Elsevier

NeuroImage

Volume 53, Issue 4, December 2010, Pages 1181-1196
NeuroImage

Highly accurate inverse consistent registration: A robust approach

https://doi.org/10.1016/j.neuroimage.2010.07.020Get rights and content

Abstract

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 Highlights

The 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.

Introduction

There is great potential utility for information extracted from neuroimaging data to serve as biomarkers, to quantify neurodegeneration, and to evaluate the efficacy of disease-modifying therapies. Currently, the accurate and reliable registration of images presents a major challenge, due to a number of factors. These include differential distortions that affect longitudinal time points in different ways; true, localized anatomical change that can cause global offsets in the computed registration, and the lack of inverse consistency in which the registration of multiple images depends on the order of processing, which can lead to algorithm-induced artifacts in detected changes. Thus, the development of an accurate, robust and inverse consistent method is a critical first step to quantify change in neuroimaging or medical image data in general.

Since the object of interest is typically located differently in each acquired image, accurate geometric transformations are necessary to register the input images into a common space. Approaches based on robust statistics are extremely useful in this domain, as they provide a mechanism for discounting regions in the images that contain true differences, and allow one to recover the correct alignment based on the remainder of the data. Inverse consistency is critical to avoid introducing bias into longitudinal studies. A lack of inverse consistency in registration is likely to bias subsequent processing and analysis, as documented in Yushkevicha et al. (2009). The goal of this work is thus to develop a robust and inverse consistent registration method for use in the analysis of neuroimaging data. The core application of this technique is intra-modality and intra-subject registration with important implications for:

  • 1.

    Motion correction and averaging of several intra-session scans to increase the signal to noise ratio,

  • 2.

    highly accurate alignment of longitudinal image data and

  • 3.

    initial registration for higher-dimensional warps.

Although the remainder of this paper deals with neuroimaging data, the method can be used for other image registration task as well.

Highly accurate rigid registrations are of importance when averaging multiple scans taken within a session to reduce the influence of noise or subject motion. Since it is nearly impossible for a person to remain motionless throughout a 20 minute scan, image quality can be increased by taking shorter scans and performing retrospective motion correction (Kochunov et al., 2006). Many common sequences are short enough to allow for several structural scans of the same modality within a session. Here even a slightly inaccurate registration will introduce additional artifacts into the final average and likely reduce the accuracy, sensitivity and robustness of downstream analysis.

Compared with cross-sectional studies, a longitudinal design can significantly reduce the confounding effect of inter-individual morphological variability by using each subject as his or her own control. As a result, longitudinal imaging studies are becoming increasingly common in clinical and scientific neuroimaging. Degeneration in subcortical structures and cortical gray matter is, for example, manifested in aging (Jack et al., 1997, Salat et al., 1999, Salat et al., 2004, Sowell et al., 2003, Sowell et al., 2004), Alzheimer's disease (Dickerson et al., 2001, Thompson et al., 2003, Lerch et al., 2005), Huntington's disease (Rosas et al., 2002), multiple sclerosis (Sailer et al., 2003) and Schizophrenia (Thompson et al., 2001, Kuperberg et al., 2003, Narr et al., 2005) and has been useful towards understanding some of the major pathophysiological mechanisms involved in these conditions. As a result, in vivo cortical thickness and subcortical volume measures are employed as biomarkers of the evolution of an array of diseases, and are thus of great utility for evaluating the efficacy of disease-modifying therapies in drug trials. To enable the information exchange at specific locations in space, highly accurate and unbiased registrations across time are necessary. They need to be capable of efficiently dealing with change in the images, which can include true neurodegeneration, differential positioning of the tongue, jaws, eyes, neck, different cutting planes as well as session-dependent imaging distortions such as susceptibility effects.

As an example see Fig. 1 showing longitudinal tumor data (same slice of five acquisitions at different times, MPRAGE, 256 × 256 × 176, 1 mm voxels) registered to the first time point (left) with the proposed robust method. The five time points are: 5 days prior to the start of treatment, 1 day prior, 1 day after the start of treatment, and 28, 56 days after the start of treatment. Despite of the significant change in these images the registration is highly accurate (verified visually in non-tumor regions). The bottom row depicts the outlier weights (red/yellow overlay), which are blurry regions of values between 0 (outlier) and 1 (regular voxel) that label differences in the images. In addition to the longitudinal change in tumor regions and consequential deformation (e.g. at the ventricles), the robust method also picks up differences in the scalp, eye region and motion artifacts in the background. In our robust approach the influence of these differences (or outliers) is reduced when constructing the registrations, while they have a detrimental influence on the final registration result in non-robust methods.

Statistically, robust parameter estimation has a history of supplying solutions to several computer vision problems (Stewart, 1999) as it is capable of estimating accurate model parameters in the presence of noise, measurement error (outliers) or true differences (e.g. change over time). The approach presented here is based on robust statistics and inspired by Nestares and Heeger (2000), who describe a robust multi-resolutional registration approach to rigidly register a set of slices to a full resolution image. Our approach, however, is designed to be inverse consistent to avoid introducing a bias. It also allows the calculation of an additional global intensity scale parameter to adjust for different intensity scalings that can be present especially in longitudinal data. A more complex intensity pre-processing is therefore not needed in most cases. Furthermore, we automatically estimate the single parameter of the algorithm that controls its sensitivity to outliers. This is a necessary addition, since a fixed parameter cannot adequately deal with different image intensity scales, which are common in MRI. In addition to the multi-resolutional approach described in Nestares and Heeger (2000) we use moments for an initial coarse alignment to allow for larger displacements and situations where source and target may not overlap. Finally, we describe the registration of two full resolution images (instead of only a set of slices) and explain how both rigid and affine transformation models can be used in the symmetric algorithm. We demonstrate that our approach yields highly accurate registrations in brain regions and outperforms existing state-of-the-art registration algorithms.

The remainder of this paper is organized as follows. After discussing related work and introducing the theoretical background, such as robust statistics in Background, we present our symmetric registration model, different transformation models as well as intensity scaling in Robust symmetric registration. Then we describe the registration algorithm in detail, taking care that the properties of the theory are carried over to the implementation (Registration algorithm). We specifically focus on maintaining inverse consistency by resampling both images into a ‘half way’ space in intermediate steps as opposed to resampling the source at the estimated target location. This asymmetric sampling, which is commonly used, introduces a bias as the target image will not be resampled at all, and will thus be less smooth than the resampled source. In Results we demonstrate the superiority of the proposed method over existing registration algorithms with respect to symmetry, robustness and accuracy on synthetic and real data as well as a motion correction application. The software implementing the presented robust registration is publicly distributed as part of the FreeSurfer (surfer.nmr.mgh.harvard.edu) software package as mri_robust_register.

Section snippets

Related work on registration

Over the last 20 years methods for the registration of images (and in particular medical images) have been studied intensely (see e.g. Maintz and Viergever, 1998, Maes et al., 1999, Hill et al., 2001 for surveys and comparisons). Many different applications domains exist for registration, including multimodal intra-subject registration, cross-subject volumetric registration, surface-based registration etc…, each of which require domain-specific approaches to maximize accuracy. Some of the most

Robust symmetric registration

As described above, the first step in constructing a robust simultaneous alignment of several images into an unbiased common space for a longitudinal study or for motion correction, is to register two images symmetrically. To avoid any bias, the resulting registration must be inverse consistent, i.e., the same registration (inverse transformation) should be computed by the algorithm if the time points are swapped.

Registration algorithm

The algorithm consists of the following steps:

  • 1.

    Initialize Gaussian pyramid: by subsampling and smoothing the images.

  • 2.

    Initialize alignment: compute a coarse initial alignment using moments at the highest resolution.

  • 3.

    Loop resolutions: iterate through pyramid (low to high resolution).

  • 4.

    Loop iterations: on each resolution level iterate registration to obtain best parameter estimate. For each iteration step:

    • a)

      Symmetry: take the current optimal alignment, map and resample both images into a half way space

Results

This section presents results quantifying the accuracy and robustness of the robust registration in comparison to other commonly used methods. As mentioned above, the robust registration is capable of ignoring outlier regions. This can be verified when checking the weights during a successful registration, as shown in Fig. 6.

The top images show the (enhanced) differences between the target and registered source. The regions that contain the strongest differences are correctly detected as

Conclusion

In this work a robust registration method based on Nestares and Heeger (2000) is presented, with additional properties such as initial coarse alignment, inverse consistency, sensitivity parameter estimation and global intensity scaling. Automatic intensity scaling is necessary for the method to function when global intensity differences exist. Similarly the automatic estimation of the saturation parameter avoids misalignment in specific image situations where a fixed value potentially ignores

Acknowledgments

Support for this research was provided in part by the National Center for Research Resources (P41-RR14075, and the NCRR BIRN Morphometric Project BIRN002, U24 RR021382), the National Institute for Biomedical Imaging and Bioengineering (R01 EB006758), the National Institute on Aging (R01 AG022381, U54 AG024904), and the National Institute for Neurological Disorders and Stroke (R01 NS052585-01, R01 NS042861, P01 NS058793). Additional support was provided by The Autism & Dyslexia Project funded by

References (74)

  • M. Jenkinson et al.

    Improved optimization for the robust and accurate linear registration and motion correction of brain images

    Neuroimage

    (2002)
  • F. Maes et al.

    Comparative evaluation of multiresolution optimization strategies for multimodality image registration by maximization of mutual information

    Med. Image Anal.

    (1999)
  • J. Maintz et al.

    A survey of medical image registration

    Med. Image Anal.

    (1998)
  • S. Periaswamy et al.

    Medical image registration with partial data. Medical Image Analysis

  • Z.S. Saad et al.

    A new method for improving functional-to-structural MRI alignment using local Pearson correlation

    Neuroimage

    (2009)
  • J. Ashburner et al.

    Nonlinear spatial normalization using basis functions

    Hum. Brain Mapp.

    (1999)
  • B. Avants et al.

    Geodesic estimation for large deformation anatomical shape averaging and interpolation

    Neuroimage

    (2004)
  • M.F. Beg et al.

    Symmetric data attachment terms for large deformation image registration

    IEEE Trans. Med. Imaging

    (2007)
  • F.L. Bookstein

    Principal warps: thin-plate splines and the decomposition of deformations

    IEEE Trans. Pattern Anal. Mach. Intell.

    (1989)
  • F.L. Bookstein

    Thin-plate splines and the atlas problem for biomedical images

  • S.H. Cheng et al.

    Approximating the logarithm of a matrix to specified accuracy

    SIAM J. Matrix Anal. Appl.

    (2001)
  • G.E. Christensen et al.

    3D brain mapping using a deformable neuroanatomy

    Phys. Med. Biol.

    (1994)
  • G.E. Christensen et al.

    Consistent image registration

    IEEE Trans. Med. Imaging

    (2001)
  • A. Collignon et al.

    Automated multi-modality image registration based on information theory

  • D. Collins et al.

    Automatic 3-D model-based neuroanatomical segmentation

    Hum. Brain Mapp.

    (1995)
  • A. Del Bimbo

    Visual Information Retrieval

    (1999)
  • A. Evans et al.

    Anatomical–functional correlative analysis of the human brain using three-dimensional imaging systems

    Proc. SPIE Int. Soc. Opt. Eng.

    (1989)
  • R. Frackowiak et al.
  • J.C. Gee et al.

    Elastically deforming a three-dimensional atlas to match anatomical brain images

    Comput. Assisted Tomogr.

    (1993)
  • P. Hellier et al.

    Hierarchical estimation of a dense deformation field for 3-D robust registration

    IEEE Trans. Med. Imaging

    (2001)
  • D.L.G. Hill et al.

    Medical image registration

    Phys. Med. Biol.

    (2001)
  • P.W. Holland et al.

    Robust regression using iteratively reweighted least-squares

    Commun. Stat. Theory Methods

    (1977)
  • P.J. Huber

    Robust estimation of a location parameter

    Ann. Math. Stat.

    (1964)
  • C.R. Jack et al.

    Medial temporal atrophy on MRI in normal aging and very mild Alzheimer's disease

    Neurology

    (1997)
  • M. Jenkinson

    Measuring transformation error by RMS deviation

  • W. Kerwin et al.

    Active edge maps for medical image registration

  • P. Kochunov et al.

    Retrospective motion correction protocol for high-resolution anatomical MRI

    Hum. Brain Mapp.

    (2006)
  • Cited by (978)

    • Privacy preserving image registration

      2024, Medical Image Analysis
    View all citing articles on Scopus
    View full text