BrainWeb Dataset Research Articles

Rubik-Net: Learning Spatial Information via Rotation-Driven Convolutions for Brain Segmentation.

The accurate segmentation of brain tissue in Magnetic Resonance Image (MRI) slices is essential for assessing neurological conditions and brain diseases. However, it is challenging to segment MRI slices because of the low contrast between different brain tissues and the partial volume effect. 2-Dimensional (2-D) convolutional networks cannot handle such volumetric image data well because they overlook spatial information between MRI slices. Although 3-Dimensional (3-D) convolutions capture volumetric spatial information, they have not been fully exploited to enhance representative ability of deep networks; moreover, they may lead to overfitting with insufficient training data. In this paper, we propose a novel convolutional mechanism, termed Rubik convolution, to capture multi-dimensional information between MRI slices. Rubik convolution rotates the axis of a set of consecutive slices, enabling 2-D convolution kernels to extract features of each axial plane simultaneously. Next, feature maps are rotated back to fuse multidimensional information by the Max-View-Maps. Furthermore, we propose an efficient 2-D convolutional network, namely Rubik-Net, where the residual connections and the bottleneck structure are used to enhance information transmission and reduce the number of network parameters. The Rubik-Net shows promising results on iSeg2017, iSeg2019, IBSR and BrainWeb datasets in terms of segmentation accuracy. In particular, we achieved the best results in 95th percentile Hausdorff distance and average surface distance in cerebrospinal fluid segmentation on the most challenging iSeg2019 dataset. The experiments indicate that Rubik-Net improves the accuracy and efficiency of medical image segmentation. Moreover, Rubik convolution can be easily embedded into existing 2-D convolutional networks.

Segmentation of brain MRI using an altruistic Harris Hawks’ Optimization algorithm

Segmentation is an essential requirement in medicine when digital images are used in illness diagnosis, especially, in posterior tasks as analysis and disease identification. An efficient segmentation of brain Magnetic Resonance Images (MRIs) is of prime concern to radiologists due to their poor illumination and other conditions related to de acquisition of the images. Thresholding is a popular method for segmentation that uses the histogram of an image to label different homogeneous groups of pixels into different classes. However, the computational cost increases exponentially according to the number of thresholds. In this paper, we perform the multi-level thresholding using an evolutionary metaheuristic. It is an improved version of the Harris Hawks Optimization (HHO) algorithm that combines the chaotic initialization and the concept of altruism. Further, for fitness assignment, we use a hybrid objective function where along with the cross-entropy minimization, we apply a new entropy function, and leverage weights to the two objective functions to form a new hybrid approach. The HHO was originally designed to solve numerical optimization problems. Earlier, the statistical results and comparisons have demonstrated that the HHO provides very promising results compared with well-established metaheuristic techniques. In this article, altruism has been incorporated into the HHO algorithm to enhance its exploitation capabilities. We evaluate the proposed method over 10 benchmark images from the WBA database of the Harvard Medical School and 8 benchmark images from the Brainweb dataset using some standard evaluation metrics. On the Harvard WBA dataset, a Peak Signal to Noise Ratio (PSNR) of 26.61 and a Structural Similarity Index (SSIM) of 0.92 are achieved using 5 thresholds. For the same scenario, using the Brainweb dataset, a PSNR of 24.77 and SSIM of 0.86 are obtained. The obtained results justify the superiority of the proposed approach compared to existing state-of-the-art methods and baseline methods. The relevant codes for the proposed approach are available at: https://github.com/Rohit-Kundu/Segmentation-HHO_Altruism.

Magnetic resonance images denoising using a wavelet solution to laplace equation associated with a new variational model

In this paper by exploiting the theory of wavelet transform, a solution of Laplace equation, after changing certain initial conditions in terms of wavelet transformation is obtained. We have further applied this solution of Laplace equation to denoise magnetic resonance (MR) images from brain web dataset at different noise levels. The denoised MR images are again denoised with the help of a new proposed variational model with certain wavelet. We compared our results with the reported results of Yadav et al along with some other recently reported results. It was found that the proposed method not only outperforms the reported methods of Yadav et al but some other recently reported results also.

On Two Algorithms for Multi-Modality Image Registration Based on Gaussian Curvature and Application to Medical Images

Registration of multi-modal images is one of the challenging problems in image processing nowadays. In this paper, two novel non-rigid registration models are proposed for multi-modality images. In model 1, mutual information of the template and reference images is used as data fitting term with Gaussian curvature regularization. This approach may not give satisfactory results in noisy images or images having bias field. To overcome this drawback, model 2 is proposed which is based on normalized gradient of both template and reference images as a data fitting term instead of mutual information. To get best transformations, both the models are minimized by using Augmented Lagrangian Method. The proposed models can register multi-modality images without effecting edges and other important fine details and are also tested on various medical images like (T1-T2 MRI, PD weighted-T2 MRI) noisy and synthetic images. The proposed models are also tested on a well known free available Brainweb dataset, where they produced satisfactory results. From experimental results, it can be observed that normalized gradient field based model gives better results than mutual information based model. Comparison is done qualitatively and quantitatively through Jaccard Similarity Coefficient.

Partially Supervised Kernel Induced Rough Fuzzy Clustering for Brain Tissue Segmentation

In modern imaging diagnosis Magnetic Resonance Imaging (MRI) is possibly one of the widely used effective techniques particularly for brain tissue segmentation. Clustering techniques may not be perfect always. Clustering can be significantly improved by supervising partially. A novel partially supervised kernel induced rough fuzzy clustering is proposed for brain tissue segmentation by employing a small quantity of labeled pixels with constraint seeded policy. Labeled pixels act as the constraints which are utilized to initialize the clustering process and guide the method towards a more accurate partitioning. Kernel trick used here enhances the possibility of linear partition of different complex segments of brain which cannot separate linearly in its original feature space. Whereas, the rough and fuzzy set handles the overlappingness, vagueness and indiscernibility of different tissue regions. A variety of benchmark brain MRI datasets are used for the experiments. The ability of the method is compared with state-of-the-art clustering segmentation techniques and evaluated using different validity indices. Experimental results confirm that the technique considerably enhances the segmentation accuracy with a little quantity of supervision. Enhancement in accuracy gained by the method compared to the other techniques are 0.3, 0.37, 1.15, and 1.03% for IBSR datasets 144, 150, 155, and 167, respectively, and 1.02% for the BrainWeb dataset 85. Statistical impact of the method is confirmed from the paired t-test results.

Fuzzy inference based contextual dissimilarity histogram equalization algorithm for image enhancement

AbstractIn order to overcome the drawback of the existing image enhancement technologies and further consider the pixel intensity expression error caused by imaging, a novel fuzzy inference‐based contextual dissimilarity histogram equalization (FICDHE) algorithm is proposed. The proposed algorithm is composed of three modules. In the first module, according to the calculated probable intensity intervals, the membership functions of intensity are generated. In the second one, fuzzy inference systems are established and the contextual dissimilarity of each pixel is calculated. In the third module, the contextual dissimilarity histograms are clipped and equalized. The fuzzy system established in this paper not only fully considers the uncertainty source of pixel gray level expression, but also has adaptability. The parameter selection of fuzzy inference system membership function in this algorithm does not need human intervention, but is automatically obtained based on the statistical information of image pixel gray. Its adaptability makes the algorithm more widely used and convenient. Experiments are conducted using four typical medical images and 800 images from the KNIX dataset and BrainWeb dataset. The performance of the proposed method was compared to a series of enhancement algorithms based on both subjective judgment and image quality measurement indexes. Experimental results demonstrate that the proposed algorithm has a better contrast enhancement ability and yields better performance.

Brain tumor segmentation of multi-modality MR images via triple intersecting U-Nets

In this paper, we propose a triple intersecting U-Nets (TIU-Nets) for brain glioma segmentation. First, the proposed TIU-Nets is composed of binary-class segmentation U-Net (BU-Net) and multi-class segmentation U-Net (MU-Net), in which MU-Net reuses multi-resolution features from BU-Net. Second, we introduce a segmentation soft-mask predicted by BU-Net, that is, candidate glioma region is generated by removing most of non-glioma backgrounds, which guides multi-category segmentation of MU-Net in a weighted manner. Third, an edge branch in MU-Net is leveraged to enhance boundary information of glioma substructure, which facilitates to locate glioma true boundaries and improve segmentation accuracy. Finally, we propose a sigmoid-evolution based polarized cross-entropy loss (S-CE) to resolve class unbalance problem, and apply S-CE loss to soft-mask prediction loss in BU-Net, multi-class segmentation loss in MU-Net and edge prediction loss in edge branch. Experimental results have demonstrated that the proposed 2D/3D TIU-Nets achieves a higher segmentation accuracy than corresponding 2D/3D state-of-the-art segmentation methods including FCN, U-Net, SegNet, CRDN, IVD-Net, FCDenseNet, DeepMedic, DMFNet, etc, evaluating on publicly available brain tumor segmentation challenge 2015 (BRATS2015) datasets. To show the universality of the proposed method, we also give a comparison of segmentation performance on BrainWeb dataset.

Semantic segmentation of cerebrospinal fluid and brain volume with a convolutional neural network in pediatric hydrocephalus\u2014transfer learning from existing algorithms

BackgroundFor the segmentation of medical imaging data, a multitude of precise but very specific algorithms exist. In previous studies, we investigated the possibility of segmenting MRI data to determine cerebrospinal fluid and brain volume using a classical machine learning algorithm. It demonstrated good clinical usability and a very accurate correlation of the volumes to the single area determination in a reproducible axial layer. This study aims to investigate whether these established segmentation algorithms can be transferred to new, more generalizable deep learning algorithms employing an extended transfer learning procedure and whether medically meaningful segmentation is possible.MethodsNinety-five routinely performed true FISP MRI sequences were retrospectively analyzed in 43 patients with pediatric hydrocephalus. Using a freely available and clinically established segmentation algorithm based on a hidden Markov random field model, four classes of segmentation (brain, cerebrospinal fluid (CSF), background, and tissue) were generated. Fifty-nine randomly selected data sets (10,432 slices) were used as a training data set. Images were augmented for contrast, brightness, and random left/right and X/Y translation. A convolutional neural network (CNN) for semantic image segmentation composed of an encoder and corresponding decoder subnetwork was set up. The network was pre-initialized with layers and weights from a pre-trained VGG 16 model. Following the network was trained with the labeled image data set. A validation data set of 18 scans (3289 slices) was used to monitor the performance as the deep CNN trained. The classification results were tested on 18 randomly allocated labeled data sets (3319 slices) and on a T2-weighted BrainWeb data set with known ground truth.ResultsThe segmentation of clinical test data provided reliable results (global accuracy 0.90, Dice coefficient 0.86), while the CNN segmentation of data from the BrainWeb data set showed comparable results (global accuracy 0.89, Dice coefficient 0.84). The segmentation of the BrainWeb data set with the classical FAST algorithm produced consistent findings (global accuracy 0.90, Dice coefficient 0.87). Likewise, the area development of brain and CSF in the long-term clinical course of three patients was presented.ConclusionUsing the presented methods, we showed that conventional segmentation algorithms can be transferred to new advances in deep learning with comparable accuracy, generating a large number of training data sets with relatively little effort. A clinically meaningful segmentation possibility was demonstrated.

RETRACTED ARTICLE: Rough fuzzy region based bounded support fuzzy C-means clustering for brain MR image segmentation

Precise brain tissue segmentation and analysis in the presence of intensity non-uniformity (INU) and noise is the challenging task due to intensity overlaps between data pixels within the image. The clustering method is commonly used to variety of applications for grouping similar data items. Specifically, the fuzzy C-means (FCM) clustering method is extensively used in many real world and research applications. In this paper, the rough fuzzy region based bounded support fuzzy C-means (RFRBSFCM) clustering method is proposed for brain MR image INU estimation and correction, and segmentation. The rough fuzzy regions are estimated based on similarity distance vector and it is determined from both local and global spatial information. In addition, the proposed algorithm incorporates bounded support vector for estimating weighted image. The objective function of proposed algorithm is minimized for segmenting different tissues in brain MR image. The RFRBSFCM algorithm is tested with recent FCM clustering techniques using simulated T1 and T2-weighted brain MR images from public BrainWeb dataset. The quantitative results confirm that the proposed algorithm achieves superior performance than other recent state-of-the-art methods.

Optimized Multistable Stochastic Resonance for the Enhancement of Pituitary Microadenoma in MRI.

Magnetic resonance imaging (MRI) is the modality of choice as far as imaging diagnosis of pathologies in the pituitary gland is concerned. Furthermore, the advent of dynamic contrast enhanced (DCE) has enhanced the capability of this modality in detecting minute benign but endocrinologically significant tumors called microadenoma. These lesions are visible with difficulty and a low confidence level in routine MRI sequences, even after administration of intravenous gadolinium. Techniques to enhance the visualization of such foci would be an asset in improving the overall accuracy of DCE-MRI for detection of pituitary microadenomas. The present study proposes an algorithm for postprocessing DCE-MRI data using multistable stochastic resonance (MSSR) technique. Multiobjective ant lion optimization optimizes the contrast enhancement factor (CEF) and anisotropy of an image by varying the parameters associated with the dynamics of MSSR. The marked regions of interest (ROIs) are labeled as normal and microadenoma of pituitary obtained with increased level of accuracy and confidence using proposed algorithm. The increased difference between the mean intensity curves obtained using these ROIs validated the obtained subjective results. Furthermore, the proposed MSSR-based algorithm has been evaluated on standard T1 and T2 weighted BrainWeb dataset images and quantified in terms of CEF, peak signal to noise ratio (PSNR), structure similarity index measure (SSIM), and universal quality index (UQI). The obtained mean values of CEF 1.22, PSNR 27.68, SSIM 0.75, UQI 0.83 for twenty dataset images were highest among considered contrast enhancement algorithms for the comparison.

Modified complex diffusion based nonlinear filter for restoration and enhancement of magnetic resonance images

In this paper, a new method is proposed for removal of Rician noise from MRI. Method is casted into a variational framework, consists of two terms wherein the first term is a data likelihood term and the second term is a prior function. The first term is obtained by minimising the negative log likelihood of Rician pdf. Owing to ill-posedness of the likelihood term, a prior function is introduced which is a nonlinear complex diffusion based filter. A regularisation parameter is used to balance the trade-off between data fidelity term and prior. The performance analysis of the proposed method with other standard methods is presented for Brain Web dataset. The values of performance measures such as PSNR, RMSE, SSIM, and CP are presented for various noise levels. From the simulation results, it is observed that the proposed method is performing better in comparison to other methods in consideration.

Identification and removal of different categories of noises from magnetic resonance image using hybrid partial differential equation-based filter

This paper identifies various types of noise present into the MRI and filters them by choosing an appropriate filter. The different categories of noises during the acquisition of MR image may be generally corrupted due to external or internal causes. The external causes lead to an additive noise pattern which follows a Gaussian distribution (pdf). Causes of internal noise in MR image are basically the intrinsic noise that is generated during the acquisition process. Normally intrinsic noise in MR image follows the Rician distribution (pdf). The proposed filter gets adapted for the removal of specific types of noise based on SNR values of image data. The performance analysis and comparative study of the proposed method with other standard methods is presented for brain web dataset at varying noise levels in terms of PSNR, MSE, SSIM and CP. From the simulation results, it is observed that the proposed framework with CD-based prior is performing better in comparison to other priors.

Unsupervised tissue classification of brain MR images for voxel-based morphometry analysis

In this article, a fully unsupervised method for brain tissue segmentation of T1-weighted MRI 3D volumes is proposed. The method uses the Fuzzy C-Means (FCM) clustering algorithm and a Fully Connected Cascade Neural Network (FCCNN) classifier. Traditional manual segmentation methods require neuro-radiological expertise and significant time while semiautomatic methods depend on parameter's setup and trial-and-error methodologies that may lead to high intraoperator/interoperator variability. The proposed method selects the most useful MRI data according to FCM fuzziness values and trains the FCCNN to learn to classify brain’ tissues into White Matter, Gray Matter, and Cerebro-Spinal Fluid in an unsupervised way. The method has been tested on the IBSR dataset, on the BrainWeb Phantom, on the BrainWeb SBD dataset, and on the real dataset “University of Palermo Policlinico Hospital” (UPPH), Italy. Sensitivity, Specificity, Dice and F-Factor scores have been calculated on the IBSR and BrainWeb datasets segmented using the proposed method, the FCM algorithm, and two state-of-the-art brain segmentation software packages (FSL and SPM) to prove the effectiveness of the proposed approach. A qualitative evaluation involving a group of five expert radiologists has been performed segmenting the real dataset using the proposed approach and the comparison algorithms. Finally, a usability analysis on the proposed method and reference methods has been carried out from the same group of expert radiologists. The achieved results show that the segmentations of the proposed method are comparable or better than the reference methods with a better usability and degree of acceptance.

Segmentation of longitudinal brain MR images using bias correction embedded fuzzy c-means with non-locally spatio-temporal regularization

We propose an automated method for segmentation of brain tissues in longitudinal MR images. In the proposed method, images acquired at each time point are first separately segmented into white matter, gray matter, and cerebrospinal fluid by bias correction embedded fuzzy c-means. Intensities differences are then defined as similarities of each voxel to the cluster centroids. After being normalized in inter-class, the similarities are incorporated into a non-local means de-noising formula to regularize the segmentation in both spatial and temporal dimensions. Non-locally regularization results are used to compute final membership functions for the segmentation. To improve time performance, we accelerate the modified de-noising algorithm using CUDA and obtain a 200× performance improvement. Quantitative comparison with the state-of-the-art methods on BrainWeb dataset demonstrate advantages of the proposed method in terms of segmentation accuracy and the ability to consistently segment brain tissues in an arbitrary number of longitudinal brain MR image series.

A partial differential equation-based general framework adapted to Rayleigh's, Rician's and Gaussian's distributed noise for restoration and enhancement of magnetic resonance image.

The proposed framework is obtained by casting the noise removal problem into a variational framework. This framework automatically identifies the various types of noise present in the magnetic resonance image and filters them by choosing an appropriate filter. This filter includes two terms: the first term is a data likelihood term and the second term is a prior function. The first term is obtained by minimizing the negative log likelihood of the corresponding probability density functions: Gaussian or Rayleigh or Rician. Further, due to the ill-posedness of the likelihood term, a prior function is needed. This paper examines three partial differential equation based priors which include total variation based prior, anisotropic diffusion based prior, and a complex diffusion (CD) based prior. A regularization parameter is used to balance the trade-off between data fidelity term and prior. The finite difference scheme is used for discretization of the proposed method. The performance analysis and comparative study of the proposed method with other standard methods is presented for brain web dataset at varying noise levels in terms of peak signal-to-noise ratio, mean square error, structure similarity index map, and correlation parameter. From the simulation results, it is observed that the proposed framework with CD based prior is performing better in comparison to other priors in consideration.

Deformable image registration based on elastodynamics

This paper presents a novel method for 2D inter-subject non-rigid image registration. It is motivated by the ideas derived from elastodynamics which is the subclass of linear elastic theory. We propose to model the non-rigid deformations as elastic waves which are characterized by elastodynamics wave equation. Our registration method is formulated in a multi-resolution manner such that it is able to recover significant deformations. The proposed scheme recovers local (non-linear) as well as global deformations. The results of image registration by the proposed technique were validated via similarity measures for real MR brain images, BrainWeb dataset and MR prostate images. We have also used overlap measures (Jaccard distance, Dice coefficient and Overlap coefficient) over the segmented deep-brain structures including: lateral and third ventricles, putamen and caudate nucleus for real image dataset to further validate our proposed registration scheme. The proposed scheme was also compared with demons, Level set and free-form deformations (cubic B-spline) algorithms implemented in ITK toolkit and symmetric normalization method available in ANTs software package. Our proposed algorithm shows marked improvement of the inter-subject non-rigid registration, both qualitatively and quantitatively, over the four above-mentioned methods. This comparison was performed on all the three types of datasets mentioned above. 2D synthetic experiments comprising of Patch-`C', Square---Rectangle and Circle-`C' were also performed with our algorithm and compared with other four methods. Both qualitative and quantitative results show that our method performs better than the other four methods. Results suggest that our proposed algorithm has improved accuracy over lemons, level set and free-form deformations (ITK implemented) and symmetric normalization method (ANTs based), tested over two types of brain and one prostate MRI. The commonly used synthetic experiments also suggest the superiority of our proposed scheme.

Brain Tissue Classification in Magnetic Resonance Images

Automatic segmentation of brain images is a challenging problem due to the complex structure of brain images, as well as to the absence of anatomy models. Brain segmentation into white matter, gray matter, and cerebral spinal fluid, is an important stage for many problems, including the studies in 3-D visualizations for disease detection and surgical planning. In this paper we present a novel fully automated framework for tissue classification of brain in MR Images that is a combination of two techniques: GLCM and SVM, each of which has been customized for the problem of brain tissue segmentation such that the results are more robust than its individual components that is demonstrated through experiments. The proposed framework has been validated on brainweb dataset of different modalities, with desirable performance in the presence of noise and bias field. To evaluate the performance of the proposed method the Kappa similarity index is computed. Our method achieves higher kappa index (91.5) compared with other methods currently in use. As an application, our method has been used for segmentation of MR images with promising results.

A NOVEL METHOD OF BRAIN TUMOR SEGMENTATION USING EXTENSIVE FEATURE SET

Brain tissue Segmentation from the MRI images is having significance in the medical research field. The accurate Segmentation of the normal as well as the abnormal tissues is the complex assignment in this process. Because of the inconsistency and difficulty of abnormal tissues, MRI Brain Image Segmentation turned into more hard procedure. In this paper, a technique is proposed for segmenting the abnormalities such as Tumor and Atrophy in the MRI Brain images. (1) Feature extraction (2) Classification (3) Segmentation are the three stages offered in this work. At first, the features such as energy, entropy, homogeneity, contrast and correlation from MRI Brain Images are extracted. Next, by utilizing Neuro-Fuzzy classifier, the Classification process is carried out and for this process, the feature set is specified as the input. From the outcome of Classification, the images are categorized into normal as well as abnormal. The further procedure Segmentation is performed according to this outcome only. The abnormal MRI images are segmented into abnormal tissues like Tumor and Atrophy using Region Growing method. Utilizing MATLAB platform the implementation of the proposed technique is made. The experimentation is carried out on the MRI Brain Images by BrainWeb data sets. The performance of our proposed technique is assessed with the help of the metrics namely FPR, FNR, Specificity, Sensitivity and Accuracy. Therefore, using our proposed technique with enhanced classification, the abnormal tissues of MRI Brain images are segmented accurately.

An efficient neuro-fuzzy based segmentation of normal tissues in brain MRI (BMRI) using extensive feature set

Brain tissue Segmentation from the MRI images is having significance in the medical research field. The accurate Segmentation of the normal as well as the abnormal tissues is the complex assignment in this process. In this paper, a technique named Neuro-Fuzzy Based Segmentation (NFBS) is proposed for segmenting the normal features such as White Matter (WM), Gray Matter (GM) and Cerebro-Spinal Fluid (CSF) in the MRI Brain images. (1) Feature extraction (2) Classification (3) Segmentation are the three stages offered in this work. At first, the features such as energy, entropy, homogeneity, contrast and correlation from MRI Brain Images are extracted. Next, by utilizing Neuro-Fuzzy classifier, the Classification process is carried out and for this process, the feature set is specified as the input. From the outcome of Classification, the images are categorized into normal as well as abnormal. The further procedure Segmentation is performed according to this outcome only. The normal MRI images are segmented into normal tissues like White Matter (WM), Gray Matter (GM) and Cerebro-Spinal Fluid (CSF). All the tissues are individually segmented by special methods such as Gradient method, Orthogonal Polynomial Transform method. Utilizing MATLAB platform, the implementation of the proposed technique is made. The experimentation is carried out on the MRI Brain Images by BrainWeb data sets. The performance of the proposed technique is assessed with the help of the metrics namely FPR, FNR, Specificity, Sensitivity and Accuracy. Therefore, using our proposed techniques with enhanced classification, the normal tissues of MRI Brain images are segmented accurately.

REGION MERGING STRATEGY FOR BRAIN MRI SEGMENTATION USING DEMPSTER-SHAFER THEORY

Detection of brain tissues using magnetic resonance imaging (MRI) is an active and challenging research area in computational neuroscience. Brain MRI artifacts lead to an uncertainty in pixel values. Therefore, brain MRI segmentation is a complicated concern which is tackled by a novel data fusion approach. The proposed algorithm has two main steps. In the first step the brain MRI is divided to some main and ancillary cluster which is done using Fuzzy c-mean (FCM). In the second step, the considering ancillary clusters are merged with main clusters employing Dempster-Shafer Theory. The proposed method was validated on simulated brain images from the commonly used BrainWeb dataset. The results of the proposed method are evaluated by using Dice and Tanimoto coefficients which demonstrate well performance and robustness of this algorithm.