Real Magnetic Resonance Imaging Research Articles

Efficient Projection onto the $\ell_{\infty,1}$ Mixed-Norm Ball Using a Newton Root Search Method

Mixed norms that promote structured sparsity have numerous applications in signal processing and machine learning problems. In this work, we present a new algorithm, based on a Newton root search technique, for computing the projection onto the $\ell_{\infty,1}$ ball, which has found application in cognitive neuroscience and classification tasks. Numerical simulations show that our proposed method is between 8 and 10 times faster on average, and up to 20 times faster for very sparse solutions, than the previous state of the art. Tests on real functional magnetic resonance image data show that, for some data distributions, our algorithm can obtain speed improvements by a factor of between 10 and 100, depending on the implementation.

Real Time Dynamic Magnetic Resonance Imaging Via Dictionary Learning and Combined Fourier Transform

Real time dynamic magnetic resonance imaging (dMRI) requires that the image acquisition and reconstruction are carried out simultaneously and the reconstruction speed catches up with imaging speed. In this paper, a novel compressed sensing (CS) reconstruction algorithm for real time dynamic MRI is proposed. The first frame with more k-space measurements is reconstructed precisely as the reference image. Different from previous methods who start their reconstructions from zero-filled k-space measurements, a Combined Fourier Transform (CFT) algorithm is implemented in our method, which can dynamically aggregate the k-space measurements from previous sampled frames to create a highly accurate predictive image for the current frame. We then combine the CFT algorithm with a 3D path-based dictionary leaning algorithm, which is named as DLCFT in our work for fast real time dMRI reconstruction. The proposed algorithm is compared with four state-of-the-art online and offline methods on two real and complex perfusion MR sequences and a real functional brain MR sequence. Experimental results show that the proposed algorithm outperforms these methods with faster convergence and higher reconstruction accuracy.

Gadolinium-enhanced MRI reveals dynamic development of endolymphatic hydrops in Ménière's disease

IntroductionMeniere's disease is associated with impaired hearing, tinnitus, vertigo, and aural fullness. Many anatomical studies have suggested idiopathic endolymphatic hydrops as the pathological basis of Meniere's disease, which now can be visualized by using gadolinium -enhanced magnetic resonance imaging of the inner ear. ObjectiveTo investigate the development of endolymphatic hydrops in Meniere's disease by monitoring the vestibules and cochleae of affected patients. MethodsInner ears of 178 patients with definite unilateral Meniere's disease diagnosis were visualized by 3-dimensional fluid-attenuated inversion recovery and three-dimensional real inversion recovery magnetic resonance imaging following bilateral gadolinium intratympanic injection. The scans were used to evaluate the presence and degree of endolymphatic hydrops in the vestibules and cochlear structures, including the cochlear apical turn, the cochlear middle turn, and the cochlear basal turn. The correlation of endolymphatic hydrops occurrence between the various parts of the inner ear was determined. ResultsSymptomatic endolymphatic hydrops was detected on the affected side in all patients, whereas asymptomatic endolymphatic hydrops was detected on the unaffected contra-lateral side in 32 patients (18.0%). On the affected side, the cochlear apical turn and the cochlear middle turn demonstrated significantly higher rates of endolymphatic hydrops than the cochlear basal turn and the vestibule. The severity of endolymphatic hydrops gradually decreased from the cochlear apical turn to the cochlear basal turn. On the contra lateral side, the incidence and degree of the detected asymptomatic endolymphatic hydrops were significantly greater in the cochleae than in the vestibules (p<0.05), with no significant difference detected between the cochlear turns. ConclusionProgression of endolymphatic hydrops appears to be directional, initiated in the cochlea. The order of endolymphatic hydrops severity gradually decreases from the cochlear apical turn to the cochlear basal turn and then to the vestibule. Endolymphatic hydrops in the vestibule is associated with symptomatic Meniere's disease.

Classification of MRI Brain Images Using DNA Genetic Algorithms Optimized Tsallis Entropy and Support Vector Machine

As a non-invasive diagnostic tool, Magnetic Resonance Imaging (MRI) has been widely used in the field of brain imaging. The classification of MRI brain image conditions poses challenges both technically and clinically, as MRI is primarily used for soft tissue anatomy and can generate large amounts of detailed information about the brain conditions of a subject. To classify benign and malignant MRI brain images, we propose a new method. Discrete wavelet transform (DWT) is used to extract wavelet coefficients from MRI images. Then, Tsallis entropy with DNA genetic algorithm (DNA-GA) optimization parameters (called DNAGA-TE) was used to obtain entropy characteristics from DWT coefficients. At last, DNA-GA optimized support vector machine (called DNAGA-KSVM) with radial basis function (RBF) kernel, is applied as a classifier. In our experimental procedure, we use two kinds of images to validate the availability and effectiveness of the algorithm. One kind of data is the Simulated Brain Database and another kind of image is real MRI images which downloaded from Harvard Medical School website. Experimental results demonstrate that our method (DNAGA-TE+KSVM) obtained better classification accuracy.

Brain Image Segmentation Based on Dual-Tree Complex Wavelet Transform and Fuzzy C-Means Clustering Algorithm

Image segmentation is an important step in most medical image analysis tasks. An effective image segmentation method helps clinicians and patients in image-guided surgery, radiotherapy, early disease detection, volumetric measurement, and three-dimensional visualization. The fuzzy c-means (FCM) clustering algorithm is one of the most popular methods used for medical image segmentation. However, it does not produce satisfactory results for images with noise and intensity inhomogeneities. Hence, a wavelet-based FCM clustering algorithm is proposed in this work. An advanced wavelet transform, such as the dual-tree complex wavelet transform (DT-CWT), is proposed to sharpen the edges and to avoid segmentation error caused by noise. An appropriate level of decomposition is selected on the basis of the images. The FCM clustering technique is applied on the wavelet transformed image by selecting an optimal number of clusters. The combination of DT-CWT and FCM clustering technique produces an effective segmentation result. The conventional discrete wavelet transform (DWT) was also tested, but it was unable to give an efficient segmentation result when combined with FCM. Experiments were conducted on real T1-weighted magnetic resonance (MR) images to validate the proposed algorithm. Moreover, a comparison was performed with different state-of-the-art algorithms to show the superiority of our proposed method.

Multilevel magnetic resonance imaging compression using compressive sensing

In this study, a multilevel compressive sensing (CS) compression for magnetic resonance imaging (MRI) images is presented. The proposed algorithm divides the image into frames of equal size, transforms the pixels inside the frame into the sparse domain, and then applies the CS compression to each frame with different level of compression. Four levels of compression are suggested, based on how sparse is the information inside the frame. The proposed algorithm is evaluated using six real MRI images showing different parts of the human body. The experimental results show a significant improvement of 7.03 dB in peak-signal-to-noise ratio and 23.76% in compression level (CL) when compared with a uniform CL algorithm.

Generative Adversarial Networks for the Creation of Realistic Artificial Brain Magnetic Resonance Images.

Even as medical data sets become more publicly accessible, most are restricted to specific medical conditions. Thus, data collection for machine learning approaches remains challenging, and synthetic data augmentation, such as generative adversarial networks (GAN), may overcome this hurdle. In the present quality control study, deep convolutional GAN (DCGAN)–based human brain magnetic resonance (MR) images were validated by blinded radiologists. In total, 96 T1-weighted brain images from 30 healthy individuals and 33 patients with cerebrovascular accident were included. A training data set was generated from the T1-weighted images and DCGAN was applied to generate additional artificial brain images. The likelihood that images were DCGAN-created versus acquired was evaluated by 5 radiologists (2 neuroradiologists [NRs], vs 3 non-neuroradiologists [NNRs]) in a binary fashion to identify real vs created images. Images were selected randomly from the data set (variation of created images, 40%–60%). None of the investigated images was rated as unknown. Of the created images, the NRs rated 45% and 71% as real magnetic resonance imaging images (NNRs, 24%, 40%, and 44%). In contradistinction, 44% and 70% of the real images were rated as generated images by NRs (NNRs, 10%, 17%, and 27%). The accuracy for the NRs was 0.55 and 0.30 (NNRs, 0.83, 0.72, and 0.64). DCGAN-created brain MR images are similar enough to acquired MR images so as to be indistinguishable in some cases. Such an artificial intelligence algorithm may contribute to synthetic data augmentation for “data-hungry” technologies, such as supervised machine learning approaches, in various clinical applications.

An Efficient Automatic Midsagittal Plane Extraction in Brain MRI

In this paper, a fully automatic and computationally efficient midsagittal plane (MSP) extraction technique in brain magnetic resonance images (MRIs) has been proposed. Automatic detection of MSP in neuroimages can significantly aid in registration of medical images, asymmetric analysis, and alignment or tilt correction (recenter and reorientation) in brain MRIs. The parameters of MSP are estimated in two steps. In the first step, symmetric features and principal component analysis (PCA)-based technique is used to vertically align the bilateral symmetric axis of the brain. In the second step, PCA is used to achieve a set of parallel lines (principal axes) from the selected two-dimensional (2-D) elliptical slices of brain MRIs, followed by a plane fitting using orthogonal regression. The developed algorithm has been tested on 157 real T1-weighted brain MRI datasets including 14 cases from the patients with brain tumors. The presented algorithm is compared with a state-of-the-art approach based on bilateral symmetry maximization. Experimental results revealed that the proposed algorithm is fast (&lt;1.04 s per MRI volume) and exhibits superior performance in terms of accuracy and precision (a mean z-distance of 0.336 voxels and a mean angle difference of 0.06).

Decentralized temporal independent component analysis: Leveraging fMRI data in collaborative settings

The field of neuroimaging has recently witnessed a strong shift towards data sharing; however, current collaborative research projects may be unable to leverage institutional architectures that collect and store data in local, centralized data centers. Additionally, though research groups are willing to grant access for collaborations, they often wish to maintain control of their data locally. These concerns may stem from research culture as well as privacy and accountability concerns. In order to leverage the potential of these aggregated larger data sets, we require tools that perform joint analyses without transmitting the data. Ideally, these tools would have similar performance and ease of use as their current centralized counterparts. In this paper, we propose and evaluate a new Algorithm, decentralized joint independent component analysis (djICA), which meets these technical requirements. djICA shares only intermediate statistics about the data, plausibly retaining privacy of the raw information to local sites, thus making it amenable to further privacy protections, for example via differential privacy. We validate our method on real functional magnetic resonance imaging (fMRI) data and show that it enables collaborative large-scale temporal ICA of fMRI, a rich vein of analysis as of yet largely unexplored, and which can benefit from the larger-N studies enabled by a decentralized approach. We show that djICA is robust to different distributions of data over sites, and that the temporal components estimated with djICA show activations similar to the temporal functional modes analyzed in previous work, thus solidifying djICA as a new, decentralized method oriented toward the frontiers of temporal independent component analysis.

Split Bregman method based level set formulations for segmentation and correction with application to MR images and color images

At present, magnetic resonance (MR) images have gradually become a major aid for clinical medicine, which has greatly improved the doctor's diagnosis rate. Accurate and fast segmentation of MR images plays an extremely important role in medical research. However, due to the influence of external factors and the defects of imaging devices, the MR images have severe intensity inhomogeneity, which poses a great challenge to accurately segment MR images. To deal with this problem, this paper presents an improved active contour model by combining the level set evolution model (LSE) and the split Bregman method, and gives the two-phase, the multi-phase and the vector-valued formulations of our model, respectively. The use of the split Bregman method accelerates the minimization process of our model by reducing the computation time and iterative times. A slowly varying bias field is added into the energy functional, which is the key to correct inhomogeneous images. By estimating the bias fields, not only can we get accurate image segmentation results, but also a homogeneous image after correction is provided. Then we apply our model to segment a large amount of synthetic and real MR images, including gray and color images. Experimental results show that our model can provide satisfactory segmentation and correction results for both gray and color images. Besides, compared with the LSE model, our model has higher accuracy and is superior to the LSE model. In addition, experimental results also demonstrate that our model has the advantages of being insensitive to initial contours and robust to noises.

An Optimized Cascaded Stochastic Resonance for the Enhancement of Brain MRI

BackgroundThe contrast enhancement of Magnetic Resonance Imaging (MRI) data is quite challenging as the noise present in this data also get amplified in this process. Dynamic Stochastic Resonance (DSR) is the technique that utilizes the noise to enhance the contrast of MRI data. MethodThe present study proposes the cascaded stochastic resonance, which exploits the properties of modified potential neuron model and quartic bistable model of DSR. The Multi-objective Particle Swarm Optimization (MOPSO) tunes the DSR parameters associated with the cascading of both the models. The MOPSO produces a set of the solution called Pareto front for the maximization of two image quality measures, i.e., contrast enhancement factor and universal image quality index. Further, the maximization of another image quality measure, i.e., anisotropy helps to obtain the optimum enhanced image from the Pareto fronts solution. ResultsThe present study included the simulated and real MRI data. The results show that the proposed method obtained mean contrast enhancement factor, universal image quality index and anisotropy equal to 1.79, 0.78 and 0.021 respectively. These values are better than those obtained for classical bistable DSR and other conventional contrast enhancement techniques. The proposed algorithm has been tested on real MRI data as well and found valuable in the diagnosis of lacunar infarct and mesial temporal sclerosis. ConclusionThe cascaded DSR based on MOPSO has shown promising results and may be highly beneficial to the diagnosis of different brain pathology.

Sparse Ordinal Logistic Regression and Its Application to Brain Decoding.

Brain decoding with multivariate classification and regression has provided a powerful framework for characterizing information encoded in population neural activity. Classification and regression models are respectively used to predict discrete and continuous variables of interest. However, cognitive and behavioral parameters that we wish to decode are often ordinal variables whose values are discrete but ordered, such as subjective ratings. To date, there is no established method of predicting ordinal variables in brain decoding. In this study, we present a new algorithm, sparse ordinal logistic regression (SOLR), that combines ordinal logistic regression with Bayesian sparse weight estimation. We found that, in both simulation and analyses using real functional magnetic resonance imaging (fMRI) data, SOLR outperformed ordinal logistic regression with non-sparse regularization, indicating that sparseness leads to better decoding performance. SOLR also outperformed classification and linear regression models with the same type of sparseness, indicating the advantage of the modeling tailored to ordinal outputs. Our results suggest that SOLR provides a principled and effective method of decoding ordinal variables.

Combining Non-negative Matrix Factorization and Sparse Coding for Functional Brain Overlapping Community Detection

The functional system of the human brain can be viewed as a complex network. Among various features of the brain functional network, community structure has raised significant interest in recent years. Increasing evidence has revealed that most realistic complex networks have an overlapping community structure. However, the overlapping community structure of the brain functional network has not been adequately studied. In this paper, we propose a novel method called sparse symmetric non-negative matrix factorization (ssNMF) to detect the overlapping community structure of the brain functional network. Specifically, it is formulated by combining the effective techniques of non-negative matrix factorization and sparse coding. Besides, the non-negative adaptive sparse representation is applied to construct the whole-brain functional network, based on which ssNMF is performed to detect the community structure. Both simulated and real functional magnetic resonance imaging data are used to evaluate ssNMF. The experimental results demonstrate that the proposed ssNMF method is capable of accurately and stably detecting the underlying overlapping community structure. Moreover, the physiological interpretation of the overlapping community structure detected by ssNMF is straightforward. This novel framework, we think, provides an effective tool to study overlapping community structure and facilitates the understanding of the network organization of the functional human brain.

Consistent adaptive sequential dictionary learning

Abstract Algorithms for learning overcomplete dictionaries for sparse signal representation are mostly iterative minimization methods that alternate between a sparse coding stage and a dictionary update stage. For most however, the notion of consistency of the learned quantities has not been addressed. Based on the observation that the observed signals can be approximated as a sum of rank one matrices, a new adaptive dictionary learning algorithm is proposed in this paper. It is derived via sequential adaptive penalized rank one matrix approximation where the l1-norm is introduced as a penalty promoting sparsity. The proposed algorithm uses a block coordinate descent approach to consistently estimate the unknowns and has the advantage of having simple closed form solutions for both the sparse coding and dictionary update stages. The consistency properties of both the estimated sparse code and dictionary atom are provided. The performance of the proposed algorithm compared to some state of the art algorithms is illustrated on both simulated data and a real functional magnetic resonance imaging (fMRI) data set from a finger tapping experiment.

Automatic MR image segmentation using maximization of mutual information

Magnetic resonance (MR) brain image segmentation is an important task for the early detection of any deformation followed by the quantitative analysis for the prediction and stage defection of brain diseases. But segmentation of the MR brain image suffers from limited accuracy as captured images have non-uniform homogeneity over an organ, presence of noise, uneven and broken boundary etc. Due to the complex structure of the brain and varieties of the captured MR images, only a single feature based MR image segmentation cannot give sufficient accurate result. In the proposed method thresholds for segmenting the MR image are computed by maximizing the mutual information for the two features, compactness and homogeneity. The proposed algorithm is tested against the real T1 MR image to asses the accuracy. Further the output is validated and compared with the ground truth and other recently reported works.

Simultaneous single- and multi-contrast super-resolution for brain MRI images based on a convolutional neural network

In magnetic resonance imaging (MRI), the acquired images are usually not of high enough resolution due to constraints such as long sampling times and patient comfort. High-resolution MRI images can be obtained by super-resolution techniques, which can be grouped into two categories: single-contrast super-resolution and multi-contrast super-resolution, where the former has no reference information, and the latter applies a high-resolution image of another modality as a reference. In this paper, we propose a deep convolutional neural network model, which performs single- and multi-contrast super-resolution reconstructions simultaneously. Experimental results on synthetic and real brain MRI images show that our convolutional neural network model outperforms state-of-the-art MRI super-resolution methods in terms of visual quality and objective quality criteria such as peak signal-to-noise ratio and structural similarity.

Multiple Matrix Gaussian Graphs Estimation.

Matrix-valued data, where the sampling unit is a matrix consisting of rows and columns of measurements, are emerging in numerous scientific and business applications. Matrix Gaussian graphical model is a useful tool to characterize the conditional dependence structure of rows and columns. In this article, we employ nonconvex penalization to tackle the estimation of multiple graphs from matrix-valued data under a matrix normal distribution. We propose a highly efficient nonconvex optimization algorithm that can scale up for graphs with hundreds of nodes. We establish the asymptotic properties of the estimator, which requires less stringent conditions and has a sharper probability error bound than existing results. We demonstrate the efficacy of our proposed method through both simulations and real functional magnetic resonance imaging analyses.

Pre-operative MRI staging of endometrial cancer in a multicentre cancer network: can we match single centre study results?

To evaluate the staging accuracy of magnetic resonance imaging (MRI) for endometrial cancer in daily practice over a 3-year period at a tertiary referral centre receiving scans from a large number of hospitals with varying protocols. To compare these daily practice results to published data from single-centre studies. After ethical approval, MRI staging records for 270 studies from nine network and three centre hospitals were retrospectively collected and compared with final operative histopathology. The International Federation of Gynaecology and Obstetrics (FIGO) stage, depth of invasion assessment and cervical stromal invasion were analysed and reasons for discrepancies reviewed. MRI-based complete FIGO stage was fully concordant with histopathology in 65.6%. MRI accuracy for depth of myometrial invasion and cervical stromal invasion was 73.3% and 89.3% respectively. Our results did not match the high accuracy previously reported in studies based on single centres. Published MRI staging accuracy from small single-centre studies were not replicated in a tertiary referral centre receiving scans with heterogeneous protocols over a 3-year period. These results highlight the challenges faced in daily practice and may reflect achievable and realistic MRI staging accuracies in large rapid throughput referral networks. Adherence to standardised high-quality protocols may help to improve future results. • Three-year MRI-staging accuracy for endometrial cancer in a multicentre cancer network • Daily practice MRI-staging accuracy did not meet results of single-centre studies • Large scale cancer network MRI-staging accuracies should be further evaluated • Treatment recommendations should be based on achievable MRI-staging accuracies.

Robust fuzzy c‐means clustering algorithm using non‐parametric Bayesian estimation in wavelet transform domain for noisy MR brain image segmentation

The major drawback of the fuzzy c-means (FCM) algorithm is its sensitivity to noise. The authors propose a new extended FCM algorithm based a non-parametric Bayesian estimation in the wavelet transform domain for segmenting noisy MR brain images. They use the Bayesian estimator to process the noisy wavelet coefficients. Before segmentation based on FCM algorithm, they use an a priori statistical model adapted to the modelisation of the wavelet coefficients of a noisy image. The main objective of this wavelet-based Bayesian statistical estimation is to recover a good quality image, from a noisy image of poor quality. Experimental results on simulated and real magnetic resonance imaging brain images show that their proposed method solves the problem of sensitivity to noise and offers a very good performance that outperforms some FCM-based algorithms.

Robustness of Representative Signals Relative to Data Loss Using Atlas-Based Parcellations.

Parcellation-based approaches are an important part of functional magnetic resonance imaging data analysis. They are a necessary processing step for sorting data in structurally or functionally homogenous regions. Real functional magnetic resonance imaging datasets usually do not cover the atlas template completely; they are often spatially constrained due to the physical limitations of MR sequence settings, the inter-individual variability in brain shape, etc. When using a parcellation template, many regions are not completely covered by actual data. This paper addresses the issue of the area coverage required in real data in order to reliably estimate the representative signal and the influence of this kind of data loss on network analysis metrics. We demonstrate this issue on four datasets using four different widely used parcellation templates. We used two erosion approaches to simulate data loss on the whole-brain level and the ROI-specific level. Our results show that changes in ROI coverage have a systematic influence on network measures. Based on the results of our analysis, we recommend controlling the ROI coverage and retaining at least 60% of the area in order to ensure at least 80% of explained variance of the original signal.