Highly accurate inverse consistent registration: A robust approach

The registration of images is a task that is at the core of many applications in computer vision. In computational neuroimaging where the automated segmentation of brain structures is frequently used to quantify change, a highly accurate registration is necessary for motion correction of images taken in the same session, or across time in longitudinal studies where changes in the images can be expected. This paper, inspired by Nestares and Heeger (2000), presents a method based on robust statistics to register images in the presence of differences, such as jaw movement, differential MR distortions and true anatomical change. The approach we present guarantees inverse consistency (symmetry), can deal with different intensity scales and automatically estimates a sensitivity parameter to detect outlier regions in the images. The resulting registrations are highly accurate due to their ability to ignore outlier regions and show superior robustness with respect to noise, to intensity scaling and outliers when compared to state-of-the-art registration tools such as FLIRT (in FSL) or the coregistration tool in SPM.

Research HighlightsThe main new contributions of this work are: ► inverse consistency (necessary to allow for unbiased downstream processing), ► automatic parameter estimation to adjust for different image situations, and ► intensity scale estimation.Applications of this method are: ► longitudinal processing of brain MRI data, ► motion correction/averaging of intra-session scans to improve SNR, and ► unbiased rigid initialization for higher-dimensional warps.Significance: ► Due to change in the images (true neurodegeneration, differential positioning of the tongue, jaws, eyes, neck, different cutting planes as well as session-dependent imaging distortions such as susceptibility effects) non-robust registration as in most standard tools cannot accurately align the images. ► The registration is significantly influenced by these ‘outlier’ voxels. ► These outliers are very common in MRI data and need to be treated for longitudinal processing or motion correction for the purpose of averaging (noise reduction). ► Furthermore, the inverse consistency is of significance to remove a bias with respect to any of the time points in a longitudinal study, that is introduced by the standard non-symmetric methods.