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

By Mattias Paul Heinrich, Mark Jenkinson, M. Bhushan, Tahreema N. Matin, F. Gleeson, M. Brady, Julia Anne Schnabel. Medical Image Anal., 2012

745 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Radiology, Nuclear Medicine and Imaging

View Paper: https://www.semanticscholar.org/paper/4f41e4f7dc303826c0cfa4181919f032730c4c6d

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Radiology, Nuclear Medicine and Imaging

Citations in the corpus

Slice-to-volume medical image registration: a survey (2017, methodology): “Another robust similarity measure, the modality independent neighborhood descriptor (MIND) was proposed by Heinrich et al. (2012) for multimodal rigid and deformable registration.”
Generative Adversarial Network in Medical Imaging: A Review (2018, background): “Hiasa et al. (2018) further incorporated gradient consistency loss in the training to improve accuracy at the boundaries.”
Medical Image Synthesis with Deep Convolutional Adversarial Networks (2018, background): “1) The brain dataset was acquired from 16 [49].”
Deep learning in medical image registration: a review (2019, methodology): “Intensitybased image appearance loss includes SSD, MSE, MAE, MI, MIND, SSIM, CC and its variants.”
MRF-Based Deformable Registration and Ventilation Estimation of Lung CT (2013, background): “Locally contrast invariant metrics, such as local cross-correlation [12], local phase [13], or modality independent neighborhood descriptors [14] may be used instead.”
MUSE: MUlti-atlas region Segmentation utilizing Ensembles of registration algorithms and parameters, and locally optimal atlas selection (2015, neutral): “…(Wang et al., 2012) (PICSL), a levelset-based label fusion and correction method (SBIA-LevelSet), a new label fusion method that uses a modality independent neighbourhood descriptor (Heinrich et al., 2012) (deedsMIND), and a label propagation method using random forests (Zikic et al., 2014)(MSRC).”
Fast and Robust Matching for Multimodal Remote Sensing Image Registration (2019, methodology): “More-over, some state-of-the-art similarity measures, such as MI, GMI [43], HOG ncc , HOPC ncc [34], and modality independent neighborhood descriptor (MIND) [37], are used for comparisons.”
Understanding and Interpreting Machine Learning in Medical Image Computing Applications (2018, methodology): “While some methods aim to find prototypes for classes, we only use methods that explain the CNN decision for a specific sample [9,10,12] (see methodological details below).”
Longitudinal multiple sclerosis lesion segmentation: Resource and challenge (2017, positive): “The MALF estimation of spatial tissue priors uses the rigid and afﬁne components of ANTs (Avants et al., 2008) and the non-linear framework of MIND (Heinrich et al., 2012, 2013).”
Deformable Medical Image Registration: A Survey (2013, background): “[373] presented a new descriptor for multi-modal registration.”
Learning deep similarity metric for 3D MR–TRUS image registration (2018, methodology): “8, 8.68] Deep Metric (mp) BFGS 14.27 0.61 [13.11, 15.25] Deep Metric (mp) Powell 11.05 5.32 [2.11, 24.09] Deep Metric (mp) DINO 3.94 1.47 [1.35, 9.37] dierences of the MIND descriptors of each image [7] using DINO. In order to demonstrate DINO’s robustness to poor initialization, initializations of ”