Brain Image Segmentation Research Articles

Multi-Atlas and Multi-Modal Hippocampus Segmentation for Infant MR Brain Images by Propagating Anatomical Labels on Hypergraph.

Accurate segmentation of hippocampus from infant magnetic resonance (MR) images is very important in the study of early brain development and neurological disorder. Recently, multi-atlas patch-based label fusion methods have shown a great success in segmenting anatomical structures from medical images. However, the dramatic appearance change from birth to 1-year-old and the poor image contrast make the existing label fusion methods less competitive to handle infant brain images. To alleviate these difficulties, we propose a novel multi-atlas and multi-modal label fusion method, which can unanimously label for all voxels by propagating the anatomical labels on a hypergraph. Specifically, we consider not only all voxels within the target image but also voxels across the atlas images as the vertexes in the hypergraph. Each hyperedge encodes a high-order correlation, among a set of vertexes, in different perspectives which incorporate 1) feature affinity within the multi-modal feature space, 2) spatial coherence within target image, and 3) population heuristics from multiple atlases. In addition, our label fusion method further allows those reliable voxels to supervise the label estimation on other difficult-to-label voxels, based on the established hyperedges, until all the target image voxels reach the unanimous labeling result. We evaluate our proposed label fusion method in segmenting hippocampus from T1 and T2 weighted MR images acquired from at 2-week-old, 3-month-old, 6-month-old, 9-month-old, and 12-month-old. Our segmentation results achieves improvement of labeling accuracy over the conventional state-of-the-art label fusion methods, which shows a great potential to facilitate the early infant brain studies.

Isointense Infant Brain Segmentation by Stacked Kernel Canonical Correlation Analysis.

Segmentation of isointense infant brain (at ~6-months-old) MR images is challenging due to the ongoing maturation and myelination process in the first year of life. In particular, signal contrast between white and gray matters inverses around 6 months of age, where brain tissues appear isointense and hence exhibit extremely low tissue contrast, thus posing significant challenges for automated segmentation. In this paper, we propose a novel segmentation method to address the above-mentioned challenges based on stacked kernel canonical correlation analysis (KCCA). Our main idea is to utilize the 12-month-old brain image with high tissue contrast to guide the segmentation of 6-month-old brain images with extremely low contrast. Specifically, we use KCCA to learn the common feature representations for both 6-month-old and the subsequent 12-month-old brain images of same subjects to make their features comparable in the common space. Note that the longitudinal 12-month-old brain images are not required in the testing stage, and they are required only in the KCCA based training stage to provide a set of longitudinal 6- and 12-month-old image pairs for training. Moreover, for optimizing the common feature representations, we propose a stacked KCCA mapping, instead of using only the conventional one-step of KCCA mapping. In this way, we can better use the 12-month-old brain images as multiple atlases to guide the segmentation of isointense brain images. Specifically, sparse patch-based multi-atlas labeling is used to propagate tissue labels in the (12-month-old) atlases and segment isointense brain images by measuring patch similarity between testing and atlas images with their learned common features. The proposed method was evaluated on 20 isointense brain images via leave-one-out cross-validation, showing much better performance than the state-of-the-art methods.

Volume Estimation of Various Brain Components Using MR Images - A Technical Report

In this paper, we present and discuss issues related to volume estimation of various brain components using MR brain images. We discuss pre-processing techniques to remove elements such as skull, blood vessels and fats from the MR images since these are non-essential to the volume calculations. The volume estimation is based on image segmentation. A challenge in MR brain image segmentation is to distinguish central gray matter from surrounding matters. This paper provides a frame work from pre-processing stage through volume estimation for white matter, gray matter and cerebrospinal fluid (CSF). The main contribution of this paper contains two parts. First, it provides a software-based method for interactive image pre-processing; second, it introduces a software-based, supervised interactive image segmentation method to deal with the segmentation of different matters, especially the central gray matter. Experiments with real data demonstrate the efficiency of our frame work.

MRI BRAIN IMAGE SEGMENTATION BASED ON CASCADED FRACTIONAL-ORDER DARWINIAN PARTICLE SWARM OPTIMIZATION AND MEAN SHIFT

Image segmentation is an initiative with massive interest in many imaging applications, suchas medical images and computer vision. It is considered as a challenging problem, so we need todevelop an efficient, fast technique for medical image segmentation. In this paper, the proposedframework is based on two segmentation methods: Fractional-order Darwinian Particle SwarmOptimization (FODPSO) and Mean Shift segmentation (MS). FODPSO is a favorable method forspecifying a predefined number of clusters and it can find the optimal set of thresholds with a higherbetween-class variance in less computational time. In the pre-processing phase,the MRI image isfiltered and the skull is removed. In the segmentation phase, the result of FODPSO is used as the inputto MS. Finally, we make a validation to thesegmented image. We compared our proposed system withsome state of the art segmentation techniques using brain benchmark data set. The experimental resultsshow that the proposed system enhances the accuracy of the MRI brain image segmentation.

Deep convolutional neural networks for multi-modality isointense infant brain image segmentation

Deep convolutional neural networks for multi-modality isointense infant brain image segmentation

Robust whole-brain segmentation: application to traumatic brain injury.

We propose a framework for the robust and fully-automatic segmentation of magnetic resonance (MR) brain images called "Multi-Atlas Label Propagation with Expectation-Maximisation based refinement" (MALP-EM). The presented approach is based on a robust registration approach (MAPER), highly performant label fusion (joint label fusion) and intensity-based label refinement using EM. We further adapt this framework to be applicable for the segmentation of brain images with gross changes in anatomy. We propose to account for consistent registration errors by relaxing anatomical priors obtained by multi-atlas propagation and a weighting scheme to locally combine anatomical atlas priors and intensity-refined posterior probabilities. The method is evaluated on a benchmark dataset used in a recent MICCAI segmentation challenge. In this context we show that MALP-EM is competitive for the segmentation of MR brain scans of healthy adults when compared to state-of-the-art automatic labelling techniques. To demonstrate the versatility of the proposed approach, we employed MALP-EM to segment 125 MR brain images into 134 regions from subjects who had sustained traumatic brain injury (TBI). We employ a protocol to assess segmentation quality if no manual reference labels are available. Based on this protocol, three independent, blinded raters confirmed on 13 MR brain scans with pathology that MALP-EM is superior to established label fusion techniques. We visually confirm the robustness of our segmentation approach on the full cohort and investigate the potential of derived symmetry-based imaging biomarkers that correlate with and predict clinically relevant variables in TBI such as the Marshall Classification (MC) or Glasgow Outcome Score (GOS). Specifically, we show that we are able to stratify TBI patients with favourable outcomes from non-favourable outcomes with 64.7% accuracy using acute-phase MR images and 66.8% accuracy using follow-up MR images. Furthermore, we are able to differentiate subjects with the presence of a mass lesion or midline shift from those with diffuse brain injury with 76.0% accuracy. The thalamus, putamen, pallidum and hippocampus are particularly affected. Their involvement predicts TBI disease progression.

LINKS: Learning-based multi-source IntegratioN frameworK for Segmentation of infant brain images

Segmentation of infant brain MR images is challenging due to insufficient image quality, severe partial volume effect, and ongoing maturation and myelination processes. In the first year of life, the image contrast between white and gray matters of the infant brain undergoes dramatic changes. In particular, the image contrast is inverted around 6–8months of age, and the white and gray matter tissues are isointense in both T1- and T2-weighted MR images and thus exhibit the extremely low tissue contrast, which poses significant challenges for automated segmentation. Most previous studies used multi-atlas label fusion strategy, which has the limitation of equally treating the different available image modalities and is often computationally expensive. To cope with these limitations, in this paper, we propose a novel learning-based multi-source integration framework for segmentation of infant brain images. Specifically, we employ the random forest technique to effectively integrate features from multi-source images together for tissue segmentation. Here, the multi-source images include initially only the multi-modality (T1, T2 and FA) images and later also the iteratively estimated and refined tissue probability maps of gray matter, white matter, and cerebrospinal fluid. Experimental results on 119 infants show that the proposed method achieves better performance than other state-of-the-art automated segmentation methods. Further validation was performed on the MICCAI grand challenge and the proposed method was ranked top among all competing methods. Moreover, to alleviate the possible anatomical errors, our method can also be combined with an anatomically-constrained multi-atlas labeling approach for further improving the segmentation accuracy.

Seeded Region Growing Based on Extension for Multispectral MR Images Classification

In the past, doctors judged images based on their own medical knowledge. Nowadays, the digital image processing technology can alleviate the burden of judging a large amount of multispectral information and lead to more effective diagnosis of the pathological tissues. In this paper, we propose a new approach of seeded region growing based on extension (SRGBE) to classify tissues from brain MRI. Based on extension, we tried to strengthen the regional definition. First, we use seeded region growing (SRG) to segment brain images. Second, the SRGBE result is further classified by K-means. Finally, we compare the images of gray matter, white matter and cerebral spinal fluid produced by both approaches to demonstrate the performance of SRGBE.

Improved fuzzy entropy clustering algorithm for MRI brain image segmentation

ABSTRACTMagnetic resonance imaging (MRI) brain image segmentation is essential at preliminary stage in the neuroscience research and computer‐aided diagnosis. However, presence of noise and intensity inhomogeneity in MRI brain images leads to improper segmentation. The fuzzy entropy clustering (FEC) is often used to deal with noisy data. One major disadvantage of the FEC algorithm is that it does not consider the local spatial information. In this article, we have proposed an improved fuzzy entropy clustering (IFEC) algorithm by introducing a new fuzzy factor, which incorporates both local spatial and gray‐level information. The IFEC algorithm is insensitive to noise, preserves the image detail during clustering, and is free of parameter selection. The efficacy of IFEC algorithm is demonstrated by comparing it quantitatively with the state‐of‐the‐art segmentation approaches in terms of similarity index on publically available real and simulated MRI brain images.

An Brain Image Segmentation Method based on Non-local MRF

Brain image segmentation is one of the most important parts of clinical diagnostic tools. However, accurate segmentation of brain images is a very difficult task due to the noise, inhomogeneity and sometimes deviation in brain images. Wells model incorporates the brain image segmentation and inhomogeneity correction into one framework to overcome influences from the intensity inhomogeneity and obtain good segmentation performance. However, the classical Wells model did not take spatial information into account, so its segmentation results are sensitive to the noise. In order to overcome this limitation, the MRF theory and the nonlocal information are used to construct a nonlocal Markov Random Field. With this nonlocal MRF, the improved Wells method can obtain much better segmentation results. The experimental results show that our method is robust to the noise and is able to simultaneously keep the image edge and slender topological structure very well.

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.

Pathological Brain Image Segmentation and Classification: A Survey

Pathological Brain Image Segmentation and Classification: A Survey

A HYBRID FIREFLY ALGORITHM WITH FUZZY-C MEAN ALGORITHM FOR MRI BRAIN SEGMENTATION

Image processing is one of the essential tasks to e xtract suspicious region and robust features from t he Magnetic Resonance Imaging (MRI). A numbers of the segmentation algorithms were developed in order to satisfy and increasing the accuracy of bra in tumor detection. In the medical image processing brain image segmentation is considered as a complex and challenging part. Fuzzy c-means is unsupervised method that has been implemented for clustering of the MRI and different purposes such as recognition of the pattern of interest and image segmentation. However; fuzzy c-means algorithm still suffers many drawbacks, such as low convergen ce rate, getting stuck in the local minima and vulnerable to initialization sensitivity. Firefly a lgorithm is a new population-based optimization method that has been used successfully for solving many complex problems. This paper proposed a new dynamic and intelligent clustering method for b rain tumor segmentation using the hybridization of Firefly Algorithm (FA) with Fuzzy C-Means algorithm (FCM). In order to automatically segment MRI brain images and improve the capability of the FCM to automatically elicit the proper number and location of cluster centres and the number of pixel s in each cluster in the abnormal (multiple scleros is lesions) MRI images. The experimental results prove d the effectiveness of the proposed FAFCM in enhancing the performance of the traditional FCM cl ustering. Moreover; the superiority of the FAFCM with other state-of-the-art segmentation methods is shown qualitatively and quantitatively. Conclusion : A novel efficient and reliable clustering algorithm presented in this work, which is called FAFCM based on the hybridization of the firefly algorithm with fuzzy c-mean clustering algorithm. Automatically; the hybridized algorithm has the cap ability to cluster and segment MRI brain images.

An Improved Gaussian Mixture Model based on NonLocal Information for Brain MR Images Segmentation

Brain image segmentation is an important part of medical image analysis. Due to the effect of imaging mechanism, MR images usually intensity in homogeneity, which is also named as bias field. Traditional Gaussian Mixed Model (GMM) method is hard to obtain satisfied segmentation results with the effect of noise and bias field. We propose a novel model based on GMM and nonlocal information. The improved method coupled segmentation and bias field correction that can manage the bias field while segmenting the image. In order to obtain a smooth bias field, we employed the Legendre Polynomials to fit it and merged it to the EM framework. We also use the non local information to deal with the noise and preserve geometrical edges information. The results show that our method can obtain more accurate results and bias field.

A study on fuzzy clustering for magnetic resonance brain image segmentation using soft computing approaches

This paper presents a novel idea of intracranial segmentation of magnetic resonance (MR) brain image using pixel intensity values by optimum boundary point detection (OBPD) method. The newly proposed (OBPD) method consists of three steps. Firstly, the brain only portion is extracted from the whole MR brain image. The brain only portion mainly contains three regions–gray matter (GM), white matter (WM) and cerebrospinal fluid (CSF). We need two boundary points to divide the brain pixels into three regions on the basis of their intensity. Secondly, the optimum boundary points are obtained using the newly proposed hybrid GA–BFO algorithm to compute final cluster centres of FCM method. For a comparison, other soft computing techniques GA, PSO and BFO are also used. Finally, FCM algorithm is executed only once to obtain the membership matrix. The brain image is then segmented using this final membership matrix. The key to our success is that we have proposed a technique where the final cluster centres for FCM are obtained using OBPD method. In addition, reformulated objective function for optimization is used. Initial values of boundary points are constrained to be in a range determined from the brain dataset. The boundary points violating imposed constraints are repaired. This method is validated by using simulated T1-weighted MR brain images from IBSR database with manual segmentation results. Further, we have used MR brain images from the Brainweb database with additional noise levels to validate the robustness of our proposed method. It is observed that our proposed method significantly improves segmentation results as compared to other methods.

A Comparison between ROI Extract and FCM Algorithms in Segmentation of Tumor Region from MR Brain Images

Segmentation of brain images based on Magnetic Resonance (MR) is used to segment and analyse the slices of brain to indicate the presence of tumour. Region of interest (ROI) extract is used to select the exact spot in an image where the tumour is present. Primary area of interest (tumour region) is chosen by the algorithm. The chosen region is completely extracted out from the input image.Fuzzy c-means algorithm is a soft clustering method in which the data elements (pixel values) are divided and clustered into two or more clusters. FCM is carried out by clustering the similar pixel values into one cluster and dissimilar pixel values into another cluster. A comparison is carried out between these two algorithms with the help of parameters such as Mean Square Error (MSE), Peak Signal to Noise Ratio (PSNR), segmentation accuracy, Jaccard index, computational time and memory requirement for processing the algorithm. The efficiency of either algorithm is proved using the comparison parameters

Automatic CT Brain Image Segmentation Using Two Level Multiresolution Mixture Model of EM

Tissue classification in computed tomography (CT) brain images is an important issue in the analysis of several brain dementias. A combination of different approaches for the segmentation of brain images is presented in this paper. A multi resolution algorithm is proposed along with scaled versions using Gaussian filter and wavelet analysis that extends expectation maximization (EM) algorithm. It is found that it is less sensitive to noise and got more accurate image segmentation than traditional EM. Moreover the algorithm has been applied on 20 sets of CT of the human brain and compared with other works. The segmentation results show the advantages of the proposed work have achieved more promising results and the results have been tested with Doctors.

Review of Segmentation Methods for Brain Tissue with Magnetic Resonance Images

Medical Magnetic Resonance Images (MRI) is characterized by a composition of small differences in signal intensities between different tissues types. Thus ambiguities and uncertainties are introduced in image formation. In this paper, review of the current approaches in the tissue segmentation of MR Brain Images has been presented. The segmentation algorithms has been divided into four categories which is able to deal with different intensity non-uniformity as adaptive spatial Fuzzy C - means, Markov Random Field, Fuzzy connectedness method and atlas based re- fuzzy connectedness. The performance of these segmentation methods have been compared in terms of validation metric as dice similarity coefficient, overlap ratio and Jaccard coefficient. The comparison of all validation metric at different levels of intensity non- uniformity shows that adaptive Fuzzy C - means clustering segmentation method give better result in segmentation of brain tissue.

Brain MR image segmentation based on local Gaussian mixture model and nonlocal spatial regularization

Brain Magnetic Resonance (MR) images often suffer from the inhomogeneous intensities caused by the bias field and heavy noise. The most widely used image segmentation algorithms, which typically rely on the homogeneity of image intensities in different regions, often fail to provide accurate segmentation results due to the existence of bias field and heavy noise. This paper proposes a novel variational approach for brain image segmentation with simultaneous bias correction. We define an energy functional with a local data fitting term and a nonlocal spatial regularization term. The local data fitting term is based on the idea of local Gaussian mixture model (LGMM), which locally models the distribution of each tissue by a linear combination of Gaussian function. By the LGMM, the bias field function in an additive form is embedded to the energy functional, which is helpful for eliminating the influence of the intensity inhomogeneity. For reducing the influence of noise and getting a smooth segmentation, the nonlocal spatial regularization is drawn upon, which is good at preserving fine structures in brain images. Experiments performed on simulated as well as real MR brain data and comparisons with other related methods are given to demonstrate the effectiveness of the proposed method.

Bayesian estimation of probabilistic atlas for tissue segmentation

Automatic anatomical brain image segmentation is still a challenge. In particular, algorithms have to address the partial volume effect (PVE) as well as the variability of the gray level of internal brain structures which may appear closer to gray matter (GM) than white matter (WM). Atlas based segmentation is one solution as it brings prior information. For such tasks, probabilistic atlases are very useful as they take account of the PVE information. In this paper, we provide a detailed analysis of a generative statistical model based on dense deformable templates that represents several tissue types observed in medical images. The inputs are gray level data whereas our atlas is composed of both an estimation of the deformation metric and probability maps of each tissue (called class). This atlas is used to guide the tissue segmentation of new images. Experiments are shown on brain T1 MRI datasets. This method only requires approximate pre-registration, as the latter is done jointly with the segmentation. Note however that an approximate registration is a reasonable pre-requisite given the application.