Validation of a Nonrigid Registration Error Detection Algorithm Using Clinical MRI Brain Data

Abstract

Identification of error in non-rigid registration is a critical problem in the medical image processing community. We recently proposed an algorithm that we call “Assessing Quality Using Image Registration Circuits” (AQUIRC) to identify non-rigid registration errors and have tested its performance using simulated cases. In this article, we extend our previous work to assess AQUIRC’s ability to detect local non-rigid registration errors and validate it quantitatively at specific clinical landmarks, namely the Anterior Commissure (AC) and the Posterior Commissure (PC). To test our approach on a representative range of error we utilize 5 different registration methods and use 100 target images and 9 atlas images. Our results show that AQUIRC’s measure of registration quality correlates with the true target registration error (TRE) at these selected landmarks with an R2 = 0.542. To compare our method to a more conventional approach, we compute Local Normalized Correlation Coefficient (LNCC) and show that AQUIRC performs similarly. However, a multi-linear regression performed with both AQUIRC’s measure and LNCC shows a higher correlation with TRE than correlations obtained with either measure alone, thus showing the complementarity of these quality measures. We conclude the article by showing that the AQUIRC algorithm can be used to reduce registration errors for all five algorithms.

Consistency-based rectification of nonrigid registrations

“…These estimations can be evaluated through ground-truth local-error magnitudes for experiments for which those are known. This allows for a comparison with AQUIRC, 19,20 which computes a dimensionless measure of local error magnitude based on (3,0)-consistency. We used our own implementation of their technique, the results of which were presented in Table 1.…”

“…4.5. We also compare our method to our own implementation of assessing quality using image registration circuits (AQUIRC), 19,20 which computes a dimensionless measure of local registration error magnitude which can be correlated with the true local error magnitudes. It is observed that the magnitude of the updates computed by CBRR correlate substantially stronger with the true error magnitudes in comparison to AQUIRC.…”

“…9,16 Recently, a method was proposed to exploit redundancy in multiple registration circles for estimating the spatial location and magnitude of errors in pair-wise registrations. 19,20 Using a simplified model for the accumulation of registration errors, this method can estimate a dimensionless measure for local registration error, which was shown therein to correlate with true errors. A similar method was also used to predict the segmentation quality for multi-atlas segmentations.…”

“…Importantly, our algorithm allows for rectifying nonrigid registrations, in contrast to a previous method for rigid registrations. 30 Our method extends an approach 19,20 presented earlier, which computes dimensionless estimates of nonrigid registration errors by optimizing a surrogate model of inconsistency in mutually registered sets of images. In contrast, our work directly minimizes actual registration inconsistencies within registrations.…”

Quadratic: quality of dice in registration circuits

“…Quantifying the error in correspondence as a result of NRR in a target image is a challenging problem, with some calling automatic registration techniques impossible to validate[5]. Datteri et al have developed the AQUIRC (Assessing Quality Using Image Registration Circuits)[6, 7] model to assess registration quality is target images using registration circuits. The authors proposed to use registration circuits to model the local error associated with NRR and have shown moderate success in applying this to multi-atlas segmentation[8].…”