review of Studies Reporting the Repeatability and Reproducibility of Diffusion Tensor Imaging

ABSTRACT

PURPOSE: Diffusion tensor imaging (DTI) data must be analyzed by an analyzer after data processing. Hence, the analyzed data of DTI might depend on the analyzer, making it a major limitation. This paper reviewed previous DTI studies reporting the repeatability and reproducibility of data from the corticospinal tract (CST), one of the most actively researched neural tracts on this topic. MATERIALS AND METHODS: Relevant studies published between January 1990 and December 2018 were identified by searching PubMed, Google Scholar, and MEDLINE electronic databases using the following keywords DTI, diffusion tensor tractography, reliability, repeatability, reproducibility, and CST. As a result, 15 studies were selected. RESULTS: Measurements of the CSTs using region of interest methods on 2-dimensional DTI images generally showed excellent repeatability and reproducibility of more than 0.8 but high variability (0.29 to 1.00) between studies. In contrast, measurements of the CST using the 3-dimensional DTT method not only revealed excellent repeatability and reproducibility of more than 0.9 but also low variability (repeatability, 0.88 to 1.00; reproducibility, 0.82 to 0.99) between studies. CONCLUSION: Both 2-dimensional DTI and 3-dimensional DTT methods appeared to be reliable for measuring the CST but the 3-dimensional DTT method appeared to be more reliable.