Hybrid genetic and variational expectation-maximization algorithm for gaussian-mixture-model-based brain MR image segmentation

Modified distance regularized level set evolution for brain ventricles segmentation

Neurodegenerative disorders are commonly characterized by atrophy of the brain which is caused by neuronal loss. Ventricles are one of the prominent structures in the brain; their shape changes, due to their content, the cerebrospinal fluid. Analyzing the morphological changes of ventricles, aids in the diagnosis of atrophy, for which the region of interest needs to be separated from the background. This study presents a modified distance regularized level set evolution segmentation method, incorporating regional intensity information. The proposed method is implemented for segmenting ventricles from brain images for normal and atrophy subjects of magnetic resonance imaging and computed tomography images. Results of the proposed method were compared with ground truth images and produced sensitivity in the range of 65%–90%, specificity in the range of 98%–99%, and accuracy in the range of 95%–98%. Peak signal to noise ratio and structural similarity index were also used as performance measures for determining segmentation accuracy: 95% and 0.95, respectively. The parameters of level set formulation vary for different datasets. An optimization procedure was followed to fine tune parameters. The proposed method was found to be efficient and robust against noisy images. The proposed method is adaptive and multimodal.

MUSIC-Expected maximization gaussian mixture methodology for clustering and detection of task-related neuronal firing rates

Researchers often rely on simple methods to identify involvement of neurons in a particular motor task. The historical approach has been to inspect large groups of neurons and subjectively separate neurons into groups based on the expertise of the investigator. In cases where neuron populations are small it is reasonable to inspect these neuronal recordings and their firing rates carefully to avoid data omissions. In this paper, a new methodology is presented for automatic objective classification of neurons recorded in association with behavioral tasks into groups. By identifying characteristics of neurons in a particular group, the investigator can then identify functional classes of neurons based on their relationship to the task. The methodology is based on integration of a multiple signal classification (MUSIC) algorithm to extract relevant features from the firing rate and an expectation-maximization Gaussian mixture algorithm (EM-GMM) to cluster the extracted features. The methodology is capable of identifying and clustering similar firing rate profiles automatically based on specific signal features. An empirical wavelet transform (EWT) was used to validate the features found in the MUSIC pseudospectrum and the resulting signal features captured by the methodology. Additionally, this methodology was used to inspect behavioral elements of neurons to physiologically validate the model. This methodology was tested using a set of data collected from awake behaving non-human primates.

Methodological challenges and analytic opportunities for modeling and interpreting Big Healthcare Data

Managing, processing and understanding big healthcare data is challenging, costly and demanding. Without a robust fundamental theory for representation, analysis and inference, a roadmap for uniform handling and analyzing of such complex data remains elusive. In this article, we outline various big data challenges, opportunities, modeling methods and software techniques for blending complex healthcare data, advanced analytic tools, and distributed scientific computing. Using imaging, genetic and healthcare data we provide examples of processing heterogeneous datasets using distributed cloud services, automated and semi-automated classification techniques, and open-science protocols. Despite substantial advances, new innovative technologies need to be developed that enhance, scale and optimize the management and processing of large, complex and heterogeneous data. Stakeholder investments in data acquisition, research and development, computational infrastructure and education will be critical to realize the huge potential of big data, to reap the expected information benefits and to build lasting knowledge assets. Multi-faceted proprietary, open-source, and community developments will be essential to enable broad, reliable, sustainable and efficient data-driven discovery and analytics. Big data will affect every sector of the economy and their hallmark will be 'team science'.

Cerebrospinal fluid image segmentation using spatial fuzzy clustering method with improved evolutionary Expectation Maximization

Visualization of cerebrospinal fluid (CSF), that flow in the brain and spinal cord, plays an important role to detect neurodegenerative diseases such as Alzheimer's disease. This is performed by measuring the substantial changes in the CSF flow dynamics, volume and/or pressure gradient. Magnetic resonance imaging (MRI) technique has become a prominent tool to quantitatively measure these changes and image segmentation method has been widely used to distinguish the CSF flows from the brain tissues. However, this is often hampered by the presence of partial volume effect in the images. In this paper, a new hybrid evolutionary spatial fuzzy clustering method is introduced to overcome the partial volume effect in the MRI images. The proposed method incorporates Expectation Maximization (EM) method, which is improved by the evolutionary operations of the Genetic Algorithm (GA) to differentiate the CSF from the brain tissues. The proposed improvement is incorporated into a spatial-based fuzzy clustering (SFCM) method to improve segmentation of the boundary curve of the CSF and the brain tissues. The proposed method was validated using MRI images of Alzheimer's disease patient. The results presented that the proposed method is capable to filter the CSF regions from the brain tissues more effectively compared to the standard EM, FCM, and SFCM methods.

Alpha shape theory for 3D visualization and volumetric measurement of brain tumor progression using magnetic resonance images

Resection of brain tumors is a tricky task in surgery due to its direct influence on the patients' survival rate. Determining the tumor resection extent for its complete information via-à-vis volume and dimensions in pre- and post-operative Magnetic Resonance Images (MRI) requires accurate estimation and comparison. The active contour segmentation technique is used to segment brain tumors on pre-operative MR images using self-developed software. Tumor volume is acquired from its contours via alpha shape theory. The graphical user interface is developed for rendering, visualizing and estimating the volume of a brain tumor. Internet Brain Segmentation Repository dataset (IBSR) is employed to analyze and determine the repeatability and reproducibility of tumor volume. Accuracy of the method is validated by comparing the estimated volume using the proposed method with that of gold-standard. Segmentation by active contour technique is found to be capable of detecting the brain tumor boundaries. Furthermore, the volume description and visualization enable an interactive examination of tumor tissue and its surrounding. Admirable features of our results demonstrate that alpha shape theory in comparison to other existing standard methods is superior for precise volumetric measurement of tumor.

Comparison of 10 brain tissue segmentation methods using revisited IBSR annotations

PURPOSE

Ground-truth annotations from the well-known Internet Brain Segmentation Repository (IBSR) datasets consider Sulcal cerebrospinal fluid (SCSF) voxels as gray matter. This can lead to bias when evaluating the performance of tissue segmentation methods. In this work we compare the accuracy of 10 brain tissue segmentation methods analyzing the effects of SCSF ground-truth voxels on accuracy estimations.

MATERIALS AND METHODS

The set of methods is composed by FAST, SPM5, SPM8, GAMIXTURE, ANN, FCM, KNN, SVPASEG, FANTASM, and PVC. Methods are evaluated using original IBSR ground-truth and ranked by means of their performance on pairwise comparisons using permutation tests. Afterward, the evaluation is repeated using IBSR ground-truth without considering SCSF.

RESULTS

The Dice coefficient of all methods is affected by changes in SCSF annotations, especially on SPM5, SPM8 and FAST. When not considering SCSF voxels, SVPASEG (0.90 ± 0.01) and SPM8 (0.91 ± 0.01) are the methods from our study that appear more suitable for gray matter tissue segmentation, while FAST (0.89 ± 0.02) is the best tool for segmenting white matter tissue.

CONCLUSION

The performance and the accuracy of methods on IBSR images vary notably when not considering SCSF voxels. The fact that three of the most common methods (FAST, SPM5, and SPM8) report an important change in their accuracy suggest to consider these differences in labeling for new comparative studies.

Automatic segmentation of corpus callosum using Gaussian mixture modeling and Fuzzy C means methods

This paper presents a comparative study of the success and performance of the Gaussian mixture modeling and Fuzzy C means methods to determine the volume and cross-sectionals areas of the corpus callosum (CC) using simulated and real MR brain images. The Gaussian mixture model (GMM) utilizes weighted sum of Gaussian distributions by applying statistical decision procedures to define image classes. In the Fuzzy C means (FCM), the image classes are represented by certain membership function according to fuzziness information expressing the distance from the cluster centers. In this study, automatic segmentation for midsagittal section of the CC was achieved from simulated and real brain images. The volume of CC was obtained using sagittal sections areas. To compare the success of the methods, segmentation accuracy, Jaccard similarity and time consuming for segmentation were calculated. The results show that the GMM method resulted by a small margin in more accurate segmentation (midsagittal section segmentation accuracy 98.3% and 97.01% for GMM and FCM); however the FCM method resulted in faster segmentation than GMM. With this study, an accurate and automatic segmentation system that allows opportunity for quantitative comparison to doctors in the planning of treatment and the diagnosis of diseases affecting the size of the CC was developed. This study can be adapted to perform segmentation on other regions of the brain, thus, it can be operated as practical use in the clinic.

A deformable cosegmentation algorithm for brain MR images

Cosegmentation aims to simultaneously segment the common parts in a pair of images, and has recently attracted increasing research attention in the field of computer vision. In this paper, we propose a novel deformable cosegmentation (D-C) algorithm to solve the brain MR image segmentation problem by cosegmenting the image and a co-registered atlas. In this manner, the prior heuristic information about brain anatomy that is embedded in the atlas can be transformed into the constraints that control the segmentation of brain MR images. Based on the multiphase Chan-Vese model, the proposed D-C algorithm is implemented using level set techniques. Then, it is compared to the protocol algorithm and the state-of-the-art GA-EM algorithm in T1-weighted brain MR images corrupted by different levels of Gaussian noise and intensity non-uniformity. Our results show that the proposed D-C algorithm can differentiate major brain structures more accuratly and produce more robust segmentation of brain MR images.

Joint probabilistic model of shape and intensity for multiple abdominal organ segmentation from volumetric CT images

We propose a novel joint probabilistic model that correlates a new probabilistic shape model with the corresponding global intensity distribution to segment multiple abdominal organs simultaneously. Our probabilistic shape model estimates the probability of an individual voxel belonging to the estimated shape of the object. The probability density of the estimated shape is derived from a combination of the shape variations of target class and the observed shape information. To better capture the shape variations, we used probabilistic principle component analysis optimized by expectation maximization to capture the shape variations and reduce computational complexity. The maximum a posteriori estimation was optimized by the iterated conditional mode-expectation maximization. We used 72 training datasets including low- and high-contrast CT images to construct the shape models for the liver, spleen and both kidneys. We evaluated our algorithm on 40 test datasets that were grouped into normal (34 normal cases) and pathologic (6 datasets) classes. The testing datasets were from different databases and manual segmentation was performed by different clinicians. We measured the volumetric overlap percentage error, relative volume difference, average square symmetric surface distance, false positive rate and false negative rate and our method achieved accurate and robust segmentation for multiple abdominal organs simultaneously.

Topology-based nonlocal fuzzy segmentation of brain MR image with inhomogeneous and partial volume intensity

PURPOSE

The aim was to automatically segment brain magnetic resonance (MR) image with inhomogeneous and partial volume (PV) intensity for brain and neurophysiology analysis.

METHODS

Rather than assuming the presence of a single bias field over the image data, we first apply a local model to MR image analysis. With the brain topology knowledge, several specific local regions are selected, and typical brain tissues are then extracted for the prior estimation of fuzzy clustering center and member function. A new nonlocal fuzzy labeling scheme is applied to global optimization segmentation based on the block comparison and distance weight, which is robust to noise and inhomogeneous intensity. The nonlocal labeling provides optimized fuzzy member value and local intensity estimation of brain tissues such as cerebrospinal fluid (CSF), white matter (WM), and gray matter (GM). In addition to inhomogeneous intensity, PV may lead to error segmentation. To correct error segmentation because of PV, this article also provides two correction schemes. The first one is to extract CSF in deep sulci, which captures more CSF candidate by intensity comparison and topology shape comparison. The local pure CSF, WM, and GM is then estimated to correct the interfaces of CSF/GM and WM/GM.

RESULTS

The segmentation experiments are performed on both brainweb-simulated images and Internet brain segmentation repository database (IBSR) real images. The experimental results demonstrate the robust and efficient performance of our approach.

CONCLUSIONS

Our approach can be applied to automatic segmentation of the brain MR image.

Diffusion-weighted imaging-based probabilistic segmentation of high- and low-proliferative areas in high-grade gliomas

The apparent diffusion coefficient (ADC) derived from diffusion-weighted imaging (DWI) correlates inversely with tumor proliferation rates. High-grade gliomas are typically heterogeneous and the delineation of areas of high and low proliferation is impeded by partial volume effects and blurred borders. Commonly used manual delineation is further impeded by potential overlap with cerebrospinal fluid and necrosis. Here we present an algorithm to reproducibly delineate and probabilistically quantify the ADC in areas of high and low proliferation in heterogeneous gliomas, resulting in a reproducible quantification in regions of tissue inhomogeneity. We used an expectation maximization (EM) clustering algorithm, applied on a Gaussian mixture model, consisting of pure superpositions of Gaussian distributions. Soundness and reproducibility of this approach were evaluated in 10 patients with glioma. High- and low-proliferating areas found using the clustering correspond well with conservative regions of interest drawn using all available imaging data. Systematic placement of model initialization seeds shows good reproducibility of the method. Moreover, we illustrate an automatic initialization approach that completely removes user-induced variability. In conclusion, we present a rapid, reproducible and automatic method to separate and quantify heterogeneous regions in gliomas.

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

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