MIND: Modality independent neighbourhood descriptor for multi-modal deformable registration

Heinrich, Mattias P.; Jenkinson, Mark; Bhushan, Manav; Matin, Tahreema; Gleeson, Fergus; Brady, Sir Michael; Schnabel, Julia A.

doi:10.1016/j.media.2012.05.008

Cited by 532 publications

(390 citation statements)

References 55 publications

(67 reference statements)

Supporting

Mentioning

386

Contrasting

Unclassified

Order By: Relevance

“…std T Hub is calculated with the same settings except that one resolution is used. The MIND feature is calculated using a [3 × 3 × 3] region as suggested by [8] and also compared with a sparse patch including 82 voxels inside a [7 × 7 × 3] box, which is physically more isotropic for our data.…”

Section: Experiments and Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Accuracy Estimation for Medical Image Registration Using Regression Forests

Sokooti

Saygılı

Glocker

et al. 2016

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. This paper reports a new automatic algorithm to estimate the misregistration in a quantitative manner. A random regression forest is constructed, predicting the local registration error. The forest is built using local and modality independent features related to the registration precision, the transformation model and intensity-based similarity after registration. The forest is trained and tested using manually annotated corresponding points between pairs of chest CT scans. The results show that the mean absolute error of regression is 0.72 ± 0.96 mm and the accuracy of classification in three classes (correct, poor and wrong registration) is 93.4%, comparing favorably to a competing method. In conclusion, a method was proposed that for the first time shows the feasibility of automatic registration assessment by means of regression, and promising results were obtained.

show abstract

Section: Experiments and Resultsmentioning

confidence: 99%

“…Difference of MIND: Heinrich et al [8] introduced the Modality Independent Neighborhood Descriptor (MIND) to register multimodal images by comparing similarities between same patches in the fixed and moving image. The output of this local self-similarity has n features for each voxel, where n is the size of the search region.…”

Section: Features and Poolingmentioning

confidence: 99%

Accuracy Estimation for Medical Image Registration Using Regression Forests

Sokooti

Saygılı

Glocker

et al. 2016

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

“…Some studies have focused on improving registration accuracy by considering the use of additional image gradient information (Pluim et al, 2000a,b), neighbor pixel information (Heinrich et al, 2012b;Kybic and Vnucko, 2012;Rueckert et al, 2000), textural information (Heinrich et al, 2012a), or different approaches to weighted MI (Park et al, 2010;Rodriguez-Carranza and Loew, 1999;Van Dalen et al, 2004). One approach to weighted MI is the regularization of MI with the use of weights based on overlaps (Rodriguez-Carranza and Loew, 1999) without including spatial information.…”

Section: II State Of the Artmentioning

confidence: 99%

Organ-focused mutual information for nonrigid multimodal registration of liver CT and Gd–EOB–DTPA-enhanced MRI

Fernandez-de-Manuel

Wollny

Kybic

et al. 2014

Medical Image Analysis

View full text Add to dashboard Cite

Accurate detection of liver lesions is of great importance in hepatic surgery planning. Recent studies have shown that the detection rate of liver lesions is significantly higher in gadoxetic acid-enhanced magnetic resonance imaging (Gd-EOB-DTPA-enhanced MRI) than in contrast-enhanced portal-phase computed tomography (CT); however, the latter remains essential because of its high specificity, good performance in estimating liver volumes and better vessel visibility. To characterize liver lesions using both the above image modalities, we propose a multimodal nonrigid registration framework using organ-focused mutual information (OF-MI). This proposal tries to improve mutual information (MI) based registration by adding spatial information, benefiting from the availability of expert liver segmentation in clinical protocols. The incorporation of an additional information channel containing liver segmentation information was studied. A dataset of real clinical images and simulated images was used in the validation process. A Gd-EOB-DTPA-enhanced MRI simulation framework is presented. To evaluate results, warping index errors were calculated for the simulated data, and landmark-based and surface-based errors were calculated for the real data. An improvement of the registration accuracy for OF-MI as compared with MI was found for both simulated and real datasets. Statistical significance of the difference was tested and confirmed in the simulated dataset (p < 0.01).

show abstract

“…spatial, information on the local structure and cannot cope with large space-variant signal and contrast differences. Although several methods introduce local information into a global objective function [15,16], it has been noticed [17] that finding an accurate correspondence remains difficult, especially due to the many local minima that generally accompany most non-rigid deformation models.…”

Section: Introductionmentioning

confidence: 99%

“…Recently, a method employing a local modality independent neighborhood descriptor (MIND) was proposed [17]. MIND was reported to be a "distinctive" descriptor, which was claimed to be important for registering images with many degrees of freedom [17].…”

Section: Introductionmentioning

confidence: 99%

A hybrid optimization strategy for registering images with large local deformations and intensity variations

Vliet

Stoker

et al. 2017

Int J CARS

View full text Add to dashboard Cite

Purpose To develop a method for intra-patient registration of pre-and post-contrast abdominal MR images with large local deformations and large intensity variations. Method A hybrid method is proposed to deal with this problem. It consists of two coupled techniques: (1) descriptor matching (DM) at the original resolution using a discrete optimization strategy to avoid getting trapped in a local minimum; (2) continuous optimization to refine the registration outcome based on autocorrelation of local image structure (ALOST). Our method-called DM-ALOST-has become insensitive to the local uptake of contrast agent by exploiting the mean phase and the phase congruency extracted from the multi-scale monogenic signal. The method was extensively tested on abdominal MR data of 30 patients with Crohn's disease. Results DM-ALOST produced significantly larger mean Dice coefficients than two state-of-the-art methods ( p < 0.05). Conclusion Both qualitative and quantitative tests demonstrated improved registration using the proposed method compared to the state-of-the-art. The DM-ALOST method facilitates measurement of corresponding features from different abdominal MR images, which can aid to assess certain diseases, particularly Crohn's disease.

show abstract

MIND: Modality independent neighbourhood descriptor for multi-modal deformable registration

Cited by 532 publications

References 55 publications

Accuracy Estimation for Medical Image Registration Using Regression Forests

Accuracy Estimation for Medical Image Registration Using Regression Forests

Organ-focused mutual information for nonrigid multimodal registration of liver CT and Gd–EOB–DTPA-enhanced MRI

A hybrid optimization strategy for registering images with large local deformations and intensity variations

Contact Info

Product

Resources

About