MRI simulation-based evaluation of image-processing and classification methods

OntoVIP: an ontology for the annotation of object models used for medical image simulation

This paper describes the creation of a comprehensive conceptualization of object models used in medical image simulation, suitable for major imaging modalities and simulators. The goal is to create an application ontology that can be used to annotate the models in a repository integrated in the Virtual Imaging Platform (VIP), to facilitate their sharing and reuse. Annotations make the anatomical, physiological and pathophysiological content of the object models explicit. In such an interdisciplinary context we chose to rely on a common integration framework provided by a foundational ontology, that facilitates the consistent integration of the various modules extracted from several existing ontologies, i.e. FMA, PATO, MPATH, RadLex and ChEBI. Emphasis is put on methodology for achieving this extraction and integration. The most salient aspects of the ontology are presented, especially the organization in model layers, as well as its use to browse and query the model repository.

New region-scalable discriminant and fitting energy functional for driving geometric active contours in medical image segmentation

We propose a novel region-based geometric active contour model that uses region-scalable discriminant and fitting energy functional for handling the intensity inhomogeneity and weak boundary problems in medical image segmentation. The region-scalable discriminant and fitting energy functional is defined to capture the image intensity characteristics in local and global regions for driving the evolution of active contour. The discriminant term in the model aims at separating background and foreground in scalable regions while the fitting term tends to fit the intensity in these regions. This model is then transformed into a variational level set formulation with a level set regularization term for accurate computation. The new model utilizes intensity information in the local and global regions as much as possible; so it not only handles better intensity inhomogeneity, but also allows more robustness to noise and more flexible initialization in comparison to the original global region and regional-scalable based models. Experimental results for synthetic and real medical image segmentation show the advantages of the proposed method in terms of accuracy and robustness.

Denoising MR images using non-local means filter with combined patch and pixel similarity

Denoising is critical for improving visual quality and reliability of associative quantitative analysis when magnetic resonance (MR) images are acquired with low signal-to-noise ratios. The classical non-local means (NLM) filter, which averages pixels weighted by the similarity of their neighborhoods, is adapted and demonstrated to effectively reduce Rician noise without affecting edge details in MR magnitude images. However, the Rician NLM (RNLM) filter usually blurs small high-contrast particle details which might be clinically relevant information. In this paper, we investigated the reason of this particle blurring problem and proposed a novel particle-preserving RNLM filter with combined patch and pixel (RNLM-CPP) similarity. The results of experiments on both synthetic and real MR data demonstrate that the proposed RNLM-CPP filter can preserve small high-contrast particle details better than the original RNLM filter while denoising MR images.

Computer simulation of magnetic resonance angiography imaging: model description and validation

With the development of medical imaging modalities and image processing algorithms, there arises a need for methods of their comprehensive quantitative evaluation. In particular, this concerns the algorithms for vessel tracking and segmentation in magnetic resonance angiography images. The problem can be approached by using synthetic images, where true geometry of vessels is known. This paper presents a framework for computer modeling of MRA imaging and the results of its validation. A new model incorporates blood flow simulation within MR signal computation kernel. The proposed solution is unique, especially with respect to the interface between flow and image formation processes. Furthermore it utilizes the concept of particle tracing. The particles reflect the flow of fluid they are immersed in and they are assigned magnetization vectors with temporal evolution controlled by MR physics. Such an approach ensures flexibility as the designed simulator is able to reconstruct flow profiles of any type. The proposed model is validated in a series of experiments with physical and digital flow phantoms. The synthesized 3D images contain various features (including artifacts) characteristic for the time-of-flight protocol and exhibit remarkable correlation with the data acquired in a real MR scanner. The obtained results support the primary goal of the conducted research, i.e. establishing a reference technique for a quantified validation of MR angiography image processing algorithms.

IMPROVING MAGNETIC RESONANCE RESOLUTION WITH SUPERVISED LEARNING

Despite ongoing improvements in magnetic resonance (MR) imaging (MRI), considerable clinical and, to a lesser extent, research data is acquired at lower resolutions. For example 1 mm isotropic acquisition of T₁-weighted (T₁-w) Magnetization Prepared Rapid Gradient Echo (MPRAGE) is standard practice, however T₂-weighted (T₂-w)-because of its longer relaxation times (and thus longer scan time)-is still routinely acquired with slice thicknesses of 2-5 mm and in-plane resolution of 2-3 mm. This creates obvious fundamental problems when trying to process T₁-w and T₂-w data in concert. We present an automated supervised learning algorithm to generate high resolution data. The framework is similar to the brain hallucination work of Rousseau, taking advantage of new developments in regression based image reconstruction. We present validation on phantom and real data, demonstrating the improvement over state-of-the-art super-resolution techniques.

MRISIMUL: a GPU-based parallel approach to MRI simulations

A new step-by-step comprehensive MR physics simulator (MRISIMUL) of the Bloch equations is presented. The aim was to develop a magnetic resonance imaging (MRI) simulator that makes no assumptions with respect to the underlying pulse sequence and also allows for complex large-scale analysis on a single computer without requiring simplifications of the MRI model. We hypothesized that such a simulation platform could be developed with parallel acceleration of the executable core within the graphic processing unit (GPU) environment. MRISIMUL integrates realistic aspects of the MRI experiment from signal generation to image formation and solves the entire complex problem for densely spaced isochromats and for a densely spaced time axis. The simulation platform was developed in MATLAB whereas the computationally demanding core services were developed in CUDA-C. The MRISIMUL simulator imaged three different computer models: a user-defined phantom, a human brain model and a human heart model. The high computational power of GPU-based simulations was compared against other computer configurations. A speedup of about 228 times was achieved when compared to serially executed C-code on the CPU whereas a speedup between 31 to 115 times was achieved when compared to the OpenMP parallel executed C-code on the CPU, depending on the number of threads used in multithreading (2-8 threads). The high performance of MRISIMUL allows its application in large-scale analysis and can bring the computational power of a supercomputer or a large computer cluster to a single GPU personal computer.

Augmented Lagrangian with variable splitting for faster non-Cartesian L1-SPIRiT MR image reconstruction

SPIRiT (iterative self-consistent parallel imaging reconstruction), and its sparsity-regularized variant L1-SPIRiT, are compatible with both Cartesian and non-Cartesian magnetic resonance imaging sampling trajectories. However, the non-Cartesian framework is more expensive computationally, involving a nonuniform Fourier transform with a nontrivial Gram matrix. We propose a novel implementation of the regularized reconstruction problem using variable splitting, alternating minimization of the augmented Lagrangian, and careful preconditioning. Our new method based on the alternating direction method of multipliers converges much faster than existing methods because of the preconditioners' heightened effectiveness. We demonstrate such rapid convergence substantially improves image quality for a fixed computation time. Our framework is a step forward towards rapid non-Cartesian L1-SPIRiT reconstructions.

SELF-SUPERVISED MRI TISSUE SEGMENTATION BY DISCRIMINATIVE CLUSTERING

The study of brain lesions can benefit from a clear identification of transitions between healthy and pathological tissues, through the analysis of brain imaging data. Current signal processing methods, able to address these issues, often rely on strong prior information. In this article, a new method for tissue segmentation is proposed. It is based on a discriminative strategy, in a self-supervised machine learning approach. This method avoids the use of prior information, which makes it very versatile, and able to cope with different tissue types. It also returns tissue probabilities for each voxel, crucial for a good characterization of the evolution of brain lesions. Simulated as well as real benchmark data were used to validate the accuracy of the method and compare it against other segmentation algorithms.

Development and evaluation of an automated histology classification system for veterinary pathology

A 2-stage algorithmic framework was developed to automatically classify digitized photomicrographs of tissues obtained from bovine liver, lung, spleen, and kidney into different histologic categories. The categories included normal tissue, acute necrosis, and inflammation (acute suppurative; chronic). In the current study, a total of 60 images per category (normal; acute necrosis; acute suppurative inflammation) were obtained from liver samples, 60 images per category (normal; acute suppurative inflammation) were obtained from spleen and lung samples, and 60 images per category (normal; chronic inflammation) were obtained from kidney samples. An automated support vector machine (SVM) classifier was trained to assign each test image to a specific category. Using 10 training images/category/organ, 40 test images/category/organ were examined. Employing confusion matrices to represent category-specific classification accuracy, the classifier-attained accuracies were found to be in the 74–90% range. The same set of test images was evaluated using a SVM classifier trained on 20 images/category/organ. The average classification accuracies were noted to be in the 84–95% range. The accuracy in correctly identifying normal tissue and specific tissue lesions was markedly improved by a small increase in the number of training images. The preliminary results from the study indicate the importance and potential use of automated image classification systems in the histologic identification of normal tissues and specific tissue lesions.

Uncertainty driven probabilistic voxel selection for image registration

This paper presents a novel probabilistic voxel selection strategy for medical image registration in time-sensitive contexts, where the goal is aggressive voxel sampling (e.g., using less than 1% of the total number) while maintaining registration accuracy and low failure rate. We develop a Bayesian framework whereby, first, a voxel sampling probability field (VSPF) is built based on the uncertainty on the transformation parameters. We then describe a practical, multi-scale registration algorithm, where, at each optimization iteration, different voxel subsets are sampled based on the VSPF. The approach maximizes accuracy without committing to a particular fixed subset of voxels. The probabilistic sampling scheme developed is shown to manage the tradeoff between the robustness of traditional random voxel selection (by permitting more exploration) and the accuracy of fixed voxel selection (by permitting a greater proportion of informative voxels).

Bloch-based MRI system simulator considering realistic electromagnetic fields for calculation of signal, noise, and specific absorption rate

PURPOSE

To describe and introduce new software capable of accurately simulating MR signal, noise, and specific absorption rate (SAR) given arbitrary sample, sequence, static magnetic field distribution, and radiofrequency magnetic and electric field distributions for each transmit and receive coil.

THEORY AND METHODS

Using fundamental equations for nuclear precession and relaxation, signal reception, noise reception, and calculation of SAR, a versatile MR simulator was developed. The resulting simulator was tested with simulation of a variety of sequences demonstrating several common imaging contrast types and artifacts. The simulation of intravoxel dephasing and rephasing with both tracking of the first order derivatives of each magnetization vector and multiple magnetization vectors was examined to ensure adequate representation of the MR signal. A quantitative comparison of simulated and experimentally measured SNR was also performed.

RESULTS

The simulator showed good agreement with our expectations, theory, and experiment.

CONCLUSION

With careful design, an MR simulator producing realistic signal, noise, and SAR for arbitrary sample, sequence, and fields has been created. It is hoped that this tool will be valuable in a wide variety of applications.

Multimodal medical volumetric data fusion using 3-D discrete shearlet transform and global-to-local rule

Traditional two-dimensional (2-D) fusion framework usually suffers from the loss of the between-slice information of the third dimension. For example, the fusion of three-dimensional (3-D) MRI slices must account for the information not only within the given slice but also the adjacent slices. In this paper, a fusion method is developed in 3-D shearlet space to overcome the drawback. On the other hand, the popularly used average-maximum fusion rule can capture only the local information but not any of the global information for it is implemented in a local window region. Thus, a global-to-local fusion rule is proposed. We firstly show the 3-D shearlet coefficients of the high-pass subbands are highly non-Gaussian. Then, we show this heavy-tailed phenomenon can be modeled by the generalized Gaussian density (GGD) and the global information between two subbands can be described by the Kullback-Leibler distance (KLD) of two GGDs. The finally fused global information can be selected according to the asymmetry of the KLD. Experiments on synthetic data and real data demonstrate that better fusion results can be obtained by the proposed method.

Automated material map generation from MRI scout pairs for preclinical PET attenuation correction

A novel method is presented to perform material map segmentation from preclinical MRI for corresponding PET attenuation correction. MRI does not provide attenuation ratio, hence segmenting a material map from it is challenging. Furthermore the MRI images often suffer from ghost artifacts. On the contrary MRI has no radiation dose. Our method operated with fast spin echo scout pairs that had perpendicular frequency directions. This way the direction of the ghost artifacts were perpendicular as well. Our body-air segmentation method built on this a priori information and successfully erased the ghost artifacts from the final binary mask. Visual and quantitative validation was performed by two preclinical specialists. Results indicate that our method is effective against MRI scout ghost artifacts and that PET attenuation correction based on MRI makes sense even on preclinical images.

Sparsity-promoting calibration for GRAPPA accelerated parallel MRI reconstruction

The amount of calibration data needed to produce images of adequate quality can prevent auto-calibrating parallel imaging reconstruction methods like generalized autocalibrating partially parallel acquisitions (GRAPPA) from achieving a high total acceleration factor. To improve the quality of calibration when the number of auto-calibration signal (ACS) lines is restricted, we propose a sparsity-promoting regularized calibration method that finds a GRAPPA kernel consistent with the ACS fit equations that yields jointly sparse reconstructed coil channel images. Several experiments evaluate the performance of the proposed method relative to unregularized and existing regularized calibration methods for both low-quality and underdetermined fits from the ACS lines. These experiments demonstrate that the proposed method, like other regularization methods, is capable of mitigating noise amplification, and in addition, the proposed method is particularly effective at minimizing coherent aliasing artifacts caused by poor kernel calibration in real data. Using the proposed method, we can increase the total achievable acceleration while reducing degradation of the reconstructed image better than existing regularized calibration methods.

MAGNETIC RESONANCE IMAGE SYNTHESIS THROUGH PATCH REGRESSION

Magnetic resonance imaging (MRI) is widely used for analyzing human brain structure and function. MRI is extremely versatile and can produce different tissue contrasts as required by the study design. For reasons such as patient comfort, cost, and improving technology, certain tissue contrasts for a cohort analysis may not have been acquired during the imaging session. This missing pulse sequence hampers consistent neuroanatomy research. One possible solution is to synthesize the missing sequence. This paper proposes a data-driven approach to image synthesis, which provides equal, if not superior synthesis compared to the state-of-the-art, in addition to being an order of magnitude faster. The synthesis transformation is done on image patches by a trained bagged ensemble of regression trees. Validation was done by synthesizing T₂-weighted contrasts from T₁-weighted scans, for phantoms and real data. We also synthesized 3 Tesla T₁-weighted magnetization prepared rapid gradient echo (MPRAGE) images from 1.5 Tesla MPRAGEs to demonstrate the generality of this approach.

Usability and potential of geostatistics for spatial discrimination of multiple sclerosis lesion patterns

BACKGROUND AND PURPOSE

In multiple sclerosis (MS) the individual disease courses are very heterogeneous among patients and biomarkers for setting the diagnosis and the estimation of the prognosis for individual patients would be very helpful. For this purpose, we are developing a multidisciplinary method and workflow for the quantitative, spatial, and spatiotemporal analysis and characterization of MS lesion patterns from MRI with geostatistics.

METHODS

We worked on a small data set involving three synthetic and three real-world MS lesion patterns, covering a wide range of possible MS lesion configurations. After brain normalization, MS lesions were extracted and the resulting binary 3-dimensional models of MS lesion patterns were subject to geostatistical indicator variography in three orthogonal directions.

RESULTS

By applying geostatistical indicator variography, we were able to describe the 3-dimensional spatial structure of MS lesion patterns in a standardized manner. Fitting a model function to the empirical variograms, spatial characteristics of the MS lesion patterns could be expressed and quantified by two parameters. An orthogonal plot of these parameters enabled a well-arranged comparison of the involved MS lesion patterns.

CONCLUSIONS

This method in development is a promising candidate to complement standard image-based statistics by incorporating spatial quantification. The work flow is generic and not limited to analyzing MS lesion patterns. It can be completely automated for the screening of radiological archives.

Closed-form relaxation for MRF-MAP tissue classification using discrete Laplace equations

While Markov random fields are very popular segmentation models in medical image processing, the associated maximum a posteriori (MAP) estimation problem is usually solved using iterative methods that are prone to local maxima. We show that a variant of the random walker algorithm can be seen as a relaxation method for the MAP problem under the Potts model. The key advantage of this technique is that it boils down to a sparse linear system with a uniquely defined explicit solution. Our experiments further demonstrate that the resulting MAP approximation can be used to improve the classical mean-field algorithm in terms of MAP estimation quality.

A virtual imaging platform for multi-modality medical image simulation

This paper presents the Virtual Imaging Platform (VIP), a platform accessible at http://vip.creatis.insa-lyon.fr to facilitate the sharing of object models and medical image simulators, and to provide access to distributed computing and storage resources. A complete overview is presented, describing the ontologies designed to share models in a common repository, the workflow template used to integrate simulators, and the tools and strategies used to exploit computing and storage resources. Simulation results obtained in four image modalities and with different models show that VIP is versatile and robust enough to support large simulations. The platform currently has 200 registered users who consumed 33 years of CPU time in 2011.

Multi-modal image registration based on gradient orientations of minimal uncertainty

In this paper, we propose a new multi-scale technique for multi-modal image registration based on the alignment of selected gradient orientations of reduced uncertainty. We show how the registration robustness and accuracy can be improved by restricting the evaluation of gradient orientation alignment to locations where the uncertainty of fixed image gradient orientations is minimal, which we formally demonstrate correspond to locations of high gradient magnitude. We also embed a computationally efficient technique for estimating the gradient orientations of the transformed moving image (rather than resampling pixel intensities and recomputing image gradients). We have applied our method to different rigid multi-modal registration contexts. Our approach outperforms mutual information and other competing metrics in the context of rigid multi-modal brain registration, where we show sub-millimeter accuracy with cases obtained from the retrospective image registration evaluation project. Furthermore, our approach shows significant improvements over standard methods in the highly challenging clinical context of image guided neurosurgery, where we demonstrate misregistration of less than 2 mm with relation to expert selected landmarks for the registration of pre-operative brain magnetic resonance images to intra-operative ultrasound images.

Improving SAR estimations in MRI using subject-specific models

To monitor and strategically control energy deposition in magnetic resonance imaging (MRI), measured as a specific absorption rate (SAR), numerical methods using generic human models have been employed to estimate worst-case values. Radiofrequency (RF) sequences are therefore often designed conservatively with large safety margins, potentially hindering the full potential of high-field systems. To more accurately predict the patient SAR values, we propose the use of image registration techniques, in conjunction with high-resolution image and tissue libraries, to create patient-specific voxel models. To test this, a matching model from the archives was first selected. Its tissue information was then warped to the patient's coordinates by registering the high-resolution library image to the pilot scan of the patient. Results from studying the models' 1 g SAR distribution suggest that the developed patient model can predict regions of elevated SAR within the patient with remarkable accuracy. Additionally, this work also proposes a voxel analytical metric that can assist in the construction of a patient library and the selection of the matching model from the library for a patient. It is hoped that, by developing voxel models with high accuracy in patient-specific anatomy and positioning, the proposed method can accurately predict the safety margins for high-field human applications and, therefore maximize the safe use of RF sequence power in high-field MRI systems.

Coaxially electrospun axon-mimicking fibers for diffusion magnetic resonance imaging

The study of brain structure and connectivity using diffusion magnetic resonance imaging (dMRI) has recently gained substantial interest. However, the use of dMRI still faces major challenges because of the lack of standard materials for validation. The present work reports on brain tissue-mimetic materials composed of hollow microfibers for application as a standard material in dMRI. These hollow fibers were fabricated via a simple and one-step coaxial electrospining (co-ES) process. Poly(ε-caprolactone) (PCL) and polyethylene oxide (PEO) were employed as shell and core materials, respectively, to achieve the most stable co-ES process. These co-ES hollow PCL fibers have different inner diameters, which mainly depend on the flow rate of the core solution and have the potential to cover the size range of the brain tissue we aimed to mimic. Co-ES aligned hollow PCL fibers were characterized using optical and electron microscopy and tested as brain white matter mimics on a high-field magnetic resonance imaging (MRI) scanner. To the best of our knowledge, this is the first time that co-ES hollow fibers have been successfully used as a tissue mimic or phantom in diffusion MRI. The results of the present study provide evidence that this phantom can mimic the dMRI behavior of cellular barriers imposed by axonal cell membranes and myelin; the measured diffusivity is compatible with that of in vivo biological tissues. Together these results suggest the potential use of co-ES hollow microfibers as tissue-mimicking phantoms in the field of medical imaging.

Efficient bias correction for magnetic resonance image denoising

Magnetic resonance imaging (MRI) is a popular radiology technique that is used for visualizing detailed internal structure of the body. Observed MRI images are generated by the inverse Fourier transformation from received frequency signals of a magnetic resonance scanner system. Previous research has demonstrated that random noise involved in the observed MRI images can be described adequately by the so-called Rician noise model. Under that model, the observed image intensity at a given pixel is a nonlinear function of the true image intensity and of two independent zero-mean random variables with the same normal distribution. Because of such a complicated noise structure in the observed MRI images, denoised images by conventional denoising methods are usually biased, and the bias could reduce image contrast and negatively affect subsequent image analysis. Therefore, it is important to address the bias issue properly. To this end, several bias-correction procedures have been proposed in the literature. In this paper, we study the Rician noise model and the corresponding bias-correction problem systematically and propose a new and more effective bias-correction formula based on the regression analysis and Monte Carlo simulation. Numerical studies show that our proposed method works well in various applications.

MR∕PET quantification tools: registration, segmentation, classification, and MR-based attenuation correction

PURPOSE

Combined MR∕PET is a relatively new, hybrid imaging modality. A human MR∕PET prototype system consisting of a Siemens 3T Trio MR and brain PET insert was installed and tested at our institution. Its present design does not offer measured attenuation correction (AC) using traditional transmission imaging. This study is the development of quantification tools including MR-based AC for quantification in combined MR∕PET for brain imaging.

METHODS

The developed quantification tools include image registration, segmentation, classification, and MR-based AC. These components were integrated into a single scheme for processing MR∕PET data. The segmentation method is multiscale and based on the Radon transform of brain MR images. It was developed to segment the skull on T1-weighted MR images. A modified fuzzy C-means classification scheme was developed to classify brain tissue into gray matter, white matter, and cerebrospinal fluid. Classified tissue is assigned an attenuation coefficient so that AC factors can be generated. PET emission data are then reconstructed using a three-dimensional ordered sets expectation maximization method with the MR-based AC map. Ten subjects had separate MR and PET scans. The PET with [(11)C]PIB was acquired using a high-resolution research tomography (HRRT) PET. MR-based AC was compared with transmission (TX)-based AC on the HRRT. Seventeen volumes of interest were drawn manually on each subject image to compare the PET activities between the MR-based and TX-based AC methods.

RESULTS

For skull segmentation, the overlap ratio between our segmented results and the ground truth is 85.2 ± 2.6%. Attenuation correction results from the ten subjects show that the difference between the MR and TX-based methods was <6.5%.

CONCLUSIONS

MR-based AC compared favorably with conventional transmission-based AC. Quantitative tools including registration, segmentation, classification, and MR-based AC have been developed for use in combined MR∕PET.

Accuracy and reliability of automated gray matter segmentation pathways on real and simulated structural magnetic resonance images of the human brain

Automated gray matter segmentation of magnetic resonance imaging data is essential for morphometric analyses of the brain, particularly when large sample sizes are investigated. However, although detection of small structural brain differences may fundamentally depend on the method used, both accuracy and reliability of different automated segmentation algorithms have rarely been compared. Here, performance of the segmentation algorithms provided by SPM8, VBM8, FSL and FreeSurfer was quantified on simulated and real magnetic resonance imaging data. First, accuracy was assessed by comparing segmentations of twenty simulated and 18 real T1 images with corresponding ground truth images. Second, reliability was determined in ten T1 images from the same subject and in ten T1 images of different subjects scanned twice. Third, the impact of preprocessing steps on segmentation accuracy was investigated. VBM8 showed a very high accuracy and a very high reliability. FSL achieved the highest accuracy but demonstrated poor reliability and FreeSurfer showed the lowest accuracy, but high reliability. An universally valid recommendation on how to implement morphometric analyses is not warranted due to the vast number of scanning and analysis parameters. However, our analysis suggests that researchers can optimize their individual processing procedures with respect to final segmentation quality and exemplifies adequate performance criteria.

Spatially regularized T(1) estimation from variable flip angles MRI

PURPOSE

To develop efficient algorithms for fast voxel-by-voxel quantification of tissue longitudinal relaxation time (T(1)) from variable flip angles magnetic resonance images (MRI) to reduce voxel-level noise without blurring tissue edges.

METHODS

T(1) estimations regularized by total variation (TV) and quadratic penalty are developed to measure T(1) from fast variable flip angles MRI and to reduce voxel-level noise without decreasing the accuracy of the estimates. First, a quadratic surrogate for a log likelihood cost function of T(1) estimation is derived based upon the majorization principle, and then the TV-regularized surrogate function is optimized by the fast iterative shrinkage thresholding algorithm. A fast optimization algorithm for the quadratically regularized T(1) estimation is also presented. The proposed methods are evaluated by the simulated and experimental MR data.

RESULTS

The means of the T(1) values in the simulated brain data estimated by the conventional, TV-regularized, and quadratically regularized methods have less than 3% error from the true T(1) in both GM and WM tissues with image noise up to 9%. The relative standard deviations (SDs) of the T(1) values estimated by the conventional method are more than 12% and 15% when the images have 7% and 9% noise, respectively. In comparison, the TV-regularized and quadratically regularized methods are able to suppress the relative SDs of the estimated T(1) to be less than 2% and 3%, respectively, regardless of the image noise level. However, the quadratically regularized method tends to overblur the edges compared to the TV-regularized method.

CONCLUSIONS

The spatially regularized methods improve quality of T(1) estimation from multiflip angles MRI. Quantification of dynamic contrast-enhanced MRI can benefit from the high quality measurement of native T(1).

3D image texture analysis of simulated and real-world vascular trees

A method is proposed for quantitative description of blood-vessel trees, which can be used for tree classification and/or physical parameters indirect monitoring. The method is based on texture analysis of 3D images of the trees. Several types of trees were defined, with distinct tree parameters (number of terminal branches, blood viscosity, input and output flow). A number of trees were computer-simulated for each type. 3D image was computed for each tree and its texture features were calculated. Best discriminating features were found and applied to 1-NN nearest neighbor classifier. It was demonstrated that (i) tree images can be correctly classified for realistic signal-to-noise ratio, (ii) some texture features are monotonously related to tree parameters, (iii) 2D texture analysis is not sufficient to represent the trees in the discussed sense. Moreover, applicability of texture model to quantitative description of vascularity images was also supported by unsupervised exploratory analysis. Eventually, the experimental confirmation was done, with the use of confocal microscopy images of rat brain vasculature. Several classes of brain tissue were clearly distinguished based on 3D texture numerical parameters, including control and different kinds of tumours - treated with NG2 proteoglycan to promote angiogenesis-dependent growth of the abnormal tissue. The method, applied to magnetic resonance imaging e.g. real neovasculature or retinal images can be used to support noninvasive medical diagnosis of vascular system diseases.

Mathematical modeling of human glioma growth based on brain topological structures: study of two clinical cases

Gliomas are the most common primary brain tumors and yet almost incurable due mainly to their great invasion capability. This represents a challenge to present clinical oncology. Here, we introduce a mathematical model aiming to improve tumor spreading capability definition. The model consists in a time dependent reaction-diffusion equation in a three-dimensional spatial domain that distinguishes between different brain topological structures. The model uses a series of digitized images from brain slices covering the whole human brain. The Talairach atlas included in the model describes brain structures at different levels. Also, the inclusion of the Brodmann areas allows prediction of the brain functions affected during tumor evolution and the estimation of correlated symptoms. The model is solved numerically using patient-specific parametrization and finite differences. Simulations consider an initial state with cellular proliferation alone (benign tumor), and an advanced state when infiltration starts (malign tumor). Survival time is estimated on the basis of tumor size and location. The model is used to predict tumor evolution in two clinical cases. In the first case, predictions show that real infiltrative areas are underestimated by current diagnostic imaging. In the second case, tumor spreading predictions were shown to be more accurate than those derived from previous models in the literature. Our results suggest that the inclusion of differential migration in glioma growth models constitutes another step towards a better prediction of tumor infiltration at the moment of surgical or radiosurgical target definition. Also, the addition of physiological/psychological considerations to classical anatomical models will provide a better and integral understanding of the patient disease at the moment of deciding therapeutic options, taking into account not only survival but also life quality.

Multi-Sectional Views Textural Based SVM for MS Lesion Segmentation in Multi-Channels MRIs

In this paper, a new technique is proposed for automatic segmentation of multiple sclerosis (MS) lesions from brain magnetic resonance imaging (MRI) data. The technique uses a trained support vector machine (SVM) to discriminate between the blocks in regions of MS lesions and the blocks in non-MS lesion regions mainly based on the textural features with aid of the other features. The classification is done on each of the axial, sagittal and coronal sectional brain view independently and the resultant segmentations are aggregated to provide more accurate output segmentation. The main contribution of the proposed technique described in this paper is the use of textural features to detect MS lesions in a fully automated approach that does not rely on manually delineating the MS lesions. In addition, the technique introduces the concept of the multi-sectional view segmentation to produce verified segmentation. The proposed textural-based SVM technique was evaluated using three simulated datasets and more than fifty real MRI datasets. The results were compared with state of the art methods. The obtained results indicate that the proposed method would be viable for use in clinical practice for the detection of MS lesions in MRI.

Multi-Sectional Views Textural Based SVM for MS Lesion Segmentation in Multi-Channels MRIs

In this paper, a new technique is proposed for automatic segmentation of multiple sclerosis (MS) lesions from brain magnetic resonance imaging (MRI) data. The technique uses a trained support vector machine (SVM) to discriminate between the blocks in regions of MS lesions and the blocks in non-MS lesion regions mainly based on the textural features with aid of the other features. The classification is done on each of the axial, sagittal and coronal sectional brain view independently and the resultant segmentations are aggregated to provide more accurate output segmentation. The main contribution of the proposed technique described in this paper is the use of textural features to detect MS lesions in a fully automated approach that does not rely on manually delineating the MS lesions. In addition, the technique introduces the concept of the multi-sectional view segmentation to produce verified segmentation. The proposed textural-based SVM technique was evaluated using three simulated datasets and more than fifty real MRI datasets. The results were compared with state of the art methods. The obtained results indicate that the proposed method would be viable for use in clinical practice for the detection of MS lesions in MRI.

Groupwise registration of multimodal images by an efficient joint entropy minimization scheme

Groupwise registration is concerned with bringing a group of images into the best spatial alignment. If images in the group are from different modalities, then the intensity correspondences across the images can be modeled by the joint density function (JDF) of the cooccurring image intensities. We propose a so-called treecode registration method for groupwise alignment of multimodal images that uses a hierarchical intensity-space subdivision scheme through which an efficient yet sufficiently accurate estimation of the (high-dimensional) JDF based on the Parzen kernel method is computed. To simultaneously align a group of images, a gradient-based joint entropy minimization was employed that also uses the same hierarchical intensity-space subdivision scheme. If the Hilbert kernel is used for the JDF estimation, then the treecode method requires no data-dependent bandwidth selection and is thus fully automatic. The treecode method was compared with the ensemble clustering (EC) method on four different publicly available multimodal image data sets and on a synthetic monomodal image data set. The obtained results indicate that the treecode method has similar and, for two data sets, even superior performances compared to the EC method in terms of registration error and success rate. The obtained good registration performances can be mostly attributed to the sufficiently accurate estimation of the JDF, which is computed through the hierarchical intensity-space subdivision scheme, that captures all the important features needed to detect the correct intensity correspondences across a multimodal group of images undergoing registration.

Improved DCT-based nonlocal means filter for MR images denoising

The nonlocal means (NLM) filter has been proven to be an efficient feature-preserved denoising method and can be applied to remove noise in the magnetic resonance (MR) images. To suppress noise more efficiently, we present a novel NLM filter based on the discrete cosine transform (DCT). Instead of computing similarity weights using the gray level information directly, the proposed method calculates similarity weights in the DCT subspace of neighborhood. Due to promising characteristics of DCT, such as low data correlation and high energy compaction, the proposed filter is naturally endowed with more accurate estimation of weights thus enhances denoising effectively. The performance of the proposed filter is evaluated qualitatively and quantitatively together with two other NLM filters, namely, the original NLM filter and the unbiased NLM (UNLM) filter. Experimental results demonstrate that the proposed filter achieves better denoising performance in MRI compared to the others.

Consensus embedding: theory, algorithms and application to segmentation and classification of biomedical data

Background

Dimensionality reduction (DR) enables the construction of a lower dimensional space (embedding) from a higher dimensional feature space while preserving object-class discriminability. However several popular DR approaches suffer from sensitivity to choice of parameters and/or presence of noise in the data. In this paper, we present a novel DR technique known as consensus embedding that aims to overcome these problems by generating and combining multiple low-dimensional embeddings, hence exploiting the variance among them in a manner similar to ensemble classifier schemes such as Bagging. We demonstrate theoretical properties of consensus embedding which show that it will result in a single stable embedding solution that preserves information more accurately as compared to any individual embedding (generated via DR schemes such as Principal Component Analysis, Graph Embedding, or Locally Linear Embedding). Intelligent sub-sampling (via mean-shift) and code parallelization are utilized to provide for an efficient implementation of the scheme.

Results

Applications of consensus embedding are shown in the context of classification and clustering as applied to: (1) image partitioning of white matter and gray matter on 10 different synthetic brain MRI images corrupted with 18 different combinations of noise and bias field inhomogeneity, (2) classification of 4 high-dimensional gene-expression datasets, (3) cancer detection (at a pixel-level) on 16 image slices obtained from 2 different high-resolution prostate MRI datasets. In over 200 different experiments concerning classification and segmentation of biomedical data, consensus embedding was found to consistently outperform both linear and non-linear DR methods within all applications considered.

Conclusions

We have presented a novel framework termed consensus embedding which leverages ensemble classification theory within dimensionality reduction, allowing for application to a wide range of high-dimensional biomedical data classification and segmentation problems. Our generalizable framework allows for improved representation and classification in the context of both imaging and non-imaging data. The algorithm offers a promising solution to problems that currently plague DR methods, and may allow for extension to other areas of biomedical data analysis.

Automatic skull stripping in MRI based on morphological filters and fuzzy c-means segmentation

In this paper a new automatic skull stripping method for T1-weighted MR image of human brain is presented. Skull stripping is a process that allows to separate the brain from the rest of tissues. The proposed method is based on a 2D brain extraction making use of fuzzy c-means segmentation and morphological operators applied on transversal slices. The approach is extended to the 3D case, taking into account the result obtained from the preceding slice to solve the organ splitting problem. The proposed approach is compared with BET (Brain Extraction Tool) implemented in MRIcro software.

Robust Student's-t mixture model with spatial constraints and its application in medical image segmentation

Finite mixture model based on the Student's-t distribution, which is heavily tailed and more robust than Gaussian, has recently received great attention for image segmentation. A new finite Student's-t mixture model (SMM) is proposed in this paper. Existing models do not explicitly incorporate the spatial relationships between pixels. First, our model exploits Dirichlet distribution and Dirichlet law to incorporate the local spatial constrains in an image. Secondly, we directly deal with the Student's-t distribution in order to estimate the model parameters, whereas, the Student's-t distributions in previous models are represented as an infinite mixture of scaled Gaussians that lead to an increase in complexity. Finally, instead of using expectation maximization (EM) algorithm, the proposed method adopts the gradient method to minimize the higher bound on the data negative log-likelihood and to optimize the parameters. The proposed model is successfully compared to the state-of-the-art finite mixture models. Numerical experiments are presented where the proposed model is tested on various simulated and real medical images.