Brain tissue classification of magnetic resonance images using partial volume modeling

Quantifying lumbar paraspinal intramuscular fat: Accuracy and reliability of automated thresholding models

Background

The reported level of lumbar paraspinal intramuscular fat (IMF) in people with low back pain (LBP) varies considerably across studies using conventional T1- and T2-weighted magnetic resonance imaging (MRI) sequences. This may be due to the different thresholding models employed to quantify IMF. In this study we investigated the accuracy and reliability of established (two-component) and novel (three-component) thresholding models to measure lumbar paraspinal IMF from T2-weighted MRI.

Methods

In this cross-sectional study, we included MRI scans from 30 people with LBP (50% female; mean (SD) age: 46.3 (15.0) years). Gaussian mixture modelling (GMM) and K-means clustering were used to quantify IMF bilaterally from the lumbar multifidus, erector spinae, and psoas major using two and three-component thresholding approaches (GMM2C; K-means2C; GMM3C; and K-means3C). Dixon fat-water MRI was used as the reference for IMF. Accuracy was measured using Bland-Altman analyses, and reliability was measured using ICC3,1. The mean absolute error between thresholding models was compared using repeated-measures ANOVA and post-hoc paired sample t-tests (α = 0.05).

Results

We found poor reliability for K-means2C (ICC3,1 ≤ 0.38), moderate to good reliability for K-means3C (ICC3,1 ≥ 0.68), moderate reliability for GMM2C (ICC3,1 ≥ 0.63) and good reliability for GMM3C (ICC3,1 ≥ 0.77). The GMM (p < .001) and three-component models (p < .001) had smaller mean absolute errors than K-means and two-component models, respectively. None of the investigated models adequately quantified IMF for psoas major (ICC3,1 ≤ 0.01).

Conclusions

The performance of automated thresholding models is strongly dependent on the choice of algorithms, number of components, and muscle assessed. Compared to Dixon MRI, the GMM performed better than K-means and three-component performed better than two-component models for quantifying lumbar multifidus and erector spinae IMF. None of the investigated models accurately quantified IMF for psoas major. Future research is needed to investigate the performance of thresholding models in a more heterogeneous clinical dataset and across different sites and vendors.

Brain tissue classification from stereoelectroencephalographic recordings

BACKGROUND

Stereoelectroencephalographic (SEEG) recordings can be performed before final resective surgery in some drug-resistant patients with focal epilepsies. For good SEEG signal interpretation, it is important to correctly identify the brain tissue in which each contact is inserted. Tissue classification is usually done with the coregistration of CT scan (with implanted SEEG electrodes) with preoperative MRI.

NEW METHOD

Brain tissue classification is done here directly from SEEG signals obtained at rest by a linear discriminant analysis (LDA) classifier using measured SEEG signals. The classification operates on features extracted from Bode plots obtained via non-parametric frequency domain transfer functions of adjacent contacts pairs. Classification results have been compared with classification from T1 MRI following the labelling procedure described in Deman et al. (2018), together with minor corrections by visual inspection by specialists.

RESULTS

With the data processed from 19 epileptic patients representing 1284 contact pairs, an accuracy of 72 ± 3% was obtained for homogeneous tissue separation. To our knowledge only one previous study conducted brain tissue classification using the power spectra of SEEG signals, and the distance between contacts on a shaft. The features proposed in our article performed better with the LDA classifier. However, the Bayesian classifier proposed in Greene et al. (2020) is more robust and could be used in a future study to enhance the classification performance.

CONCLUSIONS AND SIGNIFICANCE

Our findings suggest that careful analysis of the transfer function between adjacent contacts measuring resting activity via frequency domain identification, could allow improved interpretation of SEEG data and or their co-registration with subject's anatomy.

Relax and focus on brain tumor segmentation

In this paper, we present a Deep Convolutional Neural Networks (CNNs) for fully automatic brain tumor segmentation for both high- and low-grade gliomas in MRI images. Unlike normal tissues or organs that usually have a fixed location or shape, brain tumors with different grades have shown great variation in terms of the location, size, structure, and morphological appearance. Moreover, the severe data imbalance exists not only between the brain tumor and non-tumor tissues, but also among the different sub-regions inside brain tumor (e.g., enhancing tumor, necrotic, edema, and non-enhancing tumor). Therefore, we introduce a hybrid model to address the challenges in the multi-modality multi-class brain tumor segmentation task. First, we propose the dynamic focal Dice loss function that is able to focus more on the smaller tumor sub-regions with more complex structures during training, and the learning capacity of the model is dynamically distributed to each class independently based on its training performance in different training stages. Besides, to better recognize the overall structure of the brain tumor and the morphological relationship among different tumor sub-regions, we relax the boundary constraints for the inner tumor regions in coarse-to-fine fashion. Additionally, a symmetric attention branch is proposed to highlight the possible location of the brain tumor from the asymmetric features caused by growth and expansion of the abnormal tissues in the brain. Generally, to balance the learning capacity of the model between spatial details and high-level morphological features, the proposed model relaxes the constraints of the inner boundary and complex details and enforces more attention on the tumor shape, location, and the harder classes of the tumor sub-regions. The proposed model is validated on the publicly available brain tumor dataset from real patients, BRATS 2019. The experimental results reveal that our model improves the overall segmentation performance in comparison with the state-of-the-art methods, with major progress on the recognition of the tumor shape, the structural relationship of tumor sub-regions, and the segmentation of more challenging tumor sub-regions, e.g., the tumor core, and enhancing tumor.

Whole brain and deep gray matter structure segmentation: Quantitative comparison between MPRAGE and MP2RAGE sequences

Objective

T1-weighted MRI images are commonly used for volumetric assessment of brain structures. Magnetization prepared 2 rapid gradient echo (MP2RAGE) sequence offers superior gray (GM) and white matter (WM) contrast. This study aimed to quantitatively assess the agreement of whole brain tissue and deep GM (DGM) volumes obtained from MP2RAGE compared to the widely used MP-RAGE sequence.

Methods

Twenty-nine healthy participants were included in this study. All subjects underwent a 3T MRI scan acquiring high-resolution 3D MP-RAGE and MP2RAGE images. Twelve participants were re-scanned after one year. The whole brain, as well as DGM segmentation, was performed using CAT12, volBrain, and FSL-FAST automatic segmentation tools based on the acquired images. Finally, contrast-to-noise ratio between WM and GM (CNR_WG), the agreement between the obtained tissue volumes, as well as scan-rescan variability of both sequences were explored.

Results

Significantly higher CNR_WG was detected in MP2RAGE vs. MP-RAGE (Mean ± SD = 0.97 ± 0.04 vs. 0.8 ± 0.1 respectively; p<0.0001). Significantly higher total brain GM, and lower cerebrospinal fluid‏ volumes were obtained from MP2RAGE vs. MP-RAGE based on all segmentation methods (p<0.05 in all cases). Whole-brain voxel-wise comparisons revealed higher GM tissue probability in the thalamus, putamen, caudate, lingual gyrus, and precentral gyrus based on MP2RAGE compared with MP-RAGE. Moreover, significantly higher WM probability was observed in the cerebellum, corpus callosum, and frontal-and-temporal regions in MP2RAGE vs. MP-RAGE. Finally, MP2RAGE showed a higher mean percentage of change in total brain GM compared to MP-RAGE. On the other hand, MP-RAGE demonstrated a higher overtime percentage of change in WM and DGM volumes compared to MP2RAGE.

Conclusions

Due to its higher CNR, MP2RAGE resulted in reproducible brain tissue segmentation, and thus is a recommended method for volumetric imaging biomarkers for the monitoring of neurological diseases.

A metric learning method for estimating myelin content based on T2-weighted MRI from a de- and re-myelination model of multiple sclerosis

Myelin plays a critical role in the pathogenesis of neurological disorders but is difficult to characterize in vivo using standard analysis methods. Our goal was to develop a novel analytical framework for estimating myelin content using T2-weighted magnetic resonance imaging (MRI) based on a de- and re-myelination model of multiple sclerosis. We examined 18 mice with lysolecithin induced demyelination and spontaneous remyelination in the ventral white matter of thoracic spinal cord. Cohorts of 6 mice underwent 9.4T MRI at days 7 (peak demyelination), 14 (ongoing recovery), and 28 (near complete recovery), as well as histological analysis of myelin and the associated cellularity at corresponding timepoints. Our MRI framework took an unsupervised learning approach, including tissue segmentation using a Gaussian Markov random field (GMRF), and myelin and cellularity feature estimation based on the Mahalanobis distance. For comparison, we also investigated 2 regression-based supervised learning approaches, one using our GMRF results, and another using a freely available generalized additive model (GAM). Results showed that GMRF segmentation was 73.2% accurate, and our unsupervised learning method achieved a correlation coefficient of 0.67 (top quartile: 0.78) with histological myelin, similar to 0.70 (top quartile: 0.78) obtained using supervised analyses. Further, the area under the receiver operator characteristic curve of our unsupervised myelin feature (0.883, 95% CI: 0.874-0.891) was significantly better than any of the supervised models in detecting white matter myelin as compared to histology. Collectively, metric learning using standard MRI may prove to be a new alternative method for estimating myelin content, which ultimately can improve our disease monitoring ability in a clinical setting.

Nonmechanical parfocal and autofocus features based on wave propagation distribution in lensfree holographic microscopy

Performing long-term cell observations is a non-trivial task for conventional optical microscopy, since it is usually not compatible with environments of an incubator and its temperature and humidity requirements. Lensless holographic microscopy, being entirely based on semiconductor chips without lenses and without any moving parts, has proven to be a very interesting alternative to conventional microscopy. Here, we report on the integration of a computational parfocal feature, which operates based on wave propagation distribution analysis, to perform a fast autofocusing process. This unique non-mechanical focusing approach was implemented to keep the imaged object staying in-focus during continuous long-term and real-time recordings. A light-emitting diode (LED) combined with pinhole setup was used to realize a point light source, leading to a resolution down to 2.76 μm. Our approach delivers not only in-focus sharp images of dynamic cells, but also three-dimensional (3D) information on their (x, y, z)-positions. System reliability tests were conducted inside a sealed incubator to monitor cultures of three different biological living cells (i.e., MIN6, neuroblastoma (SH-SY5Y), and Prorocentrum minimum). Altogether, this autofocusing framework enables new opportunities for highly integrated microscopic imaging and dynamic tracking of moving objects in harsh environments with large sample areas.

Bayesian estimation of multicomponent relaxation parameters in magnetic resonance fingerprinting

PURPOSE

To estimate multiple components within a single voxel in magnetic resonance fingerprinting when the number and types of tissues comprising the voxel are not known a priori.

THEORY

Multiple tissue components within a single voxel are potentially separable with magnetic resonance fingerprinting as a result of differences in signal evolutions of each component. The Bayesian framework for inverse problems provides a natural and flexible setting for solving this problem when the tissue composition per voxel is unknown. Assuming that only a few entries from the dictionary contribute to a mixed signal, sparsity-promoting priors can be placed upon the solution.

METHODS

An iterative algorithm is applied to compute the maximum a posteriori estimator of the posterior probability density to determine the magnetic resonance fingerprinting dictionary entries that contribute most significantly to mixed or pure voxels.

RESULTS

Simulation results show that the algorithm is robust in finding the component tissues of mixed voxels. Preliminary in vivo data confirm this result, and show good agreement in voxels containing pure tissue.

CONCLUSIONS

The Bayesian framework and algorithm shown provide accurate solutions for the partial-volume problem in magnetic resonance fingerprinting. The flexibility of the method will allow further study into different priors and hyperpriors that can be applied in the model. Magn Reson Med 80:159-170, 2018. © 2017 International Society for Magnetic Resonance in Medicine.

Multi-tissue partial volume quantification in multi-contrast MRI using an optimised spectral unmixing approach

Multi-tissue partial volume estimation in MRI images is investigated with a viewpoint related to spectral unmixing as used in hyperspectral imaging. The main contribution of this paper is twofold. It firstly proposes a theoretical analysis of the statistical optimality conditions of the proportion estimation problem, which in the context of multi-contrast MRI data acquisition allows to appropriately set the imaging sequence parameters. Secondly, an efficient proportion quantification algorithm based on the minimisation of a penalised least-square criterion incorporating a regularity constraint on the spatial distribution of the proportions is proposed. Furthermore, the resulting developments are discussed using empirical simulations. The practical usefulness of the spectral unmixing approach for partial volume quantification in MRI is illustrated through an application to food analysis on the proving of a Danish pastry.

A Heavy Tailed Expectation Maximization Hidden Markov Random Field Model with Applications to Segmentation of MRI

A wide range of segmentation approaches assumes that intensity histograms extracted from magnetic resonance images (MRI) have a distribution for each brain tissue that can be modeled by a Gaussian distribution or a mixture of them. Nevertheless, intensity histograms of White Matter and Gray Matter are not symmetric and they exhibit heavy tails. In this work, we present a hidden Markov random field model with expectation maximization (EM-HMRF) modeling the components using the α-stable distribution. The proposed model is a generalization of the widely used EM-HMRF algorithm with Gaussian distributions. We test the α-stable EM-HMRF model in synthetic data and brain MRI data. The proposed methodology presents two main advantages: Firstly, it is more robust to outliers. Secondly, we obtain similar results than using Gaussian when the Gaussian assumption holds. This approach is able to model the spatial dependence between neighboring voxels in tomographic brain MRI.

Comparison of image intensity, local, and multi-atlas priors in brain tissue classification

PURPOSE

Automated and accurate tissue classification in three-dimensional brain magnetic resonance images is essential in volumetric morphometry or as a preprocessing step for diagnosing brain diseases. However, noise, intensity in homogeneity, and partial volume effects limit the classification accuracy of existing methods. This paper provides a comparative study on the contributions of three commonly used image information priors for tissue classification in normal brains: image intensity, local, and multi-atlas priors.

METHODS

We compared the effectiveness of the three priors by comparing the four methods modeling them: K-Means (KM), KM combined with a Markov Random Field (KM-MRF), multi-atlas segmentation (MAS), and the combination of KM, MRF, and MAS (KM-MRF-MAS). The key parameters and factors in each of the four methods are analyzed, and the performance of all the models is compared quantitatively and qualitatively on both simulated and real data.

RESULTS

The KM-MRF-MAS model that combines the three image information priors performs best.

CONCLUSIONS

The image intensity prior is insufficient to generate reasonable results for a few images. Introducing local and multi-atlas priors results in improved brain tissue classification. This study provides a general guide on what image information priors can be used for effective brain tissue classification.

Partial volume model for brain MRI scan using MP2RAGE

MP2RAGE is a T1 weighted MRI sequence that estimates a composite image providing much reduction of the receiver bias, has a high intensity dynamic range, and provides an estimate of T1 mapping. It is, therefore, an appealing option for brain morphometry studies. However, previous studies have reported a difference in cortical thickness computed from MP2RAGE compared with widely used Multi-Echo MPRAGE. In this article, we demonstrated that using standard segmentation and partial volume estimation techniques on MP2RAGE introduces systematic errors, and we proposed a new model to estimate partial volume of the cortical gray matter. We also included in their model a local estimate of tissue intensity to take into account the natural variation of tissue intensity across the brain. A theoretical framework is provided and validated using synthetic and physical phantoms. A repeatability experiment comparing MPRAGE and MP2RAGE confirmed that MP2RAGE using our model could be considered for structural imaging in brain morphology study, with similar cortical thickness estimate than that computed with MPRAGE. Hum Brain Mapp 38:5115-5127, 2017. © 2017 Wiley Periodicals, Inc.

Comparison of T1-weighted 2D TSE, 3D SPGR, and two-point 3D Dixon MRI for automated segmentation of visceral adipose tissue at 3 Tesla

OBJECTIVES

To evaluate and compare conventional T1-weighted 2D turbo spin echo (TSE), T1-weighted 3D volumetric interpolated breath-hold examination (VIBE), and two-point 3D Dixon-VIBE sequences for automatic segmentation of visceral adipose tissue (VAT) volume at 3 Tesla by measuring and compensating for errors arising from intensity nonuniformity (INU) and partial volume effects (PVE).

MATERIALS AND METHODS

The body trunks of 28 volunteers with body mass index values ranging from 18 to 41.2 kg/m² (30.02 ± 6.63 kg/m²) were scanned at 3 Tesla using three imaging techniques. Automatic methods were applied to reduce INU and PVE and to segment VAT. The automatically segmented VAT volumes obtained from all acquisitions were then statistically and objectively evaluated against the manually segmented (reference) VAT volumes.

RESULTS

Comparing the reference volumes with the VAT volumes automatically segmented over the uncorrected images showed that INU led to an average relative volume difference of -59.22 ± 11.59, 2.21 ± 47.04, and -43.05 ± 5.01 % for the TSE, VIBE, and Dixon images, respectively, while PVE led to average differences of -34.85 ± 19.85, -15.13 ± 11.04, and -33.79 ± 20.38 %. After signal correction, differences of -2.72 ± 6.60, 34.02 ± 36.99, and -2.23 ± 7.58 % were obtained between the reference and the automatically segmented volumes. A paired-sample two-tailed t test revealed no significant difference between the reference and automatically segmented VAT volumes of the corrected TSE (p = 0.614) and Dixon (p = 0.969) images, but showed a significant VAT overestimation using the corrected VIBE images.

CONCLUSION

Under similar imaging conditions and spatial resolution, automatically segmented VAT volumes obtained from the corrected TSE and Dixon images agreed with each other and with the reference volumes. These results demonstrate the efficacy of the signal correction methods and the similar accuracy of TSE and Dixon imaging for automatic volumetry of VAT at 3 Tesla.

Segmentation of Brain Tissues from Magnetic Resonance Images Using Adaptively Regularized Kernel-Based Fuzzy C-Means Clustering

An adaptively regularized kernel-based fuzzy C-means clustering framework is proposed for segmentation of brain magnetic resonance images. The framework can be in the form of three algorithms for the local average grayscale being replaced by the grayscale of the average filter, median filter, and devised weighted images, respectively. The algorithms employ the heterogeneity of grayscales in the neighborhood and exploit this measure for local contextual information and replace the standard Euclidean distance with Gaussian radial basis kernel functions. The main advantages are adaptiveness to local context, enhanced robustness to preserve image details, independence of clustering parameters, and decreased computational costs. The algorithms have been validated against both synthetic and clinical magnetic resonance images with different types and levels of noises and compared with 6 recent soft clustering algorithms. Experimental results show that the proposed algorithms are superior in preserving image details and segmentation accuracy while maintaining a low computational complexity.

Partial volume effect modeling for segmentation and tissue classification of brain magnetic resonance images: A review

Quantitative analysis of magnetic resonance (MR) brain images are facilitated by the development of automated segmentation algorithms. A single image voxel may contain of several types of tissues due to the finite spatial resolution of the imaging device. This phenomenon, termed partial volume effect (PVE), complicates the segmentation process, and, due to the complexity of human brain anatomy, the PVE is an important factor for accurate brain structure quantification. Partial volume estimation refers to a generalized segmentation task where the amount of each tissue type within each voxel is solved. This review aims to provide a systematic, tutorial-like overview and categorization of methods for partial volume estimation in brain MRI. The review concentrates on the statistically based approaches for partial volume estimation and also explains differences to other, similar image segmentation approaches.

Generalized method for partial volume estimation and tissue segmentation in cerebral magnetic resonance images

An artifact found in magnetic resonance images (MRI) called partial volume averaging (PVA) has received much attention since accurate segmentation of cerebral anatomy and pathology is impeded by this artifact. Traditional neurological segmentation techniques rely on Gaussian mixture models to handle noise and PVA, or high-dimensional feature sets that exploit redundancy in multispectral datasets. Unfortunately, model-based techniques may not be optimal for images with non-Gaussian noise distributions and/or pathology, and multispectral techniques model probabilities instead of the partial volume (PV) fraction. For robust segmentation, a PV fraction estimation approach is developed for cerebral MRI that does not depend on predetermined intensity distribution models or multispectral scans. Instead, the PV fraction is estimated directly from each image using an adaptively defined global edge map constructed by exploiting a relationship between edge content and PVA. The final PVA map is used to segment anatomy and pathology with subvoxel accuracy. Validation on simulated and real, pathology-free T1 MRI (Gaussian noise), as well as pathological fluid attenuation inversion recovery MRI (non-Gaussian noise), demonstrate that the PV fraction is accurately estimated and the resultant segmentation is robust. Comparison to model-based methods further highlight the benefits of the current approach.

Automatic segmentation of cerebral white matter hyperintensities using only 3D FLAIR images

Magnetic Resonance (MR) white matter hyperintensities have been shown to predict an increased risk of developing cognitive decline. However, their actual role in the conversion to dementia is still not fully understood. Automatic segmentation methods can help in the screening and monitoring of Mild Cognitive Impairment patients who take part in large population-based studies. Most existing segmentation approaches use multimodal MR images. However, multiple acquisitions represent a limitation in terms of both patient comfort and computational complexity of the algorithms. In this work, we propose an automatic lesion segmentation method that uses only three-dimensional fluid-attenuation inversion recovery (FLAIR) images. We use a modified context-sensitive Gaussian mixture model to determine voxel class probabilities, followed by correction of FLAIR artifacts. We evaluate the method against the manual segmentation performed by an experienced neuroradiologist and compare the results with other unimodal segmentation approaches. Finally, we apply our method to the segmentation of multiple sclerosis lesions by using a publicly available benchmark dataset. Results show a similar performance to other state-of-the-art multimodal methods, as well as to the human rater.

Parameterization of the distribution of white and grey matter in MRI using the α-stable distribution

This work presents a study of the distribution of the grey matter (GM) and white matter (WM) in brain magnetic resonance imaging (MRI). The distribution of GM and WM is characterized using a mixture of α-stable distributions. A Bayesian α-stable mixture model for histogram data is presented and unknown parameters are sampled using the Metropolis-Hastings algorithm. The proposed methodology is tested in 18 real images from the MRI brain segmentation repository. The GM and WM distributions are accurately estimated. The α-stable distribution mixture model presented in this paper can be used as previous step in more complex MRI segmentation procedures using spatial information. Furthermore, due to the fact that the α-stable distribution is a generalization of the Gaussian distribution, the proposed methodology can be applied instead of the Gaussian mixture model, which is widely used in segmentation of brain MRI in the literature.

Automatic segmentation of human facial tissue by MRI-CT fusion: a feasibility study

The aim of this study was to develop automatic image segmentation methods to segment human facial tissue which contains very thin anatomic structures. The segmentation output can be used to construct a more realistic human face model for a variety of purposes like surgery planning, patient specific prosthesis design and facial expression simulation. Segmentation methods developed were based on Bayesian and Level Set frameworks, which were applied on three image types: magnetic resonance imaging (MRI), computerized tomography (CT) and fusion, in which case information from both modalities were utilized maximally for every tissue type. The results on human data indicated that fusion, thickness adaptive and postprocessing options provided the best muscle/fat segmentation scores in both Level Set and Bayesian methods. When the best Level Set and Bayesian methods were compared, scores of the latter were better. Number of algorithm parameters (to be trained) and computer run time measured were also in favour of the Bayesian method.

Comparison of breast tissue measurements using magnetic resonance imaging, digital mammography and a mathematical algorithm

Women with mostly mammographically dense fibroglandular tissue (breast density, BD) have a four- to six-fold increased risk for breast cancer compared to women with little BD. BD is most frequently estimated from two-dimensional (2D) views of mammograms by a histogram segmentation approach (HSM) and more recently by a mathematical algorithm consisting of mammographic imaging parameters (MATH). Two non-invasive clinical magnetic resonance imaging (MRI) protocols: 3D gradient-echo (3DGRE) and short tau inversion recovery (STIR) were modified for 3D volumetric reconstruction of the breast for measuring fatty and fibroglandular tissue volumes by a Gaussian-distribution curve-fitting algorithm. Replicate breast exams (N = 2 to 7 replicates in six women) by 3DGRE and STIR were highly reproducible for all tissue-volume estimates (coefficients of variation <5%). Reliability studies compared measurements from four methods, 3DGRE, STIR, HSM, and MATH (N = 95 women) by linear regression and intra-class correlation (ICC) analyses. Rsqr, regression slopes, and ICC, respectively, were (1) 0.76-0.86, 0.8-1.1, and 0.87-0.92 for %-gland tissue, (2) 0.72-0.82, 0.64-0.96, and 0.77-0.91, for glandular volume, (3) 0.87-0.98, 0.94-1.07, and 0.89-0.99, for fat volume, and (4) 0.89-0.98, 0.94-1.00, and 0.89-0.98, for total breast volume. For all values estimated, the correlation was stronger for comparisons between the two MRI than between each MRI versus mammography, and between each MRI versus MATH data than between each MRI versus HSM data. All ICC values were >0.75 indicating that all four methods were reliable for measuring BD and that the mathematical algorithm and the two complimentary non-invasive MRI protocols could objectively and reliably estimate different types of breast tissues.

An open source multivariate framework for n-tissue segmentation with evaluation on public data

We introduce Atropos, an ITK-based multivariate n-class open source segmentation algorithm distributed with ANTs ( http://www.picsl.upenn.edu/ANTs). The Bayesian formulation of the segmentation problem is solved using the Expectation Maximization (EM) algorithm with the modeling of the class intensities based on either parametric or non-parametric finite mixtures. Atropos is capable of incorporating spatial prior probability maps (sparse), prior label maps and/or Markov Random Field (MRF) modeling. Atropos has also been efficiently implemented to handle large quantities of possible labelings (in the experimental section, we use up to 69 classes) with a minimal memory footprint. This work describes the technical and implementation aspects of Atropos and evaluates its performance on two different ground-truth datasets. First, we use the BrainWeb dataset from Montreal Neurological Institute to evaluate three-tissue segmentation performance via (1) K-means segmentation without use of template data; (2) MRF segmentation with initialization by prior probability maps derived from a group template; (3) Prior-based segmentation with use of spatial prior probability maps derived from a group template. We also evaluate Atropos performance by using spatial priors to drive a 69-class EM segmentation problem derived from the Hammers atlas from University College London. These evaluation studies, combined with illustrative examples that exercise Atropos options, demonstrate both performance and wide applicability of this new platform-independent open source segmentation tool.

Ligation of the jugular veins does not result in brain inflammation or demyelination in mice

An alternative hypothesis has been proposed implicating chronic cerebrospinal venous insufficiency (CCSVI) as a potential cause of multiple sclerosis (MS). We aimed to evaluate the validity of this hypothesis in a controlled animal model. Animal experiments were approved by the institutional animal care committee. The jugular veins in SJL mice were ligated bilaterally (n = 20), and the mice were observed for up to six months after ligation. Sham-operated mice (n = 15) and mice induced with experimental autoimmune encephalomyelitis (n = 8) were used as negative and positive controls, respectively. The animals were evaluated using CT venography and (99m)Tc-exametazime to assess for structural and hemodynamic changes. Imaging was performed to evaluate for signs of blood-brain barrier (BBB) breakdown and neuroinflammation. Flow cytometry and histopathology were performed to assess inflammatory cell populations and demyelination. There were both structural changes (stenosis, collaterals) in the jugular venous drainage and hemodynamic disturbances in the brain on Tc99m-exametazime scintigraphy (p = 0.024). In the JVL mice, gadolinium MRI and immunofluorescence imaging for barrier molecules did not reveal evidence of BBB breakdown (p = 0.58). Myeloperoxidase, matrix metalloproteinase, and protease molecular imaging did not reveal signs of increased neuroinflammation (all p>0.05). Flow cytometry and histopathology also did not reveal increase in inflammatory cell infiltration or population shifts. No evidence of demyelination was found, and the mice remained without clinical signs. Despite the structural and hemodynamic changes, we did not identify changes in the BBB permeability, neuroinflammation, demyelination, or clinical signs in the JVL group compared to the sham group. Therefore, our murine model does not support CCSVI as a cause of demyelinating diseases such as multiple sclerosis.

Adaptive neonate brain segmentation

Babies born prematurely are at increased risk of adverse neurodevelopmental outcomes. Recent advances suggest that measurement of brain volumes can help in defining biomarkers for neurodevelopmental outcome. These techniques rely on an accurate segmentation of the MRI data. However, due to lack of contrast, partial volume (PV) effect, the existence of both hypo- and hyper-intensities and significant natural and pathological anatomical variability, the segmentation of neonatal brain MRI is challenging. We propose a pipeline for image segmentation that uses a novel multi-model Maximum a posteriori Expectation Maximisation (MAP-EM) segmentation algorithm with a prior over both intensities and the tissue proportions, a B0 inhomogeneity correction, and a spatial homogeneity term through the use of a Markov Random Field. This robust and adaptive technique enables the segmentation of images with high anatomical disparity from a normal population. Furthermore, the proposed method implicitly models Partial Volume, mitigating the problem of neonatal white/grey matter intensity inversion. Experiments performed on a clinical cohort show expected statistically significant correlations with gestational age at birth and birthweight. Furthermore, the proposed method obtains statistically significant improvements in Dice scores when compared to the a Maximum Likelihood EM algorithm.

Brain MRI segmentation with multiphase minimal partitioning: a comparative study

This paper presents the implementation and quantitative evaluation of a multiphase three-dimensional deformable model in a level set framework for automated segmentation of brain MRIs. The segmentation algorithm performs an optimal partitioning of three-dimensional data based on homogeneity measures that naturally evolves to the extraction of different tissue types in the brain. Random seed initialization was used to minimize the sensitivity of the method to initial conditions while avoiding the need for a priori information. This random initialization ensures robustness of the method with respect to the initialization and the minimization set up. Postprocessing corrections with morphological operators were applied to refine the details of the global segmentation method. A clinical study was performed on a database of 10 adult brain MRI volumes to compare the level set segmentation to three other methods: “idealized” intensity thresholding, fuzzy connectedness, and an expectation maximization classification using hidden Markov random fields. Quantitative evaluation of segmentation accuracy was performed with comparison to manual segmentation computing true positive and false positive volume fractions. A statistical comparison of the segmentation methods was performed through a Wilcoxon analysis of these error rates and results showed very high quality and stability of the multiphase three-dimensional level set method.

Geometrically induced force interaction for three-dimensional deformable models

In this paper, we propose a novel 3-D deformable model that is based upon a geometrically induced external force field which can be conveniently generalized to arbitrary dimensions. This external force field is based upon hypothesized interactions between the relative geometries of the deformable model and the object boundary characterized by image gradient. The evolution of the deformable model is solved using the level set method so that topological changes are handled automatically. The relative geometrical configurations between the deformable model and the object boundaries contribute to a dynamic vector force field that changes accordingly as the deformable model evolves. The geometrically induced dynamic interaction force has been shown to greatly improve the deformable model performance in acquiring complex geometries and highly concave boundaries, and it gives the deformable model a high invariancy in initialization configurations. The voxel interactions across the whole image domain provide a global view of the object boundary representation, giving the external force a long attraction range. The bidirectionality of the external force field allows the new deformable model to deal with arbitrary cross-boundary initializations, and facilitates the handling of weak edges and broken boundaries. In addition, we show that by enhancing the geometrical interaction field with a nonlocal edge-preserving algorithm, the new deformable model can effectively overcome image noise. We provide a comparative study on the segmentation of various geometries with different topologies from both synthetic and real images, and show that the proposed method achieves significant improvements against existing image gradient techniques.

Component analysis approach to estimation of tissue intensity distributions of 3D images

Many segmentation algorithms in medical imaging rely on accurate modeling and estimation of tissue intensity probability density functions. Gaussian mixture modeling, currently the most common approach, has several drawbacks, such as reliance on a Gaussian model and iterative local optimization used to estimate the model parameters. It also does not take advantage of substantially larger amount of data provided by 3D acquisitions, which are becoming standard in clinical environment. We propose a novel and completely non-parametric algorithm to estimate the tissue intensity probabilities in 3D images. Instead of relying on traditional framework of iterating between classification and estimation, we pose the problem as an instance of a blind source separation problem, where the unknown distributions are treated as sources and histograms of image subvolumes as mixtures. The new approach performed well on synthetic data and real magnetic resonance imaging (MRI) scans of the brain, robustly capturing intensity distributions of even small image structures and partial volume voxels.

LoAd: a locally adaptive cortical segmentation algorithm

Thickness measurements of the cerebral cortex can aid diagnosis and provide valuable information about the temporal evolution of diseases such as Alzheimer's, Huntington's, and schizophrenia. Methods that measure the thickness of the cerebral cortex from in-vivo magnetic resonance (MR) images rely on an accurate segmentation of the MR data. However, segmenting the cortex in a robust and accurate way still poses a challenge due to the presence of noise, intensity non-uniformity, partial volume effects, the limited resolution of MRI and the highly convoluted shape of the cortical folds. Beginning with a well-established probabilistic segmentation model with anatomical tissue priors, we propose three post-processing refinements: a novel modification of the prior information to reduce segmentation bias; introduction of explicit partial volume classes; and a locally varying MRF-based model for enhancement of sulci and gyri. Experiments performed on a new digital phantom, on BrainWeb data and on data from the Alzheimer's Disease Neuroimaging Initiative (ADNI) show statistically significant improvements in Dice scores and PV estimation (p<10(-3)) and also increased thickness estimation accuracy when compared to three well established techniques.

Template-based B₁ inhomogeneity correction in 3T MRI brain studies

Low noise, high resolution, fast and accurate T₁ maps from MRI images of the brain can be performed using a dual flip angle method. However, B₁ field inhomogeneity, which is particularly problematic at high field strengths (e.g., 3T), limits the ability of the scanner to deliver the prescribed flip angle, introducing errors into the T₁ maps that limit the accuracy of quantitative analyses based on those maps. A dual repetition time method was used for acquiring a B₁ map to correct that inhomogeneity. Additional inaccuracies due to misregistration of the acquired T₁-weighted images were corrected by rigid registration, and the effects of misalignment on the T₁ maps were compared to those of B₁ inhomogeneity in 19 normal subjects. However, since B₁ map acquisition takes up precious scanning time and most retrospective studies do not have B₁ map, we designed a template-based correction strategy. B₁ maps from different subjects were aligned using a twelve-parameter affine registration. Recomputed T₁ maps showed an important improvement with respect to the noncorrected maps: histograms of all corrected maps exhibited two peaks corresponding to white and gray matter tissues, while unimodal histograms were observed in all uncorrected maps because of the inhomogeneity. A method to detect the best nonsubject-specific B₁ correction based on a set of features was designed. The optimum set of weighting factors for those features was computed. The best available B₁ correction was detected in almost all subjects while corrections comparable to the T₁ map corrected using the B₁ map from the same subject were detected in the others.

Automated brain tissue segmentation based on fractional signal mapping from inversion recovery Look-Locker acquisition

Most current automated segmentation methods are performed on T(1)- or T(2)-weighted MR images, relying on relative image intensity that is dependent on other MR parameters and sensitive to B(1) magnetic field inhomogeneity. Here, we propose an image segmentation method based on quantitative longitudinal magnetization relaxation time (T(1)) of brain tissues. Considering the partial volume effect, fractional volume maps of brain tissues (white matter, gray matter, and cerebrospinal fluid) were obtained by fitting the observed signal in an inversion recovery procedure to a linear combination of three exponential functions, which represents the relaxations of each of the tissue types. A Look-Locker acquisition was employed to accelerate the acquisition process. The feasibility and efficacy of this proposed method were evaluated using simulations and experiments. The potential applications of this method in the study of neurological disease as well as normal brain development and aging are discussed.

Segmentation of MRI brain scans using non-uniform partial volume densities

We present an algorithm that provides a partial volume segmentation of a T1-weighted image of the brain into gray matter, white matter and cerebrospinal fluid. The algorithm incorporates a non-uniform partial volume density that takes the curved nature of the cortex into account. The pure gray and white matter intensities are estimated from the image, using scanner noise and cortical partial volume effects. Expected tissue fractions are subsequently computed in each voxel. The algorithm has been tested for reliability, correct estimation of the pure tissue intensities on both real (repeated) MRI data and on simulated (brain) images. Intra-class correlation coefficients (ICCs) were above 0.93 for all volumes of the three tissue types for repeated scans from the same scanner, as well as for scans with different voxel sizes from different scanners with different field strengths. The implementation of our non-uniform partial volume density provided more reliable volumes and tissue fractions, compared to a uniform partial volume density. Applying the algorithm to simulated images showed that the pure tissue intensities were estimated accurately. Variations in cortical thickness did not influence the accuracy of the volume estimates, which is a valuable property when studying (possible) group differences. In conclusion, we have presented a new partial volume segmentation algorithm that allows for comparisons over scanners and voxel sizes.

White matter lesion extension to automatic brain tissue segmentation on MRI

A fully automated brain tissue segmentation method is optimized and extended with white matter lesion segmentation. Cerebrospinal fluid (CSF), gray matter (GM) and white matter (WM) are segmented by an atlas-based k-nearest neighbor classifier on multi-modal magnetic resonance imaging data. This classifier is trained by registering brain atlases to the subject. The resulting GM segmentation is used to automatically find a white matter lesion (WML) threshold in a fluid-attenuated inversion recovery scan. False positive lesions are removed by ensuring that the lesions are within the white matter. The method was visually validated on a set of 209 subjects. No segmentation errors were found in 98% of the brain tissue segmentations and 97% of the WML segmentations. A quantitative evaluation using manual segmentations was performed on a subset of 6 subjects for CSF, GM and WM segmentation and an additional 14 for the WML segmentations. The results indicated that the automatic segmentation accuracy is close to the interobserver variability of manual segmentations.

MR image segmentation using a power transformation approach

This study proposes a segmentation method for brain MR images using a distribution transformation approach. The method extends traditional Gaussian mixtures expectation-maximization segmentation to a power transformed version of mixed intensity distributions, which includes Gaussian mixtures as a special case. As MR intensities tend to exhibit non-Gaussianity due to partial volume effects, the proposed method is designed to fit non-Gaussian tissue intensity distributions. One advantage of the method is that it is intuitively appealing and computationally simple. To avoid performance degradation caused by intensity inhomogeneity, different methods for correcting bias fields were applied prior to image segmentation, and their correction effects on the segmentation results were examined in the empirical study. The partitions of brain tissues (i.e., gray and white matter) resulting from the method were validated and evaluated against manual segmentation results based on 38 real T1-weighted image volumes from the internet brain segmentation repository, and 18 simulated image volumes from BrainWeb. The Jaccard and Dice similarity indexes were computed to evaluate the performance of the proposed approach relative to the expert segmentations. Empirical results suggested that the proposed segmentation method yielded higher similarity measures for both gray matter and white matter as compared with those based on the traditional segmentation using the Gaussian mixtures approach.

Bayesian mixture models of variable dimension for image segmentation

We present Bayesian methodologies and apply Markov chain sampling techniques for exploring normal mixture models with an unknown number of components in the context of magnetic resonance imaging (MRI) segmentation. The experiments show that by estimating the number of components using sample-based approaches based on variable dimension models the discriminating power of the estimated components is improved. Two different MCMC methods are compared to perform the segmentation of simulated magnetic resonance brain scans, the reversible jump MCMC model and the Dirichlet process (DP) mixture model. The preference given to the Dirichlet process mixture model is discussed.

A deformable model-based system for 3D analysis and visualization of tumor in PET/CT images

This paper presents a tumor detecting system that allows interactive 3D tumor visualization and tumor volume measurements. An improved level set method is proposed to automatically segment the tumor images slice by slice. PET images are used to detect the tumor while CT images make a 3D representation of the patient's body possible. An initial slice with a seed within the tumor is firstly chosen by the operator. The system then performs automatically the tumor volume segmentation that allows the clinician to visualize the tumor, to measure it and to evaluate the best medical treatment adapted to the patient.

Robust MRI brain tissue parameter estimation by multistage outlier rejection

This article addresses the problem of the tissue type parameter estimation in brain MRI in the presence of partial volume effects. Automatic MRI brain tissue classification is hampered by partial volume effects that are caused by the finite resolution of the acquisition process. Due to this effect intensity distributions in brain MRI cannot be well modeled by a simple mixture of Gaussians and therefore more complex models have been developed. Unfortunately, these models do not seem to be robust enough for clinical conditions, as the quality of the tissue classification decreases rapidly with the image quality. Also, the application of these methods for pathological images with unmodeled intensities (e.g. MS plaques, tumors, etc.) remains uncertain. In the present work a new robust method for brain tissue characterization is presented, treating the partial volume affected voxels as outliers of the pure tissue distributions. The proposed method estimates the tissue characteristics from a reduced set of intensities belonging to a particular pure tissue class. This reduced set is selected by using a trimming procedure based on local gradient information and distributional data. This feature makes the method highly tolerant of a large amount of unexpected intensities without degrading its performance. The proposed method has been evaluated using both synthetic and real MR data and compared with state-of-the-art methods showing the best results in the comparative.

Hybrid multiresolution Slantlet transform and fuzzy c-means clustering approach for normal-pathological brain MR image segregation

The paper presents a new approach for automated segregation of brain MR images, using an improved orthogonal discrete wavelet transform (DWT), known as the Slantlet transform (ST), and a fuzzy c-means (FCM) clustering approach. ST has excellent time-frequency resolution characteristics and these can be achieved with shorter supports for the filter, compared to DWT employed for identical situations. FCM clustering, on the other hand, can provide efficient classification results, if it is implemented for well-processed input feature vectors. Thus, by combining both the ST and the FCM clustering approaches, a hybrid scheme has been developed that can segregate brain MR images. This automated tool when developed can infer whether the input image is that of a normal brain or a pathological brain. The proposed technique has been applied to several benchmark brain MR images and the results reveal excellent accuracy in characterizing human brain MR imaging.

An artificial immune-activated neural network applied to brain 3D MRI segmentation

In this paper, a new neural network model inspired by the biological immune system functions is presented. The model, termed Artificial Immune-Activated Neural Network (AIANN), extracts classification knowledge from a training data set, which is then used to classify input patterns or vectors. The AIANN is based on a neuron activation function whose behavior is conceptually modeled after the chemical bonds between the receptors and epitopes in the biological immune system. The bonding is controlled through an energy measure to ensure accurate recognition. The AIANN model was applied to the segmentation of 3-dimensional magnetic resonance imaging (MRI) data of the brain and a contextual basis was developed for the segmentation problem. Evaluation of the segmentation results was performed using both real MRI data obtained from the Center for Morphometric Analysis at Massachusetts General Hospital and simulated MRI data generated using the McGill University BrainWeb MRI simulator. Experimental results demonstrated that the AIANN model attained higher average results than those obtained using published methods for real MRI data and simulated MRI data, especially at low levels of noise.

A Dirichlet process mixture model for brain MRI tissue classification

Accurate classification of magnetic resonance images according to tissue type or region of interest has become a critical requirement in diagnosis, treatment planning, and cognitive neuroscience. Several authors have shown that finite mixture models give excellent results in the automated segmentation of MR images of the human normal brain. However, performance and robustness of finite mixture models deteriorate when the models have to deal with a variety of anatomical structures. In this paper, we propose a nonparametric Bayesian model for tissue classification of MR images of the brain. The model, known as Dirichlet process mixture model, uses Dirichlet process priors to overcome the limitations of current parametric finite mixture models. To validate the accuracy and robustness of our method we present the results of experiments carried out on simulated MR brain scans, as well as on real MR image data. The results are compared with similar results from other well-known MRI segmentation methods.

Accurate quantification methods to evaluate cervical cord atrophy in multiple sclerosis patients

OBJECT

Automatic accurate measurement techniques are needed to increase reproducibility in the quantification of cervical cord area (CCA) with magnetic resonance (MR) imaging in the assessment of central nervous system (CNS) atrophy in multiple sclerosis (MS) patients.

MATERIALS AND METHODS

Two segmentation methods were implemented: (1) spatial mean brightness level estimation (SMBLE), and (2) partial-volume modeling (PVM). These were evaluated with the inclusion of spinal cord inclination and/or partial-volume-effect corrections. An averaged manually segmented set was considered as reference. Thirty MR studies were used to compare the different methods. A set of 15 MS patients and 15 control subjects within a two-year longitudinal study were used to evaluate cord atrophy with the best method. Statistical evaluation was made by using an intraclass correlation coefficient and Bland-Altman comparisons.

RESULTS

Partial-volume modeling with spinal cord inclination correction and partial-volume spinal-cord contour contribution estimation was the most accurate method. The longitudinal test showed a 4% decrease in CCA in MS patients with no significant reduction in control subjects.

CONCLUSION

The automatic PVM cord-segmentation approach, taking into consideration the spinal-cord inclination and partial-volume treatment, provides reproducibility and increased accuracy in the evaluation of cord atrophy, allowing the monitoring of changes in MS patients.

3D Brain Segmentation Using Dual-Front Active Contours with Optional User Interaction

Important attributes of 3D brain cortex segmentation algorithms include robustness, accuracy, computational efficiency, and facilitation of user interaction, yet few algorithms incorporate all of these traits. Manual segmentation is highly accurate but tedious and laborious. Most automatic techniques, while less demanding on the user, are much less accurate. It would be useful to employ a fast automatic segmentation procedure to do most of the work but still allow an expert user to interactively guide the segmentation to ensure an accurate final result. We propose a novel 3D brain cortex segmentation procedure utilizing dual-front active contours which minimize image-based energies in a manner that yields flexibly global minimizers based on active regions. Region-based information and boundary-based information may be combined flexibly in the evolution potentials for accurate segmentation results. The resulting scheme is not only more robust but much faster and allows the user to guide the final segmentation through simple mouse clicks which add extra seed points. Due to the flexibly global nature of the dual-front evolution model, single mouse clicks yield corrections to the segmentation that extend far beyond their initial locations, thus minimizing the user effort. Results on 15 simulated and 20 real 3D brain images demonstrate the robustness, accuracy, and speed of our scheme compared with other methods.

Variational Bayes inference of spatial mixture models for segmentation

Mixture models are commonly used in the statistical segmentation of images. For example, they can be used for the segmentation of structural medical images into different matter types, or of statistical parametric maps into activating and nonactivating brain regions in functional imaging. Spatial mixture models have been developed to augment histogram information with spatial regularization using Markov random fields (MRFs). In previous work, an approximate model was developed to allow adaptive determination of the parameter controlling the strength of spatial regularization. Inference was performed using Markov Chain Monte Carlo (MCMC) sampling. However, this approach is prohibitively slow for large datasets. In this work, a more efficient inference approach is presented. This combines a variational Bayes approximation with a second-order Taylor expansion of the components of the posterior distribution, which would otherwise be intractable to Variational Bayes. This provides inference on fully adaptive spatial mixture models an order of magnitude faster than MCMC. We examine the behavior of this approach when applied to artificial data with different spatial characteristics, and to functional magnetic resonance imaging statistical parametric maps.

Comparative assessment of statistical brain MR image segmentation algorithms and their impact on partial volume correction in PET

Magnetic resonance imaging (MRI)-guided partial volume effect correction (PVC) in brain positron emission tomography (PET) is now a well-established approach to compensate the large bias in the estimate of regional radioactivity concentration, especially for small structures. The accuracy of the algorithms developed so far is, however, largely dependent on the performance of segmentation methods partitioning MRI brain data into its main classes, namely gray matter (GM), white matter (WM), and cerebrospinal fluid (CSF). A comparative evaluation of three brain MRI segmentation algorithms using simulated and clinical brain MR data was performed, and subsequently their impact on PVC in 18F-FDG and 18F-DOPA brain PET imaging was assessed. Two algorithms, the first is bundled in the Statistical Parametric Mapping (SPM2) package while the other is the Expectation Maximization Segmentation (EMS) algorithm, incorporate a priori probability images derived from MR images of a large number of subjects. The third, here referred to as the HBSA algorithm, is a histogram-based segmentation algorithm incorporating an Expectation Maximization approach to model a four-Gaussian mixture for both global and local histograms. Simulated under different combinations of noise and intensity non-uniformity, MR brain phantoms with known true volumes for the different brain classes were generated. The algorithms' performance was checked by calculating the kappa index assessing similarities with the "ground truth" as well as multiclass type I and type II errors including misclassification rates. The impact of image segmentation algorithms on PVC was then quantified using clinical data. The segmented tissues of patients' brain MRI were given as input to the region of interest (RoI)-based geometric transfer matrix (GTM) PVC algorithm, and quantitative comparisons were made. The results of digital MRI phantom studies suggest that the use of HBSA produces the best performance for WM classification. For GM classification, it is suggested to use the EMS. Segmentation performed on clinical MRI data show quite substantial differences, especially when lesions are present. For the particular case of PVC, SPM2 and EMS algorithms show very similar results and may be used interchangeably. The use of HBSA is not recommended for PVC. The partial volume corrected activities in some regions of the brain show quite large relative differences when performing paired analysis on 2 algorithms, implying a careful choice of the segmentation algorithm for GTM-based PVC.

Partial volume reduction by interpolation with reverse diffusion

Many medical images suffer from the partial volume effect where a boundary between two structures of interest falls in the midst of a voxel giving a signal value that is a mixture of the two. We propose a method to restore the ideal boundary by splitting a voxel into subvoxels and reapportioning the signal into the subvoxels. Each voxel is divided by nearest neighbor interpolation. The gray level of each subvoxel is considered as “material” able to move between subvoxels but not between voxels. A partial differential equation is written to allow the material to flow towards the highest gradient direction, creating a “reverse” diffusion process. Flow is subject to constraints that tend to create step edges. Material is conserved in the process thereby conserving signal. The method proceeds until the flow decreases to a low value. To test the method, synthetic images were downsampled to simulate the partial volume artifact and restored. Corrected images were remarkably closer both visually and quantitatively to the original images than those obtained from common interpolation methods: on simulated data standard deviation of the errors were 3.8%, 6.6%, and 7.1% of the dynamic range for the proposed method, bicubic, and bilinear interpolation, respectively. The method was relatively insensitive to noise. On gray level, scanned text, MRI physical phantom, and brain images, restored images processed with the new method were visually much closer to high-resolution counterparts than those obtained with common interpolation methods.

Comparison and validation of tissue modelization and statistical classification methods in T1-weighted MR brain images

This paper presents a validation study on statistical nonsupervised brain tissue classification techniques in magnetic resonance (MR) images. Several image models assuming different hypotheses regarding the intensity distribution model, the spatial model and the number of classes are assessed. The methods are tested on simulated data for which the classification ground truth is known. Different noise and intensity nonuniformities are added to simulate real imaging conditions. No enhancement of the image quality is considered either before or during the classification process. This way, the accuracy of the methods and their robustness against image artifacts are tested. Classification is also performed on real data where a quantitative validation compares the methods' results with an estimated ground truth from manual segmentations by experts. Validity of the various classification methods in the labeling of the image as well as in the tissue volume is estimated with different local and global measures. Results demonstrate that methods relying on both intensity and spatial information are more robust to noise and field inhomogeneities. We also demonstrate that partial volume is not perfectly modeled, even though methods that account for mixture classes outperform methods that only consider pure Gaussian classes. Finally, we show that simulated data results can also be extended to real data.

Probabilistic segmentation of brain tissue in MR imaging

A new method has been developed for probabilistic segmentation of five different types of brain structures: white matter, gray matter, cerebro-spinal fluid without ventricles, ventricles and white matter lesion in cranial MR imaging. The algorithm is based on information from T1-weighted (T1-w), inversion recovery (IR), proton density-weighted (PD), T2-weighted (T2-w) and fluid attenuation inversion recovery (FLAIR) scans. It uses the K-Nearest Neighbor classification technique that builds a feature space from spatial information and voxel intensities. The technique generates for each tissue type an image representing the probability per voxel being part of it. By application of thresholds on these probability maps, binary segmentations can be obtained. A similarity index (SI) and a probabilistic SI (PSI) were calculated for quantitative evaluation of the results. The influence of each image type on the performance was investigated by alternately leaving out one of the five scan types. This procedure showed that the incorporation of the T1-w, PD or T2-w did not significantly improve the segmentation results. Further investigation indicated that the combination of IR and FLAIR was optimal for segmentation of the five brain tissue types. Evaluation with respect to the gold standard showed that the SI-values for all tissues exceeded 0.8 and all PSI-values exceeded 0.7, implying an excellent agreement.

Muscle and fat quantification in MRI gradient echo images using a partial volume detection method. Application to the characterization of pig belly tissue

Complete dissection is the current reference method to quantify muscle and fat tissue on pig carcasses. Magnetic resonance imaging (MRI) is an appropriate nondestructive alternative method that can provide reliable and quantitative information on pig carcass composition without losing the spatial information. We have developed a method to quantify the amount of fat tissue and muscle in gradient echo MR images. This method is based on the method proposed by Shattuck et al. [12]. It provides segmentation of pure tissue and partial volume voxels, which allows separation of muscle and fat tissue including the fine insertions of intermuscular fat. Partial volume voxel signal is expected to be proportional to the signals of pure tissue constituting them or at least to vary monotonously with the proportion of each tissue. However, it is not always the case with gradient echo sequence due to the chemical shift effect. We studied this effect on a fat tissue/muscle interface model with variable proportion of water in the fat tissue and variable TE. We found that at TE=8 ms, for a 0.2-T MRI system, the requirement of Shattuck's method were filled thanks to the presence of water in fat tissue. Moreover, we extended the segmentation method with a simple correction scheme to compute more accurately the proportions of each tissue in partial volume voxels. We used this method to evaluate the fat tissue and muscle on 24 pig bellies using a gradient echo sequence (TR 700 ms, TE 8 ms, slice thickness 8 mm, number of averages 8, flip angle 90 degrees , FOV 512 mm, matrix 512*512, Rect. FOV 4/8, 19 slices, space between slices 2 mm). The image analysis results were compared with dissection results giving a prediction error of the muscle content (mean=2.7 kg) of 88.9 g and of the fat content (mean=2.7 kg) of 115.8 g without correction of the chemical shift effect in the computation of partial volume fat content. The correction scheme improved these results to, respectively, 81.5 and 107.1 g.