Multi-modality image registration by maximization of mutual information

Maes, Frederik; Collignon, André; Vandermeulen, Dirk; Marchal, Guy; Suetens, Paul

doi:10.1109/mmbia.1996.534053

Cited by 259 publications

(88 citation statements)

References 12 publications

(8 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The objective function, called stack entropy in [6] and [7], measures the compactness of the intensity distribution for a certain voxel location across different subjects. Similar to the popular entropy based metrics used in pairwise registration (e.g., mutual information [8][9]), the current formulation of stack entropy considers image intensity as the only feature, and discards local contextual information that can be provided by the voxel neighborhood. Moreover, each voxel contributes equally to the metric, regardless of its anatomical properties.…”

Section: Introductionmentioning

confidence: 99%

Attribute Vector Guided Groupwise Registration

Yap

Shen

2009

Medical Image Computing and Computer-Assisted Intervention – MICCAI 2009

View full text Add to dashboard Cite

Abstract. Groupwise registration has been recently introduced for simultaneous registration of a group of images with the goal of constructing an unbiased atlas. To this end, direct application of information-theoretic entropy measures on image intensity has achieved various successes. However, simplistic voxelwise utilization of image intensity often neglects important contextual information, which can be provided by more comprehensive geometric and statistical features. In this paper, we employ attribute vectors, instead of image intensities, to guide groupwise registration. In particular, for each voxel, the attribute vector is computed from its multiple-scale neighborhoods to capture geometric information at different scales. Moreover, the probability density function (PDF) of each attribute in the vector is then estimated from the local neighborhood, providing a statistical summary of the underlying anatomical structure. For the purpose of registration, Jensen-Shannon (JS) divergence is used to measure the PDF dissimilarity of each attribute at corresponding locations of different individual images. By minimizing the overall JS divergence in the whole image space and estimating the deformation field of each image simultaneously, we can eventually register all images and build an unbiased atlas. Experimental results indicate that our method yields better registration quality, compared with a popular groupwise registration method.

show abstract

Section: Introductionmentioning

confidence: 99%

Attribute Vector Guided Groupwise Registration

Yap

Shen

2009

Medical Image Computing and Computer-Assisted Intervention – MICCAI 2009

View full text Add to dashboard Cite

show abstract

“…At each layer of the cascade, the classification is done through the estimation of mutual information [1] between the 3D model and the image. Mutual information has shown to be a robust metric for matching involving multimodal data [5,3].…”

Section: Mutual Information Computationmentioning

confidence: 99%

Optimal Cascade Construction for Detection using 3D Models

Pong

Cham

2006

18th International Conference on Pattern Recognition (ICPR'06)

View full text Add to dashboard Cite

show abstract

“…However, mutual information is only sensitive to gray-level distribution and does not include spatial information. Therefore, it fails in some multimodality co-registration methods [13] . For example, the gray levels of the PA and LS images characterize different functional information rather than spatial information.…”

mentioning

confidence: 99%

“…Mutual information is used as a feature matching (measures) criterion [13] , which measures the statistical dependence between two images A and B. It could be described by the following equations with entropy:…”

mentioning

confidence: 99%

Co-registration method for photoacoustic imaging and laser speckle imaging

Gao¹,

Deng²,

Tong³

et al. 2012

中国光学快报

View full text Add to dashboard Cite

Combining photoacoustic (PA) imaging with laser speckle (LS) imaging (LSI) can simultaneously determine total hemoglobin concentration (HbT), hemoglobin oxygen saturation (SO2), and blood flow rates. Thus, the co-registration of PA and LS images is important in physiological studies and pathological diagnosis. This letter presents a co-registration algorithm combining mutual information with the maximum betweenclass variance segmentation method (Otsu method). The mutual information and Otsu method are used to provide the registration measure criterion and image feature recognition, respectively. The evaluation results show that the registration function possesses a single maximum peak and high smoothness across the global co-registration district, indicating a robust behavior. Moreover, this method has good registration accuracy, and the fusion result simultaneously visualizes the separate functional information of two kinds of images.OCIS codes: 110.6150, 110.5120, 100.2000codes: 110.6150, 110.5120, 100. , 170.2655 Photoacoustic (PA) imaging (PAI) is a rapidly emerging, noninvasive imaging method. It visualizes the structural and functional characteristics of biological tissues by ultrasonically detecting their optical absorption contrast of these tissues in vivo, breaking highly optical scattering limitations and achieving super-depth high-resolution optical imaging [1−3] . Thus, the method has been widely applied in monitoring angiogenesis, melanoma, hemoglobin oxygen saturation (SO2) [4] , and total hemoglobin concentration (HbT) [1,2] . Laser speckle (LS) imaging (LSI), a no-scanning, noninvasive imaging tool, could achieve highly spatial (tens of microns) and temporal (millisecond) resolutions for imaging biological tissues in vivo [5,6] . The temporal and spatial statistics of speckle pattern intensity fluctuations contain the movement information of the object observed. Local velocity distributions should be measured properly by analyzing local speckle contrast variations [6] . Thus, LSI is always used for monitoring capillary blood flow rates in the skin and cerebral blood flow rates [6−8] . Combining PA and LS images could simultaneously provide HbT, SO2, and capillary blood flow information. The multimodality image coregistration technology is the foundation of fusion, which is significant for understanding the normal and pathophysiological conditions of neurovascular, metabolic, and hemodynamic interactions [9,10] . These multimodality coregistration methods can be classified as either extrinsic, intrinsic (including landmarks, segmentation-and mutual-information-based registration), or non-imagebased registration methods [11,12] . The mutual information method does not need assumptions and limiting constraints for the image content. Thus, it is a very general and powerful criterion for multimodality image coregistration. However, mutual information is only sensitive to gray-level distribution and does not include spatial information. Therefore, it fails in some multimodality co-registration method...

show abstract

Multi-modality image registration by maximization of mutual information

Cited by 259 publications

References 12 publications

Attribute Vector Guided Groupwise Registration

Attribute Vector Guided Groupwise Registration

Optimal Cascade Construction for Detection using 3D Models

Co-registration method for photoacoustic imaging and laser speckle imaging

Contact Info

Product

Resources

About