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

Segmentation of MR Brain Images Through Hidden Markov Random Field and Hybrid Metaheuristic Algorithm

Image segmentation is one of the most critical tasks in Magnetic Resonance (MR) images analysis. Since the performance of most current image segmentation methods is suffered by noise and intensity non-uniformity artifact (INU), a precise and artifact resistant method is desired. In this work, we propose a new segmentation method combining a new Hidden Markov Random Field (HMRF) model and a novel hybrid metaheuristic method based on Cuckoo search (CS) and Particle swarm optimization algorithms (PSO). The new model uses adaptive parameters to allow balancing between the segmented components of the model. In addition, to improve the quality of searching solutions in the Maximum a posteriori (MAP) estimation of the HMRF model, the hybrid metaheuristic algorithm is introduced. This algorithm takes into account both the advantages of CS and PSO algorithms in searching ability by cooperating them with the same population in a parallel way and with a solution selection mechanism. Since CS and PSO are performing exploration and exploitation in the search space, respectively, hybridizing them in an intelligent way can provide better solutions in terms of quality. Furthermore, initialization of the population is carefully taken into account to improve the performance of the proposed method. The whole algorithm is evaluated on benchmark images including both the simulated and real MR brain images. Experimental results show that the proposed method can achieve satisfactory performance for images with noise and intensity inhomogeneity, and provides better results than its considered competitors.

An Overview on the Latest Nature-Inspired and Metaheuristics-Based Image Registration Algorithms

The development of automated image registration (IR) methods is a well-known issue within the computer vision (CV) field and it has been largely addressed from multiple viewpoints. IR has been applied to a high number of real-world scenarios ranging from remote sensing to medical imaging, artificial vision, and computer-aided design. In the last two decades, there has been an outstanding interest in the application of new optimization approaches for dealing with the main drawbacks present in the early IR methods, e.g., the Iterative Closest Point (ICP) algorithm. In particular, nature-inspired computation, e.g., evolutionary computation (EC), provides computational models that have their origin in evolution theories of nature. Moreover, other general purpose algorithms known as metaheuristics are also considered in this category of methods. Both nature-inspired and metaheuristic algorithms have been extensively adopted for tackling the IR problem, thus becoming a reliable alternative for optimization purposes. In this contribution, we aim to perform a comprehensive overview of the last decade (2009–2019) regarding the successful usage of this family of optimization approaches when facing the IR problem. Specifically, twenty-four methods (around 16 percent) of more than one hundred and fifty different contributions in the state-of-the-art have been selected. Several enhancements have been accordingly provided based on the promising outcomes shown by specific algorithmic designs. Finally, our research has shown that the field of nature-inspired and metaheuristic algorithms has increased its interest in the last decade to address the IR problem, and it has been highlighted that there is still room for improvement.

A Spatially Constrained Probabilistic Model for Robust Image Segmentation

In general, the hidden Markov random field (HMRF) represents the class label distribution of an image in probabilistic model based segmentation. The class label distributions provided by existing HMRF models consider either the number of neighboring pixels with similar class labels or the spatial distance of neighboring pixels with dissimilar class labels. Also, this spatial information is only considered for estimation of class labels of the image pixels, while its contribution in parameter estimation is completely ignored. This, in turn, deteriorates the parameter estimation, resulting in sub-optimal segmentation performance. Moreover, the existing models assign equal weightage to the spatial information for class label estimation of all pixels throughout the image, which, create significant misclassification for the pixels in boundary region of image classes. In this regard, the paper develops a new clique potential function and a new class label distribution, incorporating the information of image class parameters. Unlike existing HMRF model based segmentation techniques, the proposed framework introduces a new scaling parameter that adaptively measures the contribution of spatial information for class label estimation of image pixels. The importance of the proposed framework is depicted by modifying the HMRF based segmentation methods. The advantage of proposed class label distribution is also demonstrated irrespective of the underlying intensity distributions. The comparative performance of the proposed and existing class label distributions in HMRF model is demonstrated both qualitatively and quantitatively for brain MR image segmentation, HEp-2 cell delineation, natural image and object segmentation.

Probabilistic Graphical Models for Dynamic PET: A Novel Approach to Direct Parametric Map Estimation and Image Reconstruction

In the context of dynamic emission tomography, the conventional processing pipeline consists of independent image reconstruction of single-time frames, followed by the application of a suitable kinetic model to time-activity curves (TACs) at the voxel or region-of-interest level. Direct 4D positron emission tomography (PET) reconstruction, by contrast, seeks to move beyond this scheme and incorporate information from multiple time frames within the reconstruction task. Established direct methods are based on a deterministic description of voxelwise TACs, captured by the chosen kinetic model, considering the photon counting process the only source of uncertainty. In this paper, we introduce a new probabilistic modeling strategy based on the key assumption that activity time course would be subject to uncertainty even if the parameters of the underlying dynamic process are known. This leads to a hierarchical model that we formulate using the formalism of probabilistic graphical modeling. The inference is addressed using a new iterative algorithm, in which kinetic modeling results are treated as prior expectation of activity time course, rather than as a deterministic match, making it possible to control the trade-off between a data-driven and a model-driven reconstruction. The proposed method is flexible to an arbitrary choice of (linear and nonlinear) kinetic models, it enables the inclusion of arbitrary (sub)differentiable priors for parametric maps, and it is simple to implement. Computer simulations and an application to a real-patient scan show how the proposed method is able to generalize over conventional indirect and direct approaches, providing a bridge between them by properly tuning the impact of the kinetic modeling step on image reconstruction.

Measurements of cerebral blood volume using quantitative susceptibility mapping, R2 * relaxometry, and ferumoxytol-enhanced MRI

Ferumoxytol-enhanced MRI holds potential for the non-invasive assessment of vascular architecture using estimates of cerebral blood volume (CBV). Ferumoxytol specifically enables steady-state imaging with extended acquisition times, for substantial improvements in resolution and contrast-to-noise ratio. With such data, quantitative susceptibility mapping (QSM) can be used to obtain images of local tissue magnetic susceptibility and hence estimate the increase in blood susceptibility after administration of a contrast agent, which in turn can be correlated to tissue CBV. Here, we explore the use of QSM for CBV estimation and compare it with R₂ * (1/T₂ *)-based results. Institutional review board approval was obtained, and all subjects provided written informed consent. For this prospective study, MR images were acquired on a 3.0 T scanner in 19 healthy subjects using a multiple-echo T₂ *-weighted sequence. Scanning was performed before and after the administration of two doses of ferumoxytol (1 mg FE/kg and 4 mg FE/kg). Different QSM approaches were tested on numerical phantom simulations. Results showed that the accuracy of magnetic susceptibility measurements improved with increasing image resolution and decreasing vascular density. In vivo changes in magnetic susceptibility were measured after the administration of ferumoxytol utilizing QSM, and significantly higher QSM-based CBV was measured in gray matter compared with white matter. QSM- and R₂ *-based CBV estimates correlated well, with similar average values, but a larger variance was found in QSM-based estimates.

Multivariate Mixture Model for Myocardial Segmentation Combining Multi-Source Images

The author proposes a method for simultaneous registration and segmentation of multi-source images, using the multivariate mixture model (MvMM) and maximum of log-likelihood (LL) framework. Specifically, the method is applied to the problem of myocardial segmentation combining the complementary information from multi-sequence (MS) cardiac magnetic resonance (CMR) images. For the image misalignment and incongruent data, the MvMM is formulated with transformations and is further generalized for dealing with the hetero-coverage multi-modality images (HC-MMIs). The segmentation of MvMM is performed in a virtual common space, to which all the images and misaligned slices are simultaneously registered. Furthermore, this common space can be divided into a number of sub-regions, each of which contains congruent data, thus the HC-MMIs can be modeled using a set of conventional MvMMs. Results show that MvMM obtained significantly better performance compared to the conventional approaches and demonstrated good potential for scar quantification as well as myocardial segmentation. The generalized MvMM has also demonstrated better robustness in the incongruent data, where some images may not fully cover the region of interest, and the full coverage can only be reconstructed combining the images from multiple sources.

Four-angle method for practical ultra-high-resolution magnetic resonance mapping of brain longitudinal relaxation time and apparent proton density

Changes in longitudinal relaxation time (T₁) and proton density (PD) are sensitive indicators of microstructural alterations associated with various central nervous system diseases as well as brain maturation and aging. In this work, we introduce a new approach for rapid and accurate high-resolution (HR) or ultra HR (UHR) mapping of T₁ and apparent PD (APD) of the brain with correction of radiofrequency field, B₁, inhomogeneities. The four-angle method (FAM) uses four spoiled-gradient recalled-echo (SPGR) images acquired at different flip angles (FA) and short repetition times (TRs). The first two SPGR images are acquired at low-spatial resolution and used to accurately map the active B₁⁺ field with the recently introduced steady-state double angle method (SS-DAM). The estimated B₁⁺ map is used in conjunction with the two other SPGR images, acquired at HR or UHR, to map T₁ and APD. The method is evaluated with numerical, phantom, and in-vivo imaging measurements. Furthermore, we investigated imaging acceleration methods to further shorten the acquisition time. Our results indicate that FAM provides an accurate method for simultaneous HR or UHR mapping of T₁ and APD in human brain in clinical high-field MRI. Derived parameter maps without B₁⁺correction suffer from large inaccuracies, but this issue is well-corrected through use of the SS-DAM. Furthermore, the use of SPGR imaging with short TR and phased-array coil acquisition permits substantial imaging acceleration and enables robust HR or UHR T₁ and APD mapping in a clinically acceptable time frame, with whole brain coverage obtained in less than 2 min or 5 min, respectively. The method exhibits high reproducibility and benefits from the use of the conventional SPGR sequence, available in all preclinical and clinical MRI machines, and very simple modeling to address a critical outstanding issue in neuroimaging.

Understanding the Combined Effect of k -Space Undersampling and Transient States Excitation in MR Fingerprinting Reconstructions

Magnetic resonance fingerprinting (MRF) is able to estimate multiple quantitative tissue parameters from a relatively short acquisition. The main characteristic of an MRF sequence is the simultaneous application of 1) transient states excitation and 2) highly undersampled k -space. Despite the promising empirical results obtained with MRF, no work has appeared that formally describes the combined impact of these two aspects on the reconstruction accuracy. In this paper, a mathematical model is derived that directly relates the time-varying RF excitation and the k -space sampling to the spatially dependent reconstruction errors. A subsequent in-depth analysis identifies the mechanisms by which MRF sequence properties affect accuracy, providing a formal explanation of several empirically observed or intuitively understood facts. The new insights are obtained which show how this analytical framework could be used to improve the MRF protocol.

Systematic Differences Between Perceptually Relevant Image Statistics of Brain MRI and Natural Images

It is well-known that the human visual system is adapted to the statistical structure of natural scenes. Yet there are important classes of images - for example, medical images - that are not natural scenes, and therefore, that are expected to have statistical properties that deviate from the class of images that shaped the evolution and development of human vision. Here, focusing on structural brain MRI images, we quantify and characterize these deviations in terms of a set of local image statistics to which human visual sensitivity has been well-characterized, and that has previously been used for natural image analysis. We analyzed MRI images in multiple databases including T1-weighted and FLAIR sequence types, and simulated MRI images based on a published image simulation procedure for T1 images, which we also modified to generate FLAIR images. We first computed the power spectra of MRI images; spectral slopes were in the range -2.6 to -3.1 for T1 sequences, and -2.2 to -2.7 for FLAIR sequences. Analysis of local image statistics was then carried out on whitened images. For all of the databases as well as for the simulated images, we found that the three-point correlations contributed substantially to the differences between the "texture" of randomly selected ROIs. The informative nature of three-point correlations for brain MRI was greater than for natural images, and also disproportionate to human visual sensitivity. As this finding was consistent across databases, it is likely to result from brain geometry at the scale of brain MRI resolution, rather than characteristics of specific imaging and reconstruction methods.

A multi-objective optimization approach for brain MRI segmentation using fuzzy entropy clustering and region-based active contour methods

In this paper, we present a new multi-objective optimization approach for segmentation of Magnetic Resonance Imaging (MRI) of the human brain. The proposed algorithm not only takes advantages but also solves major drawbacks of two well-known complementary techniques, called fuzzy entropy clustering method and region-based active contour method, using multi-objective particle swarm optimization (MOPSO) approach. In order to obtain accurate segmentation results, firstly, two fitness functions with independent characteristics, compactness and separation, are derived from kernelized fuzzy entropy clustering with local spatial information and bias correction (KFECSB) and a novel adaptive energy weight combined with global and local fitting energy active contour (AWGLAC) model. Then, they are simultaneously optimized to finally produce a set of non-dominated solutions, from which L₂-metric method is used to select the best trade-off solution. Our algorithm is both verified and compared with other state-of-the-art methods using simulated MR images and real MR images from the McConnell Brain Imaging Center (BrainWeb) and the Internet Brain Segmentation Repository (IBSR), respectively. The experimental results demonstrate that the proposed technique achieves superior segmentation performance in terms of accuracy and robustness.

coreMRI: A high-performance, publicly available MR simulation platform on the cloud

Introduction

A Cloud-ORiented Engine for advanced MRI simulations (coreMRI) is presented in this study. The aim was to develop the first advanced MR simulation platform delivered as a web service through an on-demand, scalable cloud-based and GPU-based infrastructure. We hypothesized that such an online MR simulation platform could be utilized as a virtual MRI scanner but also as a cloud-based, high-performance engine for advanced MR simulations in simulation-based quantitative MR (qMR) methods.

Methods and results

The simulation framework of coreMRI was based on the solution of the Bloch equations and utilized a ground-up-approach design based on the principles already published in the literature. The development of a front-end environment allowed the connection of the end-users to the GPU-equipped instances on the cloud. The coreMRI simulation platform was based on a modular design where individual modules (such as the Gadgetron reconstruction framework and a newly developed Pulse Sequence Designer) could be inserted in the main simulation framework. Different types and sources of pulse sequences and anatomical models were utilized in this study revealing the flexibility that the coreMRI simulation platform offers to the users. The performance and scalability of coreMRI were also examined on multi-GPU configurations on the cloud, showing that a multi-GPU computer on the cloud equipped with a newer generation of GPU cards could significantly mitigate the prolonged execution times that accompany more realistic MRI and qMR simulations.

Conclusions

coreMRI is available to the entire MR community, whereas its high performance and scalability allow its users to configure advanced MRI experiments without the constraints imposed by experimentation in a true MRI scanner (such as time constraint and limited availability of MR scanners), without upfront investment for purchasing advanced computer systems and without any user expertise on computer programming or MR physics. coreMRI is available to the users through the webpage https://www.coreMRI.org.

Real-Time Filtering with Sparse Variations for Head Motion in Magnetic Resonance Imaging

Estimating a time-varying signal, such as head motion from magnetic resonance imaging data, becomes particularly challenging in the face of other temporal dynamics such as functional activation. This paper describes a new Kalman filter-like framework that includes a sparse residual term in the measurement model. This additional term allows the extended Kalman filter to generate real-time motion estimates suitable for prospective motion correction when such dynamics occur. An iterative augmented Lagrangian algorithm similar to the alterating direction method of multipliers implements the update step for this Kalman filter. This paper evaluates the accuracy and convergence rate of this iterative method for small and large motion in terms of its sensitivity to parameter selection. The included experiment on a simulated functional magnetic resonance imaging acquisition demonstrates that the resulting method improves the maximum Youden's J index of the time series analysis by 2-3% versus retrospective motion correction, while the sensitivity index increases from 4.3 to 5.4 when combining prospective and retrospective correction.

Steady-state double-angle method for rapid B1 mapping

PURPOSE

To introduce an accurate, rapid, and practical method for active B₁ field mapping based on the double-angle method (DAM) in the steady-state (SS) signal regime.

METHODS

We introduced and evaluated the performance of the SS-DAM approach to map the B₁ field and compared the results to those calculated from the conventional DAM approach. Similar to DAM, SS-DAM uses the signal intensity ratio of 2 magnitude images acquired with different flip angles using the spoiled gradient recalled echo sequence. However, unlike DAM, in SS-DAM, these 2 spoiled gradient recalled echo images are acquired with very short TR, which allows substantially reduced acquisition time. Numerical, phantom, and in vivo brain imaging analyses, representing a wide range of T₁ s and large B₁ variation, were conducted. Methods for further accelerating acquisition were also investigated.

RESULTS

Our results demonstrate the potential of the SS-DAM approach to be applied widely in the clinical setting. B₁ maps derived from SS-DAM were demonstrated to be quantitatively comparable to those derived from DAM but were derived much more rapidly. Large-volume B₁ maps were obtained at a field strength of 3 tesla within clinically acceptable acquisition times.

CONCLUSION

SS-DAM permits accurate B₁ mapping in the clinical setting, with whole-brain coverage in less than 1 min.

Advances in Computational Human Phantoms and Their Applications in Biomedical Engineering - A Topical Review

Over the past decades, significant improvements have been made in the field of computational human phantoms (CHPs) and their applications in biomedical engineering. Their sophistication has dramatically increased. The very first CHPs were composed of simple geometric volumes, e.g., cylinders and spheres, while current CHPs have a high resolution, cover a substantial range of the patient population, have high anatomical accuracy, are poseable, morphable, and are augmented with various details to perform functionalized computations. Advances in imaging techniques and semi-automated segmentation tools allow fast and personalized development of CHPs. These advances open the door to quickly develop personalized CHPs, inherently including the disease of the patient. Because many of these CHPs are increasingly providing data for regulatory submissions of various medical devices, the validity, anatomical accuracy, and availability to cover the entire patient population is of utmost importance. The article is organized into two main sections: the first section reviews the different modeling techniques used to create CHPs, whereas the second section discusses various applications of CHPs in biomedical engineering. Each topic gives an overview, a brief history, recent developments, and an outlook into the future.

Multi-Modal Medical Image Registration with Full or Partial Data: A Manifold Learning Approach

Multi-modal image registration is the primary step in integrating information stored in two or more images, which are captured using multiple imaging modalities. In addition to intensity variations and structural differences between images, they may have partial or full overlap, which adds an extra hurdle to the success of registration process. In this contribution, we propose a multi-modal to mono-modal transformation method that facilitates direct application of well-founded mono-modal registration methods in order to obtain accurate alignment of multi-modal images in both cases, with complete (full) and incomplete (partial) overlap. The proposed transformation facilitates recovering strong scales, rotations, and translations. We explain the method thoroughly and discuss the choice of parameters. For evaluation purposes, the effectiveness of the proposed method is examined and compared with widely used information theory-based techniques using simulated and clinical human brain images with full data. Using RIRE dataset, mean absolute error of 1.37, 1.00, and 1.41 mm are obtained for registering CT images with PD-, T1-, and T2-MRIs, respectively. In the end, we empirically investigate the efficacy of the proposed transformation in registering multi-modal partially overlapped images.

Intuitionistic Center-Free FCM Clustering for MR Brain Image Segmentation

In this paper, an intuitionistic center-free fuzzy c-means clustering method (ICFFCM) is proposed for magnetic resonance (MR) brain image segmentation. First, in order to suppress the effect of noise in MR brain images, a pixel-to-pixel similarity with spatial information is defined. Then, for the purpose of handling the vagueness in MR brain images as well as the uncertainty in clustering process, a pixel-to-cluster similarity measure is defined by employing the intuitionistic fuzzy membership function. These two similarities are used to modify the center-free FCM so that the ability of the method for MR brain image segmentation could be improved. Second, on the basis of the improved center-free FCM method, a local information term, which is also intuitionistic and center-free, is appended to the objective function. This generates the final proposed ICFFCM. The consideration of local information further enhances the robustness of ICFFCM to the noise in MR brain images. Experimental results on the simulated and real MR brain image datasets show that ICFFCM is effective and robust. Moreover, ICFFCM could outperform several fuzzy-clustering-based methods and could achieve comparable results to the standard published methods like statistical parametric mapping and FMRIB automated segmentation tool.

A Fast GPU-optimized 3D MRI Simulator for Arbitrary k-space Sampling

Purpose:

To develop a fast 3D MRI simulator for arbitrary k-space sampling using a graphical processing unit (GPU) and demonstrate its performance by comparing simulation and experimental results in a real MRI system.

Materials and Methods:

A fast 3D MRI simulator using a GeForce GTX 1080 GPU (NVIDIA Corporation, Santa Clara, CA, USA) was developed using C++ and the CUDA 8.0 platform (NVIDIA Corporation). The unique advantage of this simulator was that it could use the same pulse sequence as used in the experiment. The performance of the MRI simulator was measured using two GTX 1080 GPUs and 3D Cones sequences. The MRI simulation results for 3D non-Cartesian sampling trajectories like 3D Cones sequences using a numerical 3D phantom were compared with the experimental results obtained with a real MRI system and a real 3D phantom.

Results:

The performance of the MRI simulator was about 3800–4900 gigaflops for 128- to 4-shot 3D Cones sequences with 256³ voxels, which was about 60% of the performance of the previous MRI simulator optimized for Cartesian sampling calculated for a Cartesian sampling gradient-echo sequence with 256³ voxels. The effects of the static magnetic field inhomogeneity, radio-frequency field inhomogeneity, gradient field nonlinearity, and fast repetition times on the MR images were reproduced in the simulated images as observed in the experimental images.

Conclusion:

The 3D MRI simulator developed for arbitrary k-space sampling optimized using GPUs is a powerful tool for the development and evaluation of advanced imaging sequences including both Cartesian and non-Cartesian k-space sampling.

Iterative spatial fuzzy clustering for 3D brain magnetic resonance image supervoxel segmentation

BACKGROUND

Although supervoxel segmentation methods have been employed for brain Magnetic Resonance Image (MRI) processing and analysis, due to the specific features of the brain, including complex-shaped internal structures and partial volume effect, their performance remains unsatisfactory.

NEW METHODS

To address these issues, this paper presents a novel iterative spatial fuzzy clustering (ISFC) algorithm to generate 3D supervoxels for brain MRI volume based on prior knowledge. This work makes use of the common topology among the human brains to obtain a set of seed templates from a population-based brain template MRI image. After selecting the number of supervoxels, the corresponding seed template is projected onto the considered individual brain for generating reliable seeds. Then, to deal with the influence of partial volume effect, an efficient iterative spatial fuzzy clustering algorithm is proposed to allocate voxels to the seeds and to generate the supervoxels for the overall brain MRI volume.

RESULTS

The performance of the proposed algorithm is evaluated on two widely used public brain MRI datasets and compared with three other up-to-date methods.

CONCLUSIONS

The proposed algorithm can be utilized for several brain MRI processing and analysis, including tissue segmentation, tumor detection and segmentation, functional parcellation and registration.

Content-aware compressive magnetic resonance image reconstruction

This paper describes an adaptive approach to regularizing model-based reconstructions in magnetic resonance imaging to account for local structure or image content. In conjunction with common models like wavelet and total variation sparsity, this content-aware regularization avoids oversmoothing or compromising image features while suppressing noise and incoherent aliasing from accelerated imaging. To evaluate this regularization approach, the experiments reconstruct images from single- and multi-channel, Cartesian and non-Cartesian, brain and cardiac data. These reconstructions combine common analysis-form regularizers and autocalibrating parallel imaging (when applicable). In most cases, the results show widespread improvement in structural similarity and peak-signal-to-error ratio relative to the fully sampled images. These results suggest that this content-aware regularization can preserve local image structures such as edges while providing denoising power superior to sparsity-promoting or sparsity-reweighted regularization.

A simple and fast adaptive nonlocal multispectral filtering algorithm for efficient noise reduction in magnetic resonance imaging

PURPOSE

We recently introduced a multispectral (MS) nonlocal (NL) filter based on maximum likelihood estimation (MLE) of voxel intensities, termed MS-NLML. While MS-NLML provides excellent noise reduction and improved image feature preservation as compared to other NL or MS filters, it requires considerable processing time, limiting its application in routine analyses. In this work, we introduced a fast, simple, and robust filter, termed nonlocal estimation of multispectral magnitudes (NESMA), for noise reduction in multispectral (MS) magnetic resonance imaging (MRI).

METHODS

Through extensive simulation and in-vivo analyses, we compared the performance of NESMA and MS-NLML in terms of noise reduction and processing efficiency. Further, we introduce two simple adaptive methods that permit spatial variation of similar voxels, R, used in the filtering. The first method is semi-adaptive and permits variation of R across the image by using a relative Euclidean distance (RED) similarity threshold. The second method is fully adaptive and filters the raw data with several RED similarity thresholds to spatially determine the optimal threshold value using an unbiased criterion.

RESULTS

NESMA shows very similar filtering performance as compared to MS-NLML, however, with much simple implementation and very fast processing time. Further, for both filters, the adaptive methods were shown to further reduce noise in comparison with the conventional non-adaptive method in which R is set to a constant value throughout the image.

CONCLUSIONS

NESMA is fast, robust, and straightforward to implement filter. These features render it suitable for routine clinical use and analysis of large MRI datasets.

Weighted Schatten p-norm minimization for 3D magnetic resonance images denoising

Magnetic resonance (MR) imaging plays an important role in clinical diagnosis and scientific research. A clean MR image can better provide patient's information to doctors or researchers for further treatment. However, in real life, MR images are inevitably corrupted by annoying Rician noise in the process of imaging. Aiming at the Rician noise of 3D MR images, a framework is proposed to suppress noise by low-rank matrix approximation (LRMA) with weighted Schatten p-norm minimization regularization (WSNMD-3D). The proposed method not only considers the importance of different rank components, but can also approximate the true rank of the latent low-rank matrix. This approach first groups similar non-local cubic patches extracted from the noisy 3D MR image into a matrix whose columns are vectorized patches. The above matrix can be modeled as a low-rank matrix approximate model. Then weighted Schatten p-norm minimization (WSNM) is applied to the model, which shrinks different rank components with different treatments. Finally, the denoised 3D MR image is acquired by aggregating all denoised patches with weighted averaging. Experimental results on synthetic and real 3D MR data show that the proposed method obtains better results than state-of-the-art methods, both visually and quantitatively.

Medical Image Blind Integrity Verification with Krawtchouk Moments

A new blind integrity verification method for medical image is proposed in this paper. It is based on a new kind of image features, known as Krawtchouk moments, which we use to distinguish the original images from the modified ones. Basically, with our scheme, image integrity verification is accomplished by classifying images into the original and modified categories. Experiments conducted on medical images issued from different modalities verified the validity of the proposed method and demonstrated that it can be used to detect and discriminate image modifications of different types with high accuracy. We also compared the performance of our scheme with a state-of-the-art solution suggested for medical images-solution that is based on histogram statistical properties of reorganized block-based Tchebichef moments. Conducted tests proved the better behavior of our image feature set.

Rapid B1 field mapping at 3 T using the 180° signal null method with extended flip angle

PURPOSE

To extend the null signal method (NSM) for B₁ mapping to 3 T magnetic resonance imaging (MRI).

BACKGROUND

The NSM operates in the steady state regime and exploits the linearity of the spoiled gradient recalled echo (SPGR) signal around the 180° flip angle (FA). Using linear regression, B₁ maps are derived from three SPGR images acquired at different FAs with a short repetition time. While the conventional NSM allows accurate mapping of B₁ for moderate B₁ variation, we observed that this method fails for the larger B₁ variations typical of high-field MRI.

METHODS

We analyzed the effect of the FA range of the acquired SPGR images on B₁ determination using the NSM for 3 T MRI through extensive numerical and in vivo analyses. B₁ maps derived from the extended angle-range NSM (EA-NSM) were calculated and compared to those derived from the conventional, more restricted angle range, NSM, and to those derived from the reference, but much more time-consuming, double angle method (DAM). Furthermore, we investigated the compatibility of EA-NSM B₁ mapping and the half-scan and SENSE reconstruction methods for accelerating acquisition time.

RESULTS

Our results show that the use of the conventional FA range leads to substantial inaccuracies in B₁ determination. Both numerical and in vivo analyses demonstrate that expanding the FA range of the acquired SPGR images substantially improves the accuracy of B₁ maps. Furthermore, B₁ maps derived from EA-NSM were demonstrated to be quantitatively comparable to those derived from the lengthy DAM protocol. We also found that B₁ maps derived from SPGR images using the EA-NSM and imaging acceleration methods were comparable to those derived from images acquired without acceleration. Finally, the use of half scanning combined with SENSE reconstruction permits whole-brain B₁ mapping in ~1 min.

CONCLUSIONS

The EA-NSM permits accurate, fast, and practical B₁ mapping in a 3 T clinical setting.

Automatic brain tissue segmentation based on graph filter

Background

Accurate segmentation of brain tissues from magnetic resonance imaging (MRI) is of significant importance in clinical applications and neuroscience research. Accurate segmentation is challenging due to the tissue heterogeneity, which is caused by noise, bias filed and partial volume effects.

Methods

To overcome this limitation, this paper presents a novel algorithm for brain tissue segmentation based on supervoxel and graph filter. Firstly, an effective supervoxel method is employed to generate effective supervoxels for the 3D MRI image. Secondly, the supervoxels are classified into different types of tissues based on filtering of graph signals.

Results

The performance is evaluated on the BrainWeb 18 dataset and the Internet Brain Segmentation Repository (IBSR) 18 dataset. The proposed method achieves mean dice similarity coefficient (DSC) of 0.94, 0.92 and 0.90 for the segmentation of white matter (WM), grey matter (GM) and cerebrospinal fluid (CSF) for BrainWeb 18 dataset, and mean DSC of 0.85, 0.87 and 0.57 for the segmentation of WM, GM and CSF for IBSR18 dataset.

Conclusions

The proposed approach can well discriminate different types of brain tissues from the brain MRI image, which has high potential to be applied for clinical applications.

Joint sparse reconstruction of multi-contrast MRI images with graph based redundant wavelet transform

Background

Multi-contrast images in magnetic resonance imaging (MRI) provide abundant contrast information reflecting the characteristics of the internal tissues of human bodies, and thus have been widely utilized in clinical diagnosis. However, long acquisition time limits the application of multi-contrast MRI. One efficient way to accelerate data acquisition is to under-sample the k-space data and then reconstruct images with sparsity constraint. However, images are compromised at high acceleration factor if images are reconstructed individually. We aim to improve the images with a jointly sparse reconstruction and Graph-based redundant wavelet transform (GBRWT).

Methods

First, a sparsifying transform, GBRWT, is trained to reflect the similarity of tissue structures in multi-contrast images. Second, joint multi-contrast image reconstruction is formulated as a ℓ_{2, 1} norm optimization problem under GBRWT representations. Third, the optimization problem is numerically solved using a derived alternating direction method.

Results

Experimental results in synthetic and in vivo MRI data demonstrate that the proposed joint reconstruction method can achieve lower reconstruction errors and better preserve image structures than the compared joint reconstruction methods. Besides, the proposed method outperforms single image reconstruction with joint sparsity constraint of multi-contrast images.

Conclusions

The proposed method explores the joint sparsity of multi-contrast MRI images under graph-based redundant wavelet transform and realizes joint sparse reconstruction of multi-contrast images. Experiment demonstrate that the proposed method outperforms the compared joint reconstruction methods as well as individual reconstructions. With this high quality image reconstruction method, it is possible to achieve the high acceleration factors by exploring the complementary information provided by multi-contrast MRI.

Simulation of phase contrast angiography for renal arterial models

BACKGROUND

With the development of versatile magnetic resonance acquisition techniques there arises a need for more advanced imaging simulation tools to enable adequate image appearance prediction, measurement sequence design and testing thereof. Recently, there is a growing interest in phase contrast angiography (PCA) sequence due to the capabilities of blood flow quantification that it offers. Moreover, as it is a non-contrast enhanced protocol, it has become an attractive option in areas, where usage of invasive contrast agents is not indifferent for the imaged tissue. Monitoring of the kidney function is an example of such an application.

RESULTS

We present a computer framework for simulation of the PCA protocol, both conventional and accelerated with echo-planar imaging (EPI) readout, and its application to the numerical models of kidney vasculatures. Eight patient-specific renal arterial trees were reconstructed following vessel segmentation in real computed tomography angiograms. In addition, a synthetic model was designed using a vascular tree growth simulation algorithm. The results embrace a series of synthetic PCA images of the renal arterial trees giving insight into the image formation and quantification of kidney hemodynamics.

CONCLUSIONS

The designed simulation framework enables quantification of the PCA measurement error in relation to ground-truth flow velocity data. The mean velocity measurement error for the reconstructed renal arterial trees range from 1.5 to 12.8% of the aliasing velocity value, depending on image resolution and flip angle. No statistically significant difference was observed between measurements obtained using EPI with a number of echos (NETL) = 4 and conventional PCA. In case of higher NETL factors peak velocity values can be underestimated up to 34%.

Direct Patlak Reconstruction From Dynamic PET Data Using the Kernel Method With MRI Information Based on Structural Similarity

Positron emission tomography (PET) is a functional imaging modality widely used in oncology, cardiology, and neuroscience. It is highly sensitive, but suffers from relatively poor spatial resolution, as compared with anatomical imaging modalities, such as magnetic resonance imaging (MRI). With the recent development of combined PET/MR systems, we can improve the PET image quality by incorporating MR information into image reconstruction. Previously, kernel learning has been successfully embedded into static and dynamic PET image reconstruction using either PET temporal or MRI information. Here, we combine both PET temporal and MRI information adaptively to improve the quality of direct Patlak reconstruction. We examined different approaches to combine the PET and MRI information in kernel learning to address the issue of potential mismatches between MRI and PET signals. Computer simulations and hybrid real-patient data acquired on a simultaneous PET/MR scanner were used to evaluate the proposed methods. Results show that the method that combines PET temporal information and MRI spatial information adaptively based on the structure similarity index has the best performance in terms of noise reduction and resolution improvement.

Model-Based Generation of Large Databases of Cardiac Images: Synthesis of Pathological Cine MR Sequences From Real Healthy Cases

Collecting large databases of annotated medical images is crucial for the validation and testing of feature extraction, statistical analysis, and machine learning algorithms. Recent advances in cardiac electromechanical modeling and image synthesis provided a framework to generate synthetic images based on realistic mesh simulations. Nonetheless, their potential to augment an existing database with large amounts of synthetic cases requires further investigation. We build upon these works and propose a revised scheme for synthesizing pathological cardiac sequences from real healthy sequences. Our new pipeline notably involves a much easier registration problem to reduce potential artifacts, and takes advantage of mesh correspondences to generate new data from a given case without additional registration. The output sequences are thoroughly examined in terms of quality and usability on a given application: the assessment of myocardial viability, via the generation of 465 synthetic cine MR sequences (15 healthy and 450 with pathological tissue viability [random location, extent, and grade, up to myocardial infarct]). We demonstrate that: 1) our methodology improves the state-of-the-art algorithms in terms of realism and accuracy of the simulated images and 2) our methodology is well-suited for the generation of large databases at small computational cost.

Application of the 4-D XCAT Phantoms in Biomedical Imaging and Beyond

The four-dimensional (4-D) eXtended CArdiac-Torso (XCAT) series of phantoms was developed to provide accurate computerized models of the human anatomy and physiology. The XCAT series encompasses a vast population of phantoms of varying ages from newborn to adult, each including parameterized models for the cardiac and respiratory motions. With great flexibility in the XCAT's design, any number of body sizes, different anatomies, cardiac or respiratory motions or patterns, patient positions and orientations, and spatial resolutions can be simulated. As such, the XCAT phantoms are gaining a wide use in biomedical imaging research. There they can provide a virtual patient base from which to quantitatively evaluate and improve imaging instrumentation, data acquisition, techniques, and image reconstruction and processing methods which can lead to improved image quality and more accurate clinical diagnoses. The phantoms have also found great use in radiation dosimetry, radiation therapy, medical device design, and even the security and defense industry. This review paper highlights some specific areas in which the XCAT phantoms have found use within biomedical imaging and other fields. From these examples, we illustrate the increasingly important role that computerized phantoms and computer simulation are playing in the research community.

Tensor Sparse Representation for 3-D Medical Image Fusion Using Weighted Average Rule

OBJECTIVE

The technique of fusing multimodal medical images into single image has a great impact on the clinical diagnosis. The previous works mostly concern the two-dimensional (2-D) image fusion performed on each slice individually, that may destroy the 3-D correlation across adjacent slices. To address this issue, this paper proposes a novel 3-D image fusion scheme based on Tensor Sparse Representation (TSR).

METHODS

First, each medical volume is arranged as a three-order tensor, and represented by TSR with learned dictionaries. Second, a novel "weighted average" rule, calculated from the tensor sparse coefficients using 3-D local-to-global strategy. The weights are then employed to combine the multimodal medical volumes through weighted average.

RESULTS

The visual and objective comparisons show that the proposed method is competitive to the existing methods on various medical volumes in different imaging modalities.

CONCLUSION

The TSR-based 3-D fusion approach with weighted average rule can preserve the 3-D structure of medical volume, and reduce the low contrast and artifacts in fused product.

SIGNIFICANCE

The designed weights offer the effective assigned weights and accurate salience levels measure, which can improve the performance of fusion approach.

Flow MRI simulation in complex 3D geometries: Application to the cerebral venous network

PURPOSE

Develop and evaluate a complete tool to include 3D fluid flows in MRI simulation, leveraging from existing software. Simulation of MR spin flow motion is of high interest in the study of flow artifacts and angiography. However, at present, only a few simulators include this option and most are restricted to static tissue imaging.

THEORY AND METHODS

An extension of JEMRIS, one of the most advanced high performance open-source simulation platforms to date, was developed. The implementation of a Lagrangian description of the flow allows simulating any MR experiment, including both static tissues and complex flow data from computational fluid dynamics. Simulations of simple flow models are compared with real experiments on a physical flow phantom. A realistic simulation of 3D flow MRI on the cerebral venous network is also carried out.

RESULTS

Simulations and real experiments are in good agreement. The generality of the framework is illustrated in 2D and 3D with some common flow artifacts (misregistration and inflow enhancement) and with the three main angiographic techniques: phase contrast velocimetry (PC), time-of-flight, and contrast-enhanced imaging MRA.

CONCLUSION

The framework provides a versatile and reusable tool for the simulation of any MRI experiment including physiological fluids and arbitrarily complex flow motion.

Simultaneous segmentation and bias field estimation using local fitted images

Level set methods often suffer from boundary leakage and inadequate segmentation when used to segment images with inhomogeneous intensities. To handle this issue, a novel region-based level set method was developed, in which two different local fitted images are used to construct a hybrid region intensity fitting energy functional. This novel method enables simultaneous segmentation of the regions of interest and estimation of the bias fields from inhomogeneous images. Our experiments on both synthetic images and a publicly available dataset demonstrate the feasibility and reliability of the proposed method.

Functional Brain Imaging Synthesis Based on Image Decomposition and Kernel Modeling: Application to Neurodegenerative Diseases

The rise of neuroimaging in research and clinical practice, together with the development of new machine learning techniques has strongly encouraged the Computer Aided Diagnosis (CAD) of different diseases and disorders. However, these algorithms are often tested in proprietary datasets to which the access is limited and, therefore, a direct comparison between CAD procedures is not possible. Furthermore, the sample size is often small for developing accurate machine learning methods. Multi-center initiatives are currently a very useful, although limited, tool in the recruitment of large populations and standardization of CAD evaluation. Conversely, we propose a brain image synthesis procedure intended to generate a new image set that share characteristics with an original one. Our system focuses on nuclear imaging modalities such as PET or SPECT brain images. We analyze the dataset by applying PCA to the original dataset, and then model the distribution of samples in the projected eigenbrain space using a Probability Density Function (PDF) estimator. Once the model has been built, we can generate new coordinates on the eigenbrain space belonging to the same class, which can be then projected back to the image space. The system has been evaluated on different functional neuroimaging datasets assessing the: resemblance of the synthetic images with the original ones, the differences between them, their generalization ability and the independence of the synthetic dataset with respect to the original. The synthetic images maintain the differences between groups found at the original dataset, with no significant differences when comparing them to real-world samples. Furthermore, they featured a similar performance and generalization capability to that of the original dataset. These results prove that these images are suitable for standardizing the evaluation of CAD pipelines, and providing data augmentation in machine learning systems -e.g. in deep learning-, or even to train future professionals at medical school.

Robust generative asymmetric GMM for brain MR image segmentation

BACKGROUND AND OBJECTIVES

Accurate segmentation of brain tissues from magnetic resonance (MR) images based on the unsupervised statistical models such as Gaussian mixture model (GMM) has been widely studied during last decades. However, most GMM based segmentation methods suffer from limited accuracy due to the influences of noise and intensity inhomogeneity in brain MR images. To further improve the accuracy for brain MR image segmentation, this paper presents a Robust Generative Asymmetric GMM (RGAGMM) for simultaneous brain MR image segmentation and intensity inhomogeneity correction.

METHOD

First, we develop an asymmetric distribution to fit the data shapes, and thus construct a spatial constrained asymmetric model. Then, we incorporate two pseudo-likelihood quantities and bias field estimation into the model's log-likelihood, aiming to exploit the neighboring priors of within-cluster and between-cluster and to alleviate the impact of intensity inhomogeneity, respectively. Finally, an expectation maximization algorithm is derived to iteratively maximize the approximation of the data log-likelihood function to overcome the intensity inhomogeneity in the image and segment the brain MR images simultaneously.

RESULTS

To demonstrate the performances of the proposed algorithm, we first applied the proposed algorithm to a synthetic brain MR image to show the intermediate illustrations and the estimated distribution of the proposed algorithm. The next group of experiments is carried out in clinical 3T-weighted brain MR images which contain quite serious intensity inhomogeneity and noise. Then we quantitatively compare our algorithm to state-of-the-art segmentation approaches by using Dice coefficient (DC) on benchmark images obtained from IBSR and BrainWeb with different level of noise and intensity inhomogeneity. The comparison results on various brain MR images demonstrate the superior performances of the proposed algorithm in dealing with the noise and intensity inhomogeneity.

CONCLUSION

In this paper, the RGAGMM algorithm is proposed which can simply and efficiently incorporate spatial constraints into an EM framework to simultaneously segment brain MR images and estimate the intensity inhomogeneity. The proposed algorithm is flexible to fit the data shapes, and can simultaneously overcome the influence of noise and intensity inhomogeneity, and hence is capable of improving over 5% segmentation accuracy comparing with several state-of-the-art algorithms.

A segmentation of brain MRI images utilizing intensity and contextual information by Markov random field

BACKGROUND AND OBJECTIVE

Image segmentation is a preliminary and fundamental step in computer aided magnetic resonance imaging (MRI) images analysis. But the performance of most current image segmentation methods is easily depreciated by noise in MRI images. A precise and anti-noise segmentation of MRI images is desired in modern medical image diagnosis.

METHODS

This paper presents a segmentation of MRI images which combines fuzzy clustering and Markov random field (MRF). In order to utilize gray level information sufficiently and alleviate noise disturbance, fuzzy clustering is carried out on the original image and the coarse scale image of multi-scale decomposition. The spatial constraints between neighboring pixels are modeled by a defined potential function in the MRF to reduce the effect of noise and increase the integrity of segmented regions. Spatial constraints and the gray level information refined by Fuzzy C-Means (FCM) algorithm are integrated by maximum a posteriori Markov random field (MAP-MRF). In the proposed method, the fuzzy clustering membership obtained from the original image and the coarse scale image is integrated into the single-site clique potential functions by MAP-MRF. The defined potential functions and the distance weight are introduced to model the neighborhood constraint with MRF.

RESULTS

The experiments are carried out on noised synthetic images, simulated brain MR images and real MR images. The experimental results show that the proposed method has strong robustness and satisfying performance. Meanwhile the method is compared with FCM, FGFCM and FLICM algorithms visually and statistically in the experiments. In the comparison, the proposed method has achieved the best results. In the statistical comparison, the proposed method has an average similarity index of 36.8%, 33.7%, 2.75% increase against FCM, FGFCM and FLICM.

CONCLUSIONS

This paper proposes a MRI segmentation method combining fuzzy clustering and Markov random field. The method is tested in the noised image databases and comparison experiments, which shows that it is a precise and robust MRI segmentation method.