Markov random field segmentation of brain MR images

Held, Karsten; Kops, Elena Rota; Krause, Bernd J.; Wells, Iii. William M.; Kikinis, Ron; Müller‐Gärtner, H.‐W.

doi:10.1109/42.650883

Cited by 425 publications

(242 citation statements)

References 28 publications

Supporting

Mentioning

240

Contrasting

Unclassified

Order By: Relevance

“…Originally proposed in a brain MRI context by Wells III et al (1996), and further improved by many others (e.g. : Ashburner, 2000;Ashburner and Friston, 2000;Guillemaud and Brady, 1997;Held et al, 1997;Pohl et al, 2002;Schroeter et al, 1998;Van Leemput et al, 1999a,b), these methods interleave intensity non-uniformity field estimation (correction) and classification, in an iterative fashion.…”

Section: Introductionmentioning

confidence: 99%

“…Several approaches have been proposed to address this limitation of intensity-based classification. For example, post-classification morphological operations or contextual classifiers (Choi et al, 1991;Held et al, 1997;Rajapakse et al, 1997;Udupa and Samarasekera, 1996;Yan and Karp, 1995). Moreover, a number of researchers have proposed continuous classifiers which attempt to estimate the mixing proportions of several tissues in a voxel (i.e.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A fully automatic and robust brain MRI tissue classification method

Cocosco

Zijdenbos

Evans

2003

Medical Image Analysis

297

184

View full text Add to dashboard Cite

A novel, fully automatic, adaptive, robust procedure for brain tissue classification from 3D magnetic resonance head images (MRI) is described in this paper. The procedure is adaptive in that it customizes a training set, by using a "pruning" strategy, such that the classification is robust against anatomical variability and pathology. Starting from a set of samples generated from prior tissue probability maps (a "model") in a standard, brain-based coordinate system ("stereotaxic space"), the method first reduces the fraction of incorrectly labeled samples in this set by using a minimum spanning tree graph-theoretic approach. Then, the corrected set of samples is used by a supervised kNN classifier for classifying the entire 3D image. The classification procedure is robust against variability in the image quality through a non-parametric implementation: no assumptions are made about the tissue intensity distributions. The performance of this brain tissue classification procedure is demonstrated through quantitative and qualitative validation experiments on both simulated MRI data (10 subjects) and real MRI data (43 subjects). A significant improvement in output quality was observed on subjects who exhibit morphological deviations from the model due to aging and pathology.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

A fully automatic and robust brain MRI tissue classification method

Cocosco

Zijdenbos

Evans

2003

Medical Image Analysis

297

184

View full text Add to dashboard Cite

show abstract

“…However, such MRF models can over regularize the fine structured borders, e.g. the interface between gray and white matter; therefore, it is often necessary to impose additional, heuristic constraints [6,3]. Active contour models [9,10] have also been used to impose smoothness constraints for segmentation.…”

Section: Related Workmentioning

confidence: 99%

“…Others have used non-linear diffusion for image denoising as a pre-processing step [4]. However, for probabilistic algorithms, it is intuitive to incorporate spatial smoothness constraints directly into the segmentation process via Markov random field (MRF) models [5,6,3,7,8]. These methods modify single-pixel tissue-probabilities with energies defined on local configurations of segmentation labels.…”

Section: Related Workmentioning

confidence: 99%

MRI Tissue Classification with Neighborhood Statistics: A Nonparametric, Entropy-Minimizing Approach

Taşdizen

Awate

Whitaker

et al. 2005

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. We introduce a novel approach for magnetic resonance image (MRI) brain tissue classification by learning image neighborhood statistics from noisy input data using nonparametric density estimation. The method models images as random fields and relies on minimizing an entropy-based metric defined on high dimensional probability density functions. Combined with an atlas-based initialization, it is completely automatic. Experiments on real and simulated data demonstrate the advantages of the method in comparison to other approaches.

show abstract

“…By means of Markov Random field in [15] and [21] are described fully automatic 3D-segmentation techniques especially designed for brain MRI images. This techniques captures three main spatial features of MRI images: non-parametric distribution of tissue intensities, neighborhood correlations and signal inhomogeneities.…”

Section: Introductionmentioning

confidence: 99%