## Multimodality Image Registration by Maximization of Mutual Information (1997)

### Cached

### Download Links

- [web.mit.edu]
- [www.csee.wvu.edu]
- [idealab.ucdavis.edu]
- DBLP

### Other Repositories/Bibliography

Venue: | IEEE transactions on Medical Imaging |

Citations: | 528 - 8 self |

### BibTeX

@ARTICLE{Maes97multimodalityimage,

author = {Frederik Maes and André Collignon and Dirk V and Guy Marchal and Paul Suetens},

title = {Multimodality Image Registration by Maximization of Mutual Information},

journal = {IEEE transactions on Medical Imaging},

year = {1997},

volume = {16},

pages = {187--198}

}

### Years of Citing Articles

### OpenURL

### Abstract

Abstract — A new approach to the problem of multimodality medical image registration is proposed, using a basic concept from information theory, mutual information (MI), or relative entropy, as a new matching criterion. The method presented in this paper applies MI to measure the statistical dependence or information redundancy between the image intensities of corresponding voxels in both images, which is assumed to be maximal if the images are geometrically aligned. Maximization of MI is a very general and powerful criterion, because no assumptions are made regarding the nature of this dependence and no limiting constraints are imposed on the image content of the modalities involved. The accuracy of the MI criterion is validated for rigid body registration of computed tomography (CT), magnetic resonance (MR), and photon emission tomography (PET) images by comparison with the stereotactic registration solution, while robustness is evaluated with respect to implementation issues, such as interpolation and optimization, and image content, including partial overlap and image degradation. Our results demonstrate that subvoxel accuracy with respect to the stereotactic reference solution can be achieved completely automatically and without any prior segmentation, feature extraction, or other preprocessing steps which makes this method very well suited for clinical applications. Index Terms—Matching criterion, multimodality images, mutual information, registration. I.

### Citations

8867 |
Elements of Information Theory
- Cover, Thomas
- 1991
(Show Context)
Citation Context ...which reduces its applicability to some very specific multimodality combinations (PET/MR). In this paper, we propose to use the much more general notion of mutual information (MI) or relative entropy =-=[8]-=-, [22] to describe the dispersive behavior of the 2-D histogram. MI is a basic concept from information theory, measuring the statistical dependence between two random variables or the amount of infor... |

4050 |
Pattern Classification and Scene Analysis
- Duda, Hart
- 1973
(Show Context)
Citation Context ...inear image intensity transformations on the behavior of the MI registration criterion. Other schemes can be used to estimate the image intensity distributions, for instance by using Parzen windowing =-=[9]-=- on a set of samples taken from the overlapping part of both images. This approach was used by Viola et al. [27], who also use stochastic sampling of the floating image to increase speed performance. ... |

795 | Alignment by maximization of mutual information
- Viola, Wells
- 1995
(Show Context)
Citation Context ...t prior segmentation on a large variety of applications. This paper expands on the ideas first presented by Collignon et al. [7]. Related work in this area includes the work by Viola and Wells et al. =-=[27]-=-, [28] and by Studholme et al. [21]. The theoretical concept of MI is presented in Section II, while the implementation of the registration algorithm is described in Section III. In Sections IV, V, an... |

729 | A survey of image registration techniques
- Brown
- 1992
(Show Context)
Citation Context ...lisher Item Identifier S 0278-0062(97)02397-5. 0278–0062/97$10.00 © 1997 IEEE obtained from positron emission tomography (PET) images, etc. The bulk of registration algorithms in medical imaging (see =-=[3]-=-, [16], and [23] for an overview) can be classified as being either frame based, point landmark based, surface based, or voxel based. Stereotactic frame-based registration is very accurate, but inconv... |

362 | Multi-modal volume registration by maximization of mutual information - Viola, Wells - 1996 |

153 |
MRI-PET registration with automated algorithm
- Woods, Mazziotta, et al.
- 1993
(Show Context)
Citation Context ...orrelate equally well. A multiresolution optimization strategy is therefore required, which is not necessarily a disadvantage, as it can be computationally attractive. In the approach of Woods et al. =-=[30]-=- and Hill et al. [12], [13], misregistration is measured by the dispersion of the two-dimensional (2-D) histogram of the image intensities of corresponding voxel pairs, which is assumed to be minimal ... |

68 |
Theory of Statistical Inference and Information
- Vajda
- 1989
(Show Context)
Citation Context ... reduces its applicability to some very specific multimodality combinations (PET/MR). In this paper, we propose to use the much more general notion of mutual information (MI) or relative entropy [8], =-=[22]-=- to describe the dispersive behavior of the 2-D histogram. MI is a basic concept from information theory, measuring the statistical dependence between two random variables or the amount of information... |

58 |
et al., “Comparison and evaluation of retrospective intermodality brain image registration techniques
- West
- 1997
(Show Context)
Citation Context ...ndence between them. Additional experiments on nine other datasets similar to dataset C within the Retrospective Registration Evaluation Project by Fitzpatrick et al. [10] have verified these results =-=[29]-=-, [14]. Moreover, Section VI-C demonstrated the robustness of the method with respect to partial overlap, while it was shown in Section VI-D that large image degradations, such as noise and intensity ... |

53 |
3D multimodality medical image registration using feature space clustering
- Collignon, Vandermeulen, et al.
- 1995
(Show Context)
Citation Context ... 10 (see the parameters in Table III). If both marginal distributions and can be considered to be independent of the registration parameters , the MI criterion reduces to minimizing the joint entropy =-=[6]-=-. If either or is independent of , which is the case if one of the images is always completely contained in the other, the MI criterion reduces to minimizing the conditional entropy or . However, if b... |

48 |
Voxel similarity measures for automated image registration
- Hill, Hawkes
- 1994
(Show Context)
Citation Context ...ltiresolution optimization strategy is therefore required, which is not necessarily a disadvantage, as it can be computationally attractive. In the approach of Woods et al. [30] and Hill et al. [12], =-=[13]-=-, misregistration is measured by the dispersion of the two-dimensional (2-D) histogram of the image intensities of corresponding voxel pairs, which is assumed to be minimal in the registered position.... |

35 |
Multiresolution Voxel Similarity Measures for MR-PET Registration
- Studholme, Hill, et al.
- 1995
(Show Context)
Citation Context ...ON MEDICAL IMAGING, VOL. 16, NO. 2, APRIL 1997 propose are largely heuristic. Hill’s criterion requires segmentation of the images or delineation of specific histogram regions to make the method work =-=[20]-=-, while Woods’ criterion is based on additional assumptions concerning the relationship between the grey-values in the different modalities, which reduces its applicability to some very specific multi... |

24 | An error metric for binary images - BADDELEY - 1992 |

24 |
Automatic registration of CT and MR brain images using correlation of geometrical features
- Elsen, Maintz, et al.
- 1995
(Show Context)
Citation Context ...case of intermodality registration. Crosscorrelation of feature images derived from the original image data has been applied to CT/MR matching using geometrical features such as edges [15] and ridges =-=[24]-=- or using especially designed intensity transformations [25]. But feature extraction may introduce new geometrical errors and requires extra calculation time. Furthermore, correlation of sparse featur... |

16 | Statistical Pattern Recognition - Chen - 1978 |

16 |
Automated multimodality medical image registration using information theCopyright © 2012 SciRes
- Collignon, Maes, et al.
- 1995
(Show Context)
Citation Context ...terion is very general and powerful and can be applied automatically without prior segmentation on a large variety of applications. This paper expands on the ideas first presented by Collignon et al. =-=[7]-=-. Related work in this area includes the work by Viola and Wells et al. [27], [28] and by Studholme et al. [21]. The theoretical concept of MI is presented in Section II, while the implementation of t... |

16 |
J.R.: Grey value correlation techniques used for automatic matching of
- Elsen, Pol, et al.
(Show Context)
Citation Context ...ure images derived from the original image data has been applied to CT/MR matching using geometrical features such as edges [15] and ridges [24] or using especially designed intensity transformations =-=[25]-=-. But feature extraction may introduce new geometrical errors and requires extra calculation time. Furthermore, correlation of sparse features like edges and ridges may have a very peaked optimum at t... |

16 |
Roucayrol, "A New Class of Similarity Measures for Robust Image Registration
- Venot, Lebruchec, et al.
- 1984
(Show Context)
Citation Context ...been proposed that optimize some global measure of the absolute difference between image intensities of corresponding voxels within overlapping parts or in a region of interest (ROI) [5], [11], [19], =-=[26]-=-. These criteria all rely on the assumption that the intensities of the two images are linearly correlated, which is generally not satisfied in the case of intermodality registration. Crosscorrelation... |

11 |
A strategy for automated multimodality image registration incorporating anatomical knowledge and imager characteristics
- Hill, Hawkes, et al.
- 1993
(Show Context)
Citation Context .... A multiresolution optimization strategy is therefore required, which is not necessarily a disadvantage, as it can be computationally attractive. In the approach of Woods et al. [30] and Hill et al. =-=[12]-=-, [13], misregistration is measured by the dispersion of the two-dimensional (2-D) histogram of the image intensities of corresponding voxel pairs, which is assumed to be minimal in the registered pos... |

8 | Comparison of feature-based matching of CT and MR brain images
- Maintz, Elsen, et al.
- 1995
(Show Context)
Citation Context ...atisfied in the case of intermodality registration. Crosscorrelation of feature images derived from the original image data has been applied to CT/MR matching using geometrical features such as edges =-=[15]-=- and ridges [24] or using especially designed intensity transformations [25]. But feature extraction may introduce new geometrical errors and requires extra calculation time. Furthermore, correlation ... |

7 |
Bizais Y.: Registration of Multimodality Medical Images Using Region Overlap Criterion
- Gerlot-Chiron
- 1992
(Show Context)
Citation Context ...ethods have been proposed that optimize some global measure of the absolute difference between image intensities of corresponding voxels within overlapping parts or in a region of interest (ROI) [5], =-=[11]-=-, [19], [26]. These criteria all rely on the assumption that the intensities of the two images are linearly correlated, which is generally not satisfied in the case of intermodality registration. Cros... |

7 |
Pseudocorrelation: A fast, robust, absolute, grey-level image alignment algorithm
- Radcliffe, Rajapakshe, et al.
- 1994
(Show Context)
Citation Context ... have been proposed that optimize some global measure of the absolute difference between image intensities of corresponding voxels within overlapping parts or in a region of interest (ROI) [5], [11], =-=[19]-=-, [26]. These criteria all rely on the assumption that the intensities of the two images are linearly correlated, which is generally not satisfied in the case of intermodality registration. Crosscorre... |

6 |
On the problem of geometric distortion in magnetic resonance images for stereotactic neurosurgery
- Michiels, Bosmans, et al.
- 1994
(Show Context)
Citation Context ...of the inhomogeneity. 3) Geometric Distortion: Geometric distortions and were applied to the original MR image according to a slice-by-slice planar quadratic model of the magnetic field inhomogeneity =-=[17]-=- (16) (17) (18) with the image coordinates of the center of each image plane and a scale parameter. Fig. 9(d) shows traces of the registration criterion for various amounts of distortion . As expected... |

4 |
A Review of medical image registration, Interactive Image-Guided Surgery
- Maurer, Fitzpatrick
- 1993
(Show Context)
Citation Context ...r Item Identifier S 0278-0062(97)02397-5. 0278–0062/97$10.00 © 1997 IEEE obtained from positron emission tomography (PET) images, etc. The bulk of registration algorithms in medical imaging (see [3], =-=[16]-=-, and [23] for an overview) can be classified as being either frame based, point landmark based, surface based, or voxel based. Stereotactic frame-based registration is very accurate, but inconvenient... |

4 | image matching—A review with classification - A - 1993 |

3 |
Coincident bit counting—A new criterion for image registration
- Chiang, Sullivan
- 1993
(Show Context)
Citation Context ...VSB methods have been proposed that optimize some global measure of the absolute difference between image intensities of corresponding voxels within overlapping parts or in a region of interest (ROI) =-=[5]-=-, [11], [19], [26]. These criteria all rely on the assumption that the intensities of the two images are linearly correlated, which is generally not satisfied in the case of intermodality registration... |

3 |
Automated 3D Registration of Truncated
- Studholme, Hill, et al.
- 1995
(Show Context)
Citation Context ...iety of applications. This paper expands on the ideas first presented by Collignon et al. [7]. Related work in this area includes the work by Viola and Wells et al. [27], [28] and by Studholme et al. =-=[21]-=-. The theoretical concept of MI is presented in Section II, while the implementation of the registration algorithm is described in Section III. In Sections IV, V, and VI we evaluate the accuracy and t... |

2 |
Entropy correlation coefficient, a measure of statistical dependence for categorized data
- Astola, Virtanen
- 1982
(Show Context)
Citation Context ...a family of measures of statistical dependence or information redundancy (see Appendix C). We have experimented with , which can be shown to be a metric [8], and , the entropy correlation coefficient =-=[1]-=-. In some cases these measures performed better than the original MI criterion, but we could not establish a clear preference for either of these. Furthermore, the use of MI for multimodality image re... |

1 |
Evaluation of retrospective image registration,” Vanderbilt Univ
- Fitzpatrick
- 1994
(Show Context)
Citation Context ...ptimizing the parameters in the order ( , ). The results are summarized in Table III by the parameters of the transformation that 1 Data provided by van den Elsen [25]. 2 Data provided by Fitzpatrick =-=[10]-=-. Fig. 4. The bounding box of the central eighth of the floating image defines eight points near the brain surface at which the difference between different registration transforms is evaluated. takes... |