Segmentation of brain MR images through a hidden Markov random field model and the expectation-maximization algorithm
By Yongyue Zhang, M. Brady, Stephen M. Smith. IEEE Transactions on Medical Imaging, 2001
7,371 citations recorded by the source corpus.
The finite mixture (FM) model is the most commonly used model for statistical segmentation of brain magnetic resonance (MR) images because of its simple mathematical form and the piecewise constant nature of ideal brain MR images. However, being a histogram-based model, the FM has an intrinsic limitation-no spatial information is taken into account. This causes the FM model to work only on well-defined images with low levels of noise; unfortunately, this is often not the the case due to artifacts such as partial volume effect and bias field distortion. Under these conditions, FM model-based methods produce unreliable results. Here, the authors propose a novel hidden Markov random field (HMRF) model, which is a stochastic process generated by a MRF whose state sequence cannot be observed directly but which can be indirectly estimated through observations. Mathematically, it can be shown that the FM model is a degenerate version of the HMRF model. The advantage of the HMRF model derives from the way in which the spatial information is encoded through the mutual influences of neighboring sites. Although MRF modeling has been employed in MR image segmentation by other researchers, most reported methods are limited to using MRF as a general prior in an FM model-based approach. To fit the HMRF model, an EM algorithm is used. The authors show that by incorporating both the HMRF model and the EM algorithm into a HMRF-EM framework, an accurate and robust segmentation can be achieved. More importantly, the HMRF-EM framework can easily be combined with other techniques. As an example, the authors show how the bias field correction algorithm of Guillemaud and Brady (1997) can be incorporated into this framework to achieve a three-dimensional fully automated approach for brain MR image segmentation.
Subfields: Artificial Intelligence, Media Technology, Signal Processing
View Paper: https://www.semanticscholar.org/paper/7d5aedecdfc4e8f83638bac47eb7cf2f860ec51c
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Dinggang Shen
- Chitta Baral
- Stephen M. Smith
- Daoqiang Zhang
- Christian Gaser
- Christian F. Beckmann
- Heung-Il Suk
- Seong-Whan Lee
In Media Technology
- Dinggang Shen
- Stephen M. Smith
- Alan C. Evans
- Christian F. Beckmann
- Xiahai Zhuang
- Bruce Fischl
- Daoqiang Zhang
- Matthew F. Glasser
In Signal Processing
- Stephen M. Smith
- Christian F. Beckmann
- Dinggang Shen
- Vince D. Calhoun
- Matthew F. Glasser
- Mark William Woolrich
- Mark Jenkinson
- Xiahai Zhuang
Citations in the corpus
- Local-Global Parcellation of the Human Cerebral Cortex From Intrinsic Functional Connectivity MRI (2017, methodology): “For example, the FSL FAST (Zhang et al., 2001) MRF model balances a global similarity objective that encourages voxels with similar T1 intensities to be labeled the same structure with a local smoothness objective that encourages neighboring voxels to have”
- MRI Segmentation of the Human Brain: Challenges, Methods, and Applications (2015, methodology): “Several methods have been proposed to address the problem of PVE in MRI segmentation for both adult and neonatal brains [3, 42, 103, 121, 122].”
- Hierarchical feature representation and multimodal fusion with deep learning for AD/MCI diagnosis (2014, positive): “Then, FAST in FSL package(7) (Zhang et al., 2001) was used to segment the structural MR images into three tissue types of Gray Matter (GM), White Matter (WM), and CerebroSpinal Fluid (CSF).”
- Accurate and Robust Brain Image Alignment using Boundary-based Registration (2009, methodology): “For this we used the FMRIB's Automated Segmentation Tool (FAST; Zhang et al., 2001) distributed with FSL.”
- Cortical Thickness or Grey Matter Volume? The Importance of Selecting the Phenotype for Imaging Genetics Studies (2009, neutral): “Thus, the assignment of a voxel to a class depends on its immediate spatial neighbours in the threedimensional space, and the problem is treated by obtaining the maximum likelihood estimate of the parameters using an expectation-maximization (EM) algorithm (Zhang et al., 2001).”
- Brain anatomy differences in childhood stuttering (2008, methodology): “Individual skull-stripped images were segmented into gray matter, white matter, and cerebrospinal fluid using FSL's automated segmentation algorithm (Zhang et al., 2001) and normalized into the pediatric template space using parameters derived from the initial registration of raw images to the…”
- A survey of GPU-based medical image computing techniques. (2012, background): “In addition, Markov random field (MRF) (149), Expectation-Maximization (EM) algorithm (150), and Hidden Markov Chains (HMC) (151) have been used to improve image segmentation performance.”
- Machine Learning and Radiology (2012, methodology): “As introduced in Section 2, Markov random fields (MRFs) have wide applications in medical image segmentation (Held et al., 1997; Towhidkhah et al., 2008; Zhang et al., 2001).”
- Segmentation of X‐ray computed tomography images of porous materials: A crucial step for characterization and quantitative analysis of pore structures (2009, methodology): “More recent developments provide options to use MRF in unsupervised, data-driven manner [e.g., Zhang et al., 2001; Peng et al., 2005; Xia et al., 2006], potentially reducing operator bias.”
- Age-related differences in white matter microstructure: Region-specific patterns of diffusivity (2010, methodology): “We estimated WMH volume using a semi-automated procedure based on FMRIB's Automated Segmentation Tool (FAST 4.1, Zhang et al. 2001).”
- Particle swarm optimization method for image clustering (2005, methodology): “Experiments conducted by Zhang et al [50] [49] and Hamerly el al [17] [18] showed that KHM outperforms K-means, FCM (according to Hamerly [17]) and EM.”
- Multimodal Classification of Alzheimer’s Disease and Mild Cognitive Impairment (2011, methodology): “After removal of cerebellum, FAST in the FSL package (Zhang et al., 2001) is used to segment structural MR images into three different tissues: gray matter (GM), white matter (WM), and cerebrospinal fluid (CSF).”
- Harmonization of multi-site diffusion tensor imaging data (2017, methodology): “A 3-tissue class T1-w segmentation was performed using FSL's FAST tool (Zhang et al., 2001) (v4.”
- Impact of In-Scanner Head Motion on Multiple Measures of Functional Connectivity: Relevance for Studies of Neurodevelopment in Youth (2012, methodology): “CSF and white matter timecourses were extracted from subject-specific tissue-segments of the T1-weighted image created using FAST (Zhang et al., 2001).”
- Measurement of atrophy in multiple sclerosis: pathological basis, methodological aspects and clinical relevance. (2002, methodology): “MRF (MRI random ®eld model)-based segmentation (Zhang et al., 2001) is used to segment the brain image into different tissue types, including partial volume modelling, giving a total normalized brain volume as output.”
- Non-invasive mapping of connections between human thalamus and cortex using diffusion imaging (2003, methodology): “We performed probabilistic tissue-type segmentation and partial volume estimation on the T1-weighted imag”
- Anatomically-constrained tractography: Improved diffusion MRI streamlines tractography through effective use of anatomical information (2012, methodology): “Intensity-based classifications such as FAST do however have difficulty in achieving meaningful segmentation of sub-cortical grey matter structures.”