New variants of a method of MRI scale standardization

By László G. Nyúl, Jayaram Koteshwar Udupa, Xuan Zhang. IEEE Transactions on Medical Imaging, 2000

1,006 citations recorded by the source corpus.

Subfields: Radiology, Nuclear Medicine and Imaging

View Paper: https://www.semanticscholar.org/paper/e7e245cf1e9508b82138f2b406f14cc7ca7884ea

How researchers see this paper

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

In Radiology, Nuclear Medicine and Imaging

Citations in the corpus

On the Compactness, Efficiency, and Representation of 3D Convolutional Networks: Brain Parcellation as a Pretext Task (2017, methodology): “At image-level, we adopted the histogram-based scale standardisation method [17] to normalised the intensity histograms.”
Brain Tumor Segmentation Using Convolutional Neural Networks in MRI Images (2016, methodology): “In fact, it can vary even if the image of the same patient is acquired in the same scanner in different time points, or in the presence of a pathology [7], [38].”
Evaluation of prostate segmentation algorithms for MRI: the PROMISE12 challenge (2013, background): “Additionally, signal intensity values are not standardized (Nyúl and Udupa, 1999; Nyúl et al., 2000).”
Harmonization of cortical thickness measurements across scanners and sites (2017, methodology): “For instance, for conventional MRI studies, intensity normalization techniques have been developed to make the image intensities comparable across studies, including histogram matching (Nyúl et al., 2000), WhiteStripe (Shinohara et al., 2014) and RAVEL (Fortin et al., 2016).”
A deep learning model integrating FCNNs and CRFs for brain tumor segmentation (2017, methodology): “The proposed brain tumor segmentation method consists of three main steps: pre-processing, segmentation using a deep learning model with integrated FCNNs and CRF-RNN, and post-processing.”
Stochastic Filter Groups for Multi-Task CNNs: Learning Specialist and Generalist Convolution Kernels (2019, methodology): “The input MRI scans were ﬁrst normalised using histogram normalisation based on the 1 st and 99 th percentile [40].”
Multifractal Texture Estimation for Detection and Segmentation of Brain Tumors (2013, methodology): “2) Correct bias and intensity inhomogeneity across all the slices of all the patients for each MRI modality using two-step normalization method [39].”
Fast and robust multi-atlas segmentation of brain magnetic resonance images (2010, methodology): “One strategy is to align the intensity histograms of images; Nyúl et al. (2000) defined several landmarks (percentiles and modes) from histograms and matched the landmarks.”
Probabilistic segmentation of white matter lesions in MR imaging (2004, methodology): “To correct for MR inhomogeneities, a method was used that resulted in similar gray values of major anatomical structures in different patients per image type (Nyúl and Udupa, 1999; Nyúl et al., 2000).”
The Multimodal Brain Tumor Image Segmentation Benchmark (BRATS) (2015, methodology): “Image pre-processing involves bias ﬁeld correction using N3 [117], intra-subject multispectral volume registration [118], non-uniformity correction [119], and intensity normalization [104] The algorithm consists of two stages.”
Stain Specific Standardization of Whole-Slide Histopathological Images (2016, methodology): “The normalized median intensity (NMI) measure [15], [28] was then chosen to evaluate color constancy of the nuclei.”
An adaptive spatial fuzzy clustering algorithm for 3-D MR image segmentation (2003, methodology): “Data preprocessing using a tissue intensity calibration procedure as proposed in [6] prior to the segmentation process would ensure a more accurate segmentation.”
Review of automatic segmentation methods of multiple sclerosis white matter lesions on conventional magnetic resonance imaging (2013, methodology): “…cMRI is not quantitative, differences in brightness and contrast that are not related to any biological change can be found even between two images from the same patient, especially when they come from different scanners, and can be reduced by intensity normalization algorithms(Nyul et al., 2000).”
Ensembles of Multiple Models and Architectures for Robust Brain Tumour Segmentation (2017, methodology): “3) Bias field correction, followed by piece-wise linear normalisation [29], followed by (1).”