Search

Background Mental fatigue is one of the most causes of road accidents. Identification of biological tools and methods such as electroencephalogram (EEG) are invaluable to detect them at early stage in hazard situations. Methods In this paper, an expert automatic method based on brain region connectivity for detecting fatigue is proposed. The recorded general data during driving in both fatigue (the last five minutes) and alert (at the beginning of driving) states are used in analyzing the method. In this process, the EEG data during continuous driving in one to two hours are noted. The new feature of Gaussian Copula Mutual Information (GCMI) based on wavelet coefficients is calculated to detect brain region connectivity. Classification for each subject is then done through selected optimal features using the support vector machine (SVM) with linear kernel. Results The designed technique can classify trials with 98.1% accuracy. The most significant contributions to the selected features are the wavelet coefficients details 1−2 (corresponding to the Beta and Gamma frequency bands) in the central and temporal regions. In this paper, a new algorithm for channel selection is introduced that has been able to achieve 97.2% efficiency by selecting eight channels from 30 recorded channels. Conclusion The obtained results from the classification are compared with other methods, and it is proved that the proposed method accuracy is higher from others at a significant level. The technique is completely automatic, while the calculation load could be reduced remarkably through selecting the optimal channels implementing in real-time systems.

Hyperspectral image (HSI) and multispectral image (MSI) fusion aiming to improve HSI spatial resolution has attracted increasing research interests in the recent decade. Fusing hyperspectral and mu...

Robust data hiding methods are basically used when rightful ownership is intended and/or transmitting channel is prone to noise. In this paper, a robust and secure data hiding method is proposed in Tchebichef domain. For feature extraction, remainders of Tchebichef coefficients are calculated according to Chinese remainder theorem (CRT). Distances of these remainders are utilized as the extracted features. These distances are then modified according to the hidden bits, to complete the embedding process. To reduce the chance of illegal access and increase the security of the proposed method, a closed-loop chaotic encryption system (CLCES) has been introduced. The proposed CLCES requires very little side information and produces extremely sensitive noise-like sequences with high security. Robustness and security of the proposed method have been analyzed. Comparison results with other related methods in the current literature confirm the superiority of the proposed method. Due to its simplicity, the proposed method is a good candidate for real-time data hiding applications.

Brain extraction is one of the most important preprocessing steps in cerebral magnetic resonance (MR) image analysis. Brain extraction from neonatal MR images is particularly challenging due to significant differences in head size and shape between neonates and rapid changes in neonatal brain structure in the weeks and months after birth. In this work, a multi-atlas-based neonatal brain extraction method using atlas library clustering and local label fusion (NOBELL) is presented. In NOBELL, an affinity propagation (AP) approach is first applied to cluster images of an atlas library into clusters represented by exemplars, which are used to select best matching clusters for target images. A local weighted voting strategy based on Jacobian determinant ranking is then employed to extract brain from target images using training images in best matching clusters. The performance of NOBELL was evaluated on T2- and T1-weighted scans of 40 neonates aged between 37 and 44 weeks. NOBELL outperformed two popular brain extraction tools, FSL’s Brain Extraction Tool (BET) and BrainSuite’s Brain Surface Extractor (BSE), and achieved higher accuracy with brain masks very close to manually extracted ones. NOBELL showed an average Jaccard coefficient of 0.974 (0.942) on T2 (T1)-weighted images in comparison with 0.908 (0.602) and 0.845 (0.762) achieved by BSE, and BET, respectively. NOBELL allows for accurate and efficient brain extraction, a crucial step in brain MRI applications such as accurate brain tissue segmentation and volume estimation as well as accurate cortical surface delineation in neonates.

Multilevel thresholding using Otsu or Kapur methods is widely used in the context of image segmentation. These methods select optimal thresholds in gray level images by maximizing between-class variance or entropy criterion. These methods become time consuming and less efficient with increasing number of thresholds. To increase the efficiency of the image segmentation using multilevel thresholding based on Kapur and Otsu methods, we developed a hybrid optimization algorithm named BMO-DE based on bird mating optimization (BMO) and differential evolutionary (DE) algorithms. The efficiency of the proposed method was evaluated on eight standard benchmark images. The proposed method achieved better segmentation results in term of solution quality and stability in comparison with other well-known techniques including bacterial foraging (BF), modified bacterial foraging (MBF), particle swarm optimization (PSO), genetic algorithm (GA) and hybrid algorithm named PSO-DE.

One of the challenges in improving the performance of brain computer interface systems is to overcome the large number of extracted features from EEG signals. Feature selection can reduce noisy data, overtraining effects, necessary storage, computational complexity, and can improve the performance of the classifier. Different feature selection methods have been used to achieve these goals. In this study, a new hybrid feature selection method is proposed. It employs a filter bank common spatial pattern for feature extraction and a grey wolf optimization algorithm to search and generate optimal feature subset with performance evaluated by support vector machine classifier. Also, In order to increase the search performance of the proposed feature selection algorithm, a new parallel combined grey wolf and differential evolution optimization algorithm is proposed. Experimental results show that the proposed methods improve the performance of motor imagery brain computer interface system in comparison to the state-of-the-art methods, even with small training data.

Automatic sleep stage scoring is of great importance for investigating sleep architecture during infancy. In this work, we introduce a novel multichannel approach based on deep learning networks and hidden Markov models (HMM) to improve the accuracy of sleep stage classification in term neonates. The classification performance was evaluated on quiet sleep (QS) and active sleep (AS) stages, each with two sub-states, using multichannel EEG data recorded from sixteen neonates with postmenstrual age of 38–40 weeks. A comprehensive set of linear and nonlinear features were extracted from thirty-second EEG segments. The feature space dimensionality was then reduced by using an evolutionary feature selection method called MGCACO (Modified Graph Clustering Ant Colony Optimization) based on the relevance and redundancy analysis. A bi-directional long-short time memory (BiLSTM) network was trained for sleep stage classification. The number of channels was optimized using the sequential forward selection method to reduce the spatial space. Finally, an HMM-based postprocessing stage was used to reduce false positives by incorporating the knowledge of transition probabilities between stages into the classification process. The method performance was evaluated using the K-fold (KFCV) and leave-one-out cross-validation (LOOCV) strategies. Using six-bipolar channels, our method achieved a mean kappa and an overall accuracy of 0.71-0.76 and 78.9%-82.4% using the KFCV and LOOCV strategies, respectively. The presented automatic sleep stage scoring method can be used to study the neurodevelopmental process and to diagnose brain abnormalities in term neonates.

Gene clustering is a prerequisite in the analysis of microarray data where sets of co-expressed genes are clustered. In this paper, a multi-objective clonal selection optimization algorithm (MCSOA) is developed based on the immune system behavior for gene clustering purposes in which the number of clusters can vary in a predefined range. To achieve a reliable clustering outcome on various gene expression (GE) datasets, the most effective clustering validity indexes are incorporated and represented in terms of two conflicting objective functions. For the sake of fast convergence to the optimal solutions, a new population updating mechanism is iteratively applied to select the less-dominate solutions of the previous iteration. The proposed clustering technique is implemented on various publicly available microarray datasets. Comparing the results with those of the widely used gene clustering techniques confirms the superiority and efficacy of the proposed technique.

Dimensionality reduction is an important preprocessing step to improve the performance of machine learning algorithms. Feature selection methods can efficiently speed up the learning process and improve the overall classification accuracy by reducing the computational complexity. Among the feature selection methods, multivariate methods are more effective in removing irrelevant and redundant features. An efficient multivariate feature selection method, optimization method, called ‘graph clustering based ant colony optimization (GCACO)’ has been recently introduced and shown to outperform other well-known feature selection methods. In the GCACO, features are divided into communities (clusters) in the entire feature space represented as a graph by an efficient community detection algorithm. An ACO-based search strategy is then used to select an optimal feature subset from the initial set of features. In this paper, a modified GCACO algorithm called MGCACO is presented to significantly improve the performance of the GCACO. Performance of the MGCACO algorithm was assessed by testing it on several standard benchmark datasets and sleep EEG data. The performance of the MGCACO was compared to those obtained using the original GCACO and other well-known filtering methods available in the literature. The MGCACO achieved superior performance over the GCACO and other univariate and multivariate algorithms with up to 10%. The MGCACO also exhibited higher efficiency in reducing the number of features all by keeping the classification accuracy maximum.

In this paper, a bottom-up salient object detection method is proposed by modeling image as a random graph. The proposed method starts with portioning input image into superpixels and extracting color and spatial features for each superpixel. Then, a complete graph is constructed by employing superpixels as nodes. A high edge weight is assigned into a pair of superpixels if they have high similarity. Next, a random walk prior on nodes is assumed to generate the probability distribution on edges. On the other hand, a complete directed graph is created that each edge weight represents the probability for transmitting random walker from current node to next node. By considering a threshold and eliminating edges with higher probability than the threshold, a random graph is created to model input image. The inbound degree vector of a random graph is computed to determine the most salient nodes (regions). Finally, a propagation technique is used to form saliency map. Experimental results on two challenging datasets: MSRA10K and SED2 demonstrate the efficiency of the proposed unsupervised RG method in comparison with the state-of-the-art unsupervised methods.

An efficient fuzzy interactive multi-objective optimization method is proposed to select the sub-optimal subset of genes from large-scale gene expression data. It is based on the binary particle swarm optimization (BPSO) algorithm tuned by a chaotic method. The proposed method is able to select the sub-optimal subset of genes with the least number of features that can accurately distinguish between the two classes, e.g. the normal and cancerous samples. The proposed method is evaluated on several publicly available microarray and RNA-sequencing gene expression datasets such as leukemia, colon cancer, central nervous system, lung cancer, ovarian cancer, prostate cancer and RNA-seq lung disease. The results indicate that the proposed method can identify the minimum number of genes to achieve the most accuracy, sensitivity and specificity in the classification process. Achieving 100% accuracy in six out of the seven datasets investigated in this study, demonstrates the high capacity of the proposed algorithm to find the sub-optimal subset of genes. This approach is useful in clinical applications to extract the most influential genes on a disease and to find the treatment procedure for the disease.

Background Brain tumor extraction from magnetic resonance (MR) images is challenging due to variations in the location, shape, size and intensity of tumors. Manual delineation of brain tumors from MR images is time-consuming and prone to human errors. Method In this paper, we present a method for automatic tumor extraction from multimodal MR images. Brain tumors are first detected using k-means clustering. A morphological region-based active contour model is then used for tumor extraction using an initial contour defined based on the boundary of the detected brain tumor regions. The contour evolution for tumor extraction was performed using successive application of morphological operators. In our model, a Gaussian distribution was used to model local image intensities. The spatial correlation between neighboring voxels was also modeled using Markov random field. Results The proposed method was evaluated on BraTS 2013 dataset including patients with high-grade and low-grade tumors. In comparison with other active contour based methods, the proposed method yielded better performance on tumor segmentation with mean Dice similarity coefficients of 0.9179 ( ± 0.025) and 0.8910 ( ± 0.042) obtained on high-grade and low-grade tumors, respectively. Conclusion The proposed method achieved higher accuracies for brain tumor extraction in comparison to other contour-based methods.

In this paper, we propose a novel multi-graph-based method for salient object detection in natural images. Starting from image decomposition via a superpixel generation algorithm, we utilize color, spatial and background label to calculate edge weight matrix of the graphs. By considering superpixels as the nodes and region similarities as the edge weights, local, global and high contrast graphs are created. Then, an integration technique is applied to form the saliency maps using degree vectors of the graphs. Extensive experiments on three challenging datasets show that the proposed unsupervised method outperforms the several different state-of-the-art unsupervised methods.

Objective Sleep stage scoring is essential for diagnosing sleep disorders. Visual scoring of sleep stages is very time-consuming and prone to human errors. In this work, we introduce an efficient approach to improve the accuracy of sleep stage scoring and classification for sleep analysis. Method In this approach, a set of optimal features was first selected from a pool of features extracted from sleep EEG epochs by using a feature selection method based on the relevance and redundancy analysis. EEG segments were then classified using a random forest classifier. Finally, a Hidden Markov Model (HMM) was used to reduce false positives by incorporating knowledge of the temporal structure of transitions between sleep stages. We evaluated the proposed method using single-channel EEG signals from four public sleep EEG datasets scored according to R&K and AASM guidelines. We compared the performance of our method with existing methods using different cross validation strategies. Results Using a leave-one-out validation strategy, our method achieved overall accuracies in the range of (79.4–87.4%) and (77.6–80.4%) with Kappa values in the range of 0.7–0.85 for six-stage (R&K) and five-stage (AASM) classification, respectively. Our method showed a reduction in overall accuracy up to 8% using the cross-dataset validation strategy in comparison with the subject cross-validation method. Comparison with existing method(s) Our method outperformed the existing methods for all multi-stage classification. Conclusions The proposed single-channel method can be used for robust and reliable sleep stage scoring with high accuracy and relatively low complexity required for real time applications.

Fusing a low spatial resolution hyperspectral image (HSI) with a high spatial resolution multispectral image (MSI) to produce a fused high spatio-spectal resolution one, referred to as HSI super-resolution, has recently attracted increasing research interests. In this paper, a new method based on coupled non-negative tensor decomposition (CNTD) is proposed. The proposed method uses tucker tensor factorization for low resolution hyperspectral image (LR-HSI) and high resolution multispectral image (HR-MSI) under the constraint of non-negative tensor ecomposition (NTD). The conventional non-negative matrix factorization (NMF) method essentially loses spatio-spectral joint structure information when stacking a 3D data into a matrix form. On the contrary, in NMF-based methods, the spectral, spatial, or their joint structures must be imposed from outside as a constraint to well pose the NMF problem, The proposed CNTD method blindly brings the advantage of preserving the spatio-spectral joint structure of HSIs. In this paper, the NTD is imposed on the coupled tensor of HIS and MSI straightly. Hence the intrinsic spatio-spectral joint structure of HSI can be losslessly expressed and interdependently exploited. Furthermore, multilinear interactions of different modes of the HSIs can be exactly modeled by means of the core tensor of the Tucker tensor decomposition. The proposed method is completely straight forward and easy to implement. Unlike the other state-of-the-art methods, the complexity of the proposed CNTD method is quite linear with the size of the HSI cube. Compared with the state-of-the-art methods experiments on two well-known datasets, give promising results with lower complexity order.

The majority of network studies of human brain structural connectivity are based on single-shell diffusion-weighted imaging (DWI) data. Recent advances in imaging hardware and software capabilities have made it possible to acquire multishell (b-values) high-quality data required for better characterization of white-matter crossing-fiber microstructures. The purpose of this study was to investigate the extent to which brain structural organization and network topology are affected by the choice of diffusion magnetic resonance imaging (MRI) acquisition strategy and parcellation scale. We performed graph-theoretical network analysis using DWI data from 35 Human Connectome Project subjects. Our study compared four single-shell (b = 1000, 3000, 5000, 10,000 s/mm2) and multishell sampling schemes and six parcellation scales (68, 200, 400, 600, 800, 1000 nodes) using five graph metrics, including small-worldness, clustering coefficient, characteristic path length, modularity and global efficiency. Rich-club analysis was also performed to explore the rich-club organization of brain structural networks. Our results showed that the parcellation scale and imaging protocol have significant effects on the network attributes, with the parcellation scale having a substantially larger effect. Regardless of the parcellation scale, the brain structural networks exhibited a rich-club organization with similar cortical distributions across the parcellation scales involving at least 400 nodes. Compared to single b-value diffusion acquisitions, the deterministic tractography using multishell diffusion imaging data consisting of shells with b-values higher than 5000 s/mm2 resulted in significantly improved fiber-tracking results at the locations where fiber bundles cross each other. Brain structural networks constructed using the multishell acquisition scheme including high b-values also exhibited significantly shorter characteristic path lengths, higher global efficiency and lower modularity. Our results showed that both parcellation scale and sampling protocol can significantly impact the rich-club organization of brain structural networks. Therefore, caution should be taken concerning the reproducibility of connectivity results with regard to the parcellation scale and sampling scheme.

The three soft brain tissues white matter (WM), gray matter (GM), and cerebral spinal fluid (CSF) identified in a magnetic resonance (MR) image via image segmentation techniques can aid in structural and functional brain analysis, brain's anatomical structures measurement and visualization, neurodegenerative disorders diagnosis, and surgical planning and image-guided interventions, but only if obtained segmentation results are correct. This paper presents a multiple-classifier-based system for automatic brain tissue segmentation from cerebral MR images. The developed system categorizes each voxel of a given MR image as GM, WM, and CSF. The algorithm consists of preprocessing, feature extraction, and supervised classification steps. In the first step, intensity non-uniformity in a given MR image is corrected and then non-brain tissues such as skull, eyeballs, and skin are removed from the image. For each voxel, statistical features and non-statistical features were computed and used a feature vector representing the voxel. Three multilayer perceptron (MLP) neural networks trained using three different datasets were used as the base classifiers of the multiple-classifier system. The output of the base classifiers was fused using majority voting scheme. Evaluation of the proposed system was performed using Brainweb simulated MR images with different noise and intensity non-uniformity and internet brain segmentation repository (IBSR) real MR images. The quantitative assessment of the proposed method using Dice, Jaccard, and conformity coefficient metrics demonstrates improvement (around 5 % for CSF) in terms of accuracy as compared to single MLP classifier and the existing methods and tools such FSL-FAST and SPM. As accurately segmenting a MR image is of paramount importance for successfully promoting the clinical application of MR image segmentation techniques, the improvement obtained by using multiple-classifier-based system is encouraging.

This study aims to develop a semi-automatic system for brain tumor segmentation in 3D MR images. For a given image, noise was corrected using SUSAN algorithm first. A specific region of interest (ROI) that contains tumor was identified and then the intensity non-uniformity in ROI was corrected via the histogram normalization and intensity scaling. Each voxel in ROI was presented using 22 features and then was categorized as tumor or non-tumor by a multiple-classifier system. T1- and T2-weighted images and fluid-attenuated inversion recovery (FLAIR) were examined. The system performance in terms of Dice index (DI), sensitivity (SE) and specificity (SP) was evaluated using 150 simulated and 30 real images from the BraTS 2012 database. The results showed that the presented system with an average DI > 0.85, SE > 0.90, and SP > 0.98 for simulated data and DI > 0.80, SE > 0.84, and SP > 0.98 for real data might be used for accurate extraction of the brain tumors. Moreover, this system is 6 times faster than a similar system that processes the whole image. In comparison with two state-of-the-art tumor segmentation methods, our system improved DI (e.g., by 0.31 for low-grade tumors) and outperformed these algorithms. Considering the costs of imaging procedures, tumor identification accuracy and computation times, the proposed system that augmented general pathological information about tumors and used only 4 features of FLAIR images can be suggested as a brain tumor segmentation system for clinical applications.

Here, a two-phase search strategy is proposed to identify the biomarkers in gene expression data set for the prostate cancer diagnosis. A statistical filtering method is initially employed to remove the noisiest data. In the first phase of the search strategy, a multi-objective optimisation based on the binary particle swarm optimisation algorithm tuned by a chaotic method is proposed to select the optimal subset of genes with the minimum number of genes and the maximum classification accuracy. Finally, in the second phase of the search strategy, the cache-based modification of the sequential forward floating selection algorithm is used to find the most discriminant genes from the optimal subset of genes selected in the first phase. The results of applying the proposed algorithm on the available challenging prostate cancer data set demonstrate that the proposed algorithm can perfectly identify the informative genes such that the classification accuracy, sensitivity, and specificity of 100% are achieved with only nine biomarkers.

Compared to adults, brain extraction from magnetic resonance (MR) images of newborn infants is particularly challenging due to their smaller brain size, which causes lower spatial resolution, lower tissue contrast and ambiguous tissue intensity distribution. In this work, a multi-atlas patch-based label fusion method is presented for automatic brain extraction from neonatal head MR images. In this method, a number of atlases are first selected uniformly among a set of training images. After nonlinear alignment of the selected atlas images to the target image, a probabilistic gray level-coded brain mask is created and used to assign brain/non-brain labels to voxels with different degree of uncertainty using a modified non-local patch-based label fusion method based on the integration of low-level and in-depth search patch selection strategies. Experiments with 40 neonates aged between 37 and 44 weeks showed an average Dice, Jaccard and Conformity coefficients of 0.993, 0.986 and 0.986, respectively. We compared the performance of the proposed method with two multi atlas-based methods, i.e. NLPB and MASS, and two popular non-learning-based methods, i.e. BSE and BET. Compared to these methods, our method achieved higher accuracy with brain masks very close to manually extracted ones and produced lower false negative and false positive rates. Our proposed method allows for accurate and efficient brain extraction, a crucial step in brain MRI applications such as brain tissue segmentation and volume estimation in neonates.

In this paper, we present a new semi-automatic brain tissue segmentation method based on a hybrid hierarchical approach that combines a brain atlas as a priori information and a least-square support vector machine (LS-SVM). The method consists of three steps. In the first two steps, the skull is removed and the cerebrospinal fluid (CSF) is extracted. These two steps are performed using the toolbox FMRIB′s automated segmentation tool integrated in the FSL software (FSL-FAST) developed in Oxford Centre for functional MRI of the brain (FMRIB). Then, in the third step, the LS-SVM is used to segment grey matter (GM) and white matter (WM). The training samples for LS-SVM are selected from the registered brain atlas. The voxel intensities and spatial positions are selected as the two feature groups for training and test. SVM as a powerful discriminator is able to handle nonlinear classification problems; however, it cannot provide posterior probability. Thus, we use a sigmoid function to map the SVM output into probabilities. The proposed method is used to segment CSF, GM and WM from the simulated magnetic resonance imaging (MRI) using Brainweb MRI simulator and real data provided by Internet Brain Segmentation Repository. The semi-automatically segmented brain tissues were evaluated by comparing to the corresponding ground truth. The Dice and Jaccard similarity coefficients, sensitivity and specificity were calculated for the quantitative validation of the results. The quantitative results show that the proposed method segments brain tissues accurately with respect to corresponding ground truth.

The newborn’s cranium is composed of flat cranial bone and fontanels forming together the envelope of the cerebral cavity. The fontanels are relatively flexible since they consist of fibrous membrane that ossifies during maturation becoming flat cranial bone as well. Fontanels give less contrast in computerized tomography (CT) images; they can be identified as gaps between the cranial bones. In this paper, we propose an automatic model-based method using variational level set to segment the skull and fontanels from CT images. In this approach, firstly a skull model consisting of cranial bones and fontanels is created and then used as constraint for level set evolution. Then, by removing the cranial bones from the segmented skulls, the fontanels are obtained. To verify the validity of the achieved results, automatically segmented skull and fontanels have been compared with the ones manually segmented by an expert using Dice similarity and Hausdorff dissimilarity measures, which show the good agreement between them. Furthermore, the surface areas of cranium and fontanel have been determined for these segmentations. The results for both, manual and automatic segmentation, are in good agreement.

This paper presents the design and construction of a 3D digital neonatal neurocranial phantom and its application for the simulation of brain magnetic resonance (MR) images. Commonly used digital brain phantoms (e.g. BrainWeb) are based on the adult brain. With the growing interest in computer-aided methods for neonatal MR image processing, there is a growing demand a digital phantom and brain MR image simulator especially for the neonatal brains. This is due to the pronounced differences between adult and neonatal brains not only in terms of size but also, more importantly, in terms of geometrical proportions and the need to subdivide white matter into two different tissue types in neonates. Therefore the neonatal brain phantom created in the here presented work consists of 9 different tissue types: skin, fat, muscle, skull, dura mater, gray matter, myelinated white matter, nonmyelinated white matter and cerebrospinal fluid. Each voxel has a vector consisting of 9 components, one for each of these nine tissue types. This digital phantom can be used to map simulated magnetic resonance signal intensities resulting in simulated MR images of the newborns head. These images with controlled degradation of the image data present a representative, reproducible data set ideal for development and evaluation of neonatal MRI analysis methods, e.g. segmentation and registration algorithms.

We present a novel system for detecting electroencephalographic transient events in neonates and older children. The detection system consists of three major elements: (i) a preprocessing stage for filtering EEG and detecting artifacts, (ii) a hierarchical course-to-fine temporal event detection stage and (iii) a hierarchical course-to-fine spatial event selection stage to incorporate spatial contextual information for rejection of spurious events. The output consists of homogeneous EEG events and their corresponding dipole clusters. The system was evaluated on EEG signals recorded in four neonates and six older children. There was a high degree of correlation between system-detected and expert-marked events for all patients. Mean sensitivities of 84.9% and 91.9% and mean selectivities of 86.3% and 90.6% were obtained for the neonates and the older children, respectively. This tool is appropriate for the detection and selection of homogeneous EEG events prior to source localization. Quantitative spatial analysis of dipoles may facilitate the physician's assessment of patients' brain dysfunction.

Commonly used brain templates are based on adults' or children's brains. In this study, we create a neonatal brain template. This becomes necessary because of the pronounced differences not only in size but even more importantly in geometrical proportions of the brains of adults and children as compared to the ones of newborns. The template is created based on high resolution T1 magnetic resonance images of 7 individuals with gestational ages between 39 and 42 weeks at the dates of examination. As usual, the created template presents two characteristics in a single image: an average intensity and an average shape. The normalization process to map subjects to the same space is done using SPM2 (Statistical Parametric Mapping) and its deformation toolbox. It consists of two steps: an affine and a nonlinear registration for global and local alignments, respectively. The template was evaluated by (i) study of anatomical local deviations and (ii) amount of local deformations of brain tissues in normalized neonatal images. The extracted results were compared with the ones obtained by normalization using adult and pediatric templates. It was shown that the application of our neonatal brain template for alignment of neonatal images results in a pronounced increase in performance of the normalization procedure as indicated by reduction of deviation of anatomical equivalent structures. The neonatal atlas template is freely downloadable from http://www.u-picardie.fr/labo/GRAMFC.

In this paper, we propose an unsupervised bottom-up method which formulates salient object detection problem as finding salient vertices of a graph. Global contrast is extracted in a novel graph-based framework to determine localization of salient objects. Saliency values are assigned to regions in terms of nodes degrees on graph. The proposed method has been applied on SED2 dataset. The qualitative and quantitative evaluation of the proposed method show that it can detect the salient objects appropriately in comparison with 5 state-of-art saliency models.

The fact that very small aspect changes can drastically alter the profile of an aircraft implies that a huge number of profiles would be required to characterize an aircraft from all possible aspects. This make a look-up table approach to classification impractical in the short-wavelength limit. A deterministic model and a conditionally Gaussian model for high-resolution range profiles (HRRP) are used, and the recognition performance under each model are compared by using one and more than one range profile. The experimental results show that assumption of Gaussian distribution by using 10 range profiles improve the recognition performance.

Since the human brain has a certain degree of asymmetry, asymmetry analysis based on an asymmetric brain template may result in ambiguity. Therefore, in this paper we present our framework to create a symmetric brain template from a group of newborns aged between 39 to 42 weeks after gestation. Consequently, the resulting template shows no difference between its original and its flipped version as is the case for an asymmetric template. The resulting symmetric template can be used e.g. for brain asymmetry analysis in language development.

A major drawback of the orthogonal frequency division multiplexing (OFDM) systems is Peak-to-Average Power Ratio (PAPR) of the transmitted signal, which leads to a very costly and inefficient straightforward implementation. A new OFDM system, which exploited three-dimensional (3-D) signal mapper and 2-D IDFT modulator, and called it 3-D OFDM, has much improved performance as compared with conventional OFDM. In this paper the PAPR of a 3-D OFDM technique have investigated and it is shown that 3-D OFDM system also improved PAPR as compared with conventional OFDM.

Neonatal skull is different from adults' skull. It is composed of ossified parts of flat bone and fontanels. Fontanels are fibrous membranes that at this stage of development connect the already ossified flat bones of the cranium. Since these two different tissue types have different electrical conductivities, it is important to model the geometry of the fontanels if one aims to solve the inverse problem as e.g. for source localization. In neonatal Computer Tomography (CT) images fontanels are identifiable as gaps between the bones forming the cranium. In this paper we propose an automatic method based on a level set algorithm with shape prior to segment skull and fontanels from neonatal CT images. Quantitative evaluation based on similarity between automatically and manually segmented skulls and fontanels using ten subjects show that the proposed method can appropriately segment skull and fontanels.

The newborn's skull is composed of already ossified parts of the flat bone connected by areas of fibrous membrane not yet ossified, which are called fontanels. At birth, an infant has six of such fontanels. These two different tissue types forming the outer part of the neuro-cranium have different electrical conductivities. Thus, it is important to determine the exact geometry of the fontanels if one aims to solve the inverse problem as e.g. for source localization. Computer Tomography (CT) imaging provides an excellent tool for the non-invasive study of bone which here can easily be identified due to its high contrast as compared to other tissue. Fontanels correspond to not yet ossified cartilage and give less contrast, thus they can be indirectly reconstructed by extrapolation for closing of the gaps between the flat bones forming the skull. In this paper, we propose an automatic model based method using level set to extract the fontanels from CT images. The automatically determined fontanels show good agreement with the manually extracted ones.

Accuracy of automated segmentation methods is of great importance in brain image analysis. In this paper, a comparison evaluation of three common software packages in brain image segmentation is presented. Performance analysis of segmentation methods integrated with the latest versions of SPM, FSL and BrainSuite is considered. In this comparison study, quantitative and qualitative assessment of segmentation algorithms is performed for both simulated and real data. Results obtained in this paper can be utilized to assist the users for selecting the appropriate software for the application of brain tissue segmentation.

This study presents an automatic model based technique for brain tissue segmentation from cerebral magnetic resonance (MR) images. In this paper, support vector machine (SVM) based classifier, as a new and powerful kind of supervised machine learning with high generalization characteristics, is employed. Here, least-square SVM (LS-SVM) in conjunction with brain probabilistic atlas as a priori information is applied to obtain class probabilities for three tissues of cerebrospinal fluid (CSF), white matter (WM) and grey matter (GM). The entire process of brain segmentation is performed in an iterative procedure, so that the probabilistic maps of brain tissues will be updated at any iteration. The quantitative and qualitative results indicate excellent performance of the applied method.

A realistic head model is needed for source localization methods used for the study of epilepsy in neonates applying Electroencephalographic (EEG) measurements from the scalp. The earliest models consider the head as a series of concentric spheres, each layer corresponding to a different tissue whose conductivity is assumed to be homogeneous. The results of the source reconstruction depend highly on the electric conductivities of the tissues forming the head.The most used model is constituted of three layers (scalp, skull, and intracranial). Most of the major bones of the neonates’ skull are ossified at birth but can slightly move relative to each other. This is due to the sutures, fibrous membranes that at this stage of development connect the already ossified flat bones of the neurocranium. These weak parts of the neurocranium are called fontanels. Thus it is important to enter the exact geometry of fontaneles and flat bone in a source reconstruction because they show pronounced in conductivity. Computer Tomography (CT) imaging provides an excellent tool for non-invasive investigation of the skull which expresses itself in high contrast to all other tissues while the fontanels only can be identified as absence of bone, gaps in the skull formed by flat bone. Therefore, the aim of this paper is to extract the fontanels from CT images applying a variational level set method. We applied the proposed method to CT-images of five different subjects. The automatically extracted fontanels show good agreement with the manually extracted ones. QC 20120312

Inappropriate results may be produced if one uses adult or pediatric atlases for evaluation of neonatal cerebral images for morphological studies. This is mainly due to anatomical particularities typical for this early stage of development. In this paper, we describe the construction of a digital neonatal brain atlas from a set of images of neonates aged between 39 and 42 weeks. It consists of probabilistic models for brain, cerebrospinal fluid (CSF) and skull. In the first step, the selected images are segmented automatically followed by manual correction. In the second step, the images are normalized to a stereotaxic space defined by the neonatal brain atlas template GRAMFC_T(39-42) using a popular normalization algorithm implemented in Statistical Parametric Mapping (SPM). The normalization parameters of individual subjects are then used to resample the corresponding brain, CSF and skull. Finally, to construct the probabilistic models, the average is computed for each voxel location. The atlas might be used for different applications such as source localization or neonatal structural image analysis.

In this paper, we present a novel automatic algorithm for scalp and skull segmentation in T1-weighted neonatal head MR images. First, the probabilistic scalp and skull atlases are constructed. Second, the scalp outer surface is extracted based on an active mesh method. Third, maximum number of boundary points corresponding to the scalp inner surface is extracted using the constructed scalp probabilistic atlas and a set of knowledge based rules. In the next step, the skull inner surface and maximum number of boundary points of the outer surface are extracted using a priori information of the head anatomy and the constructed skull probabilistic atlas. Finally, the fast sweeping, tagging and level set methods are applied to reconstruct surfaces from the detected points in three-dimensional space. The results of the new segmentation algorithm on MRI data acquired from nine newborns (including three atlas and six test subjects) were compared with manual segmented data provided by an expert radiologist. The average similarity indices for the scalp and skull segmented regions were equal to 89% and 71% for the atlas and 84% and 63% for the test data, respectively.

In this paper, an algorithm for multifocus image fusion is presented. In this algorithm, the discrete multiwavelet transform (DMWT) has been used as decomposition tool to obtain multiresolution data. A major advantage of this method over discrete wavelet transform (DWT) is its sensitivity to different kinds of object features in digital images. The developed algorithm consists of a registration step followed by a decomposition step using DMWT. Then, an optimized coefficient selection algorithm consisting of coefficient grouping and consistency verification is used. Finally, the fused image is obtained by reconstruction of the selected coefficients using inverse DMWT. The algorithm has been applied to some multifocus images. Quantitative and qualitative analyses of the results demonstrate higher performance of the algorithm compared to the same method using DWT.

This paper represents a new region-based active contour model that can be used to segment images with intensity non-uniformity and high-level noise. The main idea of our proposed method is to use Gaussian distributions with different means and variances with incorporation of intensity non-uniformity model for image segmentation. In order to integrate the spatial information between neighboring pixels in our proposed method, we use Markov Random Field. Our experiments on synthetic images and cerebral magnetic resonance images show the advantages of the proposed method over state-of-art methods, i.e. local Gaussian distribution fitting.