Classification of Electroencephalogram Signals Using LSTM and SVM Based on Fast Walsh-Hadamard Transform

Classification of electroencephalogram (EEG) signals for humans can be achieved via artificial intelligence (AI) techniques. Especially, the EEG signals associated with seizure epilepsy can be detected to distinguish between epileptic and non-epileptic regions. From this perspective, an automated AI technique with a digital processing method can be used to improve these signals. This paper proposes two classifiers: long short-term memory (LSTM) and support vector machine (SVM) for the classification of seizure and non-seizure EEG signals. These classifiers are applied to a public dataset, namely the University of Bonn, which consists of 2 classes –seizure and non-seizure. In addition, a fast Walsh-Hadamard Transform (FWHT) technique is implemented to analyze the EEG signals within the recurrence space of the brain. Thus, Hadamard coefficients of the EEG signals are obtained via the FWHT. Moreover, the FWHT is contributed to generate an efficient derivation of seizure EEG recordings from non-seizure EEG recordings. Also, a k-fold cross-validation technique is applied to validate the performance of the proposed classifiers. The LSTM classifier provides the best performance, with a testing accuracy of 99.00%. The training and testing loss rates for the LSTM are 0.0029 and 0.0602, respectively, while the weighted average precision, recall, and F1-score for the LSTM are 99.00%. The results of the SVM classifier in terms of accuracy, sensitivity, and specificity reached 91%, 93.52%, and 91.3%, respectively. The computational time consumed for the training of the LSTM and SVM is 2000 and 2500 s, respectively. The results show that the LSTM classifier provides better performance than SVM in the classification of EEG signals. Eventually, the proposed classifiers provide high classification accuracy compared to previously published classifiers.     

Enhancing level set brain tumor segmentation using fuzzy shape prior information and deep learning

Magnetic resonance imaging (MRI) brain tumor segmentation is a crucial task for clinical treatment. However, it is challenging owing to variations in type, size, and location of tumors. In addition, anatomical variation in individuals, intensity non-uniformity, and noises adversely affect brain tumor segmentation. To address these challenges, an automatic region-based brain tumor segmentation approach is presented in this paper which combines fuzzy shape prior term and deep learning. We define a new energy function in which an Adaptively Regularized Kernel-Based Fuzzy C-Means (ARKFCM) Clustering algorithm is utilized for inferring the shape of the tumor to be embedded into the level set method. In this way, some shortcomings of traditional level set methods such as contour leakage and shrinkage have been eliminated. Moreover, a fully automated method is achieved by using U-Net to obtain the initial contour, reducing sensitivity to initial contour selection. The proposed method is validated on the BraTS 2017 benchmark dataset for brain tumor segmentation. Average values of Dice, Jaccard, Sensitivity and specificity are 0.93 ± 0.03, 0.86 ± 0.06, 0.95 ± 0.04, and 0.99 ± 0.003, respectively. Experimental results indicate that the proposed method outperforms the other state-of-the-art methods in brain tumor segmentation.     

Adaptive robust control of uncertain systems with state and input delay

In this paper, adaptive robust control of uncertain systems with multiple time delays in states and input is considered. It is assumed that the parameter uncertainties are time varying norm-bounded whose bounds are unknown but their functional properties are known. To overcome the effect of input delay on the closed loop system stability, new Lyapunov Krasovskii functional will be introduced. It is shown that the proposed adaptive robust controller guarantees globally uniformly exponentially convergence of all system solutions to a ball with any certain convergence rate. Moreover, if there is no disturbance in the system, asymptotic stability of the closed loop system will be established. The proposed design condition is formulated in terms of linear matrix inequality (LMI) which can be easily solved by LMI Toolbox in Matlab. Finally, an illustrative example is included to show the effectiveness of results developed in this paper.     

A parameter-tuned genetic algorithm for the resource investment problem with discounted cash flows and generalized precedence relations

A resource investment problem with discounted cash flows (RIPDCF) is a project-scheduling problem in which (a) the availability levels of the resources are considered decision variables and (b) the goal is to find a schedule such that the net present value of the project cash flows optimizes. In this paper, the RIPDCF in which the activities are subject to generalized precedence relations is first modeled. Then, a genetic algorithm (GA) is proposed to solve this model. In addition, design of experiments and response surface methodology are employed to both tune the GA parameters and to evaluate the performance of the proposed method in 240 test problems. The results of the performance analysis show that the efficiency of the proposed GA method is relatively well.     

A domain decomposition strategy for alignment of multiple biological sequences on multiprocessor platforms

Multiple Sequences Alignment (MSA) of biological sequences is a fundamental problem in computational biology due to its critical significance in wide ranging applications including haplotype reconstruction, sequence homology, phylogenetic analysis, and prediction of evolutionary origins. The MSA problem is considered NP-hard and known heuristics for the problem do not scale well with increasing numbers of sequences. On the other hand, with the advent of a new breed of fast sequencing techniques it is now possible to generate thousands of sequences very quickly. For rapid sequence analysis, it is therefore desirable to develop fast MSA algorithms that scale well with an increase in the dataset size. In this paper, we present a novel domain decomposition based technique to solve the MSA problem on multiprocessing platforms. The domain decomposition based technique, in addition to yielding better quality, gives enormous advantages in terms of execution time and memory requirements. The proposed strategy allows one to decrease the time complexity of any known heuristic of O(N)^x$O{\left(N\right)}^x$ complexity by a factor of O(1/p)^x$O{(1/p)}^x$, where N$N$ is the number of sequences, x$x$ depends on the underlying heuristic approach, and p$p$ is the number of processing nodes. In particular, we propose a highly scalable algorithm, Sample-Align-D, for aligning biological sequences using Muscle system as the underlying heuristic. The proposed algorithm has been implemented on a cluster of workstations using the MPI library. Experimental results for different problem sizes are analyzed in terms of quality of alignment, execution time and speed-up.     

Dynamic hybrid automatic repeat request (DHARQ) for WiMAX – Mobile multihop relay using adaptive power control

Wireless multihop network is currently attracting much attention as a new wireless broadband access technology due to numerous benefits. This work proposes a power control scheme for WiMAX multihop relay system. In contrast to existing power control and optimization approaches, our proposed method uses an adaptive Channel Quality Measurement for a relay station to reduce interferences to other mobile station (MS) or relay station (RS) within the same cell and hence increase the number of hops per link and consequently maximize the spatial reuse. The proposed power control is applied to a new dynamic HARQ algorithm for adaptive channel quality enhancement. Simulation results have indicated that the proposed approach achieves superior BER/PER performance enhancement in comparison to previous related works.     

Short-term and long-term memory analysis of learning using 2D and 3D educational contents

ABSTRACT

The effect of 2D and 3D educational content learning on memory has been studied using electroencephalography (EEG) brain signal. A hypothesis is set that the 3D materials are better than the 2D materials for learning and memory recall. To test the hypothesis, we proposed a classification system that will predict true or false recall for short-term memory (STM) and long-term memory (LTM) after learning by either 2D or 3D educational contents. For this purpose, EEG brain signals are recorded during learning and testing; the signals are then analysed in the time domain using different types of features in various frequency bands. The features are then fed into a support vector machine (SVM)-based classifier. The experimental results indicate that the learning and memory recall using 2D and 3D contents do not have significant differences for both the STM and the LTM.     

SimClus: an effective algorithm for clustering with a lower bound on similarity

Clustering algorithms generally accept a parameter k from the user, which determines the number of clusters sought. However, in many application domains, like document categorization, social network clustering, and frequent pattern summarization, the proper value of k is difficult to guess. An alternative clustering formulation that does not require k is to impose a lower bound on the similarity between an object and its corresponding cluster representative. Such a formulation chooses exactly one representative for every cluster and minimizes the representative count. It has many additional benefits. For instance, it supports overlapping clusters in a natural way. Moreover, for every cluster, it selects a representative object, which can be effectively used in summarization or semi-supervised classification task. In this work, we propose an algorithm, SimClus, for clustering with lower bound on similarity. It achieves a O(log n) approximation bound on the number of clusters, whereas for the best previous algorithm the bound can be as poor as O(n). Experiments on real and synthetic data sets show that our algorithm produces more than 40% fewer representative objects, yet offers the same or better clustering quality. We also propose a dynamic variant of the algorithm, which can be effectively used in an on-line setting.     

A hermetic glass-silicon package formed using localizedaluminum/silicon-glass bonding

A hermetic package based on localized aluminum/silicon-to-glass bonding has been successfully demonstrated. Less than 0.2 MPa contact pressure with 46 mA current input for two parallel 3.5-μm-wide polysilicon on-chip microheaters can raise the temperature of the bonding region to 700°C bonding temperature and achieve a strong and reliable bond in 7.5 min. The formation of aluminum oxide with silicon precipitate composite layer is believed to be the source of the strong bond. Accelerated testing in an autoclave shows some packages survive more than 450 h under 3 atm, 100% RH and 128°C. Premature failure has been attributed to some unbonded regions on the failed samples. The bonding yield and reliability have been improved by increasing bonding time and applied pressure