November 2024
Longitudinal structural MRI (sMRI) may be used to characterize brain morphological changes over time. A key requirement for this approach is accurate rigid-body alignment of longitudinal sMRI. We have recently developed the automatic temporal registration algorithm (ATRA) for this purpose. ATRA is a landmark-based approach capable of registering dozens of serial sMRI simultaneously in an unbiased manner. The aim of the research presented in this paper was to evaluate the accuracy and inverse-consistency of ATRA in comparison to three commonly used sMRI alignment methods: FSL, FreeSurfer, and ANTS. In the absence of a ground truth, it is only possible to quantitatively determine the degree of discrepancy between two algorithms. We propose that if the discrepancy exceeds a certain threshold, the relative accuracy of the two algorithms could be determined visually. We computed the discrepancy between ATRA and each of the three other methods for the alignment of 150 pairs of sMRI taken roughly one year apart. We visually rated the accuracy of alignments in cases where the discrepancy was greater than .5 mm while the rater was agnostic to the registration method. In those instances, ATRA was considered more accurate than FSL in 46 out of 48 cases, more accurate than FreeSurfer in 6 out of 7 cases, and more accurate than ANTS in all 6 cases. ATRA was also the most inverse-consistent method. In addition to being capable of performing unbiased group-wise registration, ATRA is the most accurate algorithm in comparison to several commonly used rigid-body alignment methods.