Search

Automatic epileptic seizure detection based on electroencephalogram is crucial to epilepsy diagnosis and treatment. However, the large numbers of time series make it quite challenging to establish a high performance automatic detection method. Considering different physiological states of the brain could be characterized by distinct combinations or interactions of similar discontinuous local temporal patterns, a novel framework based on biclustering for automatic epileptic seizure detection is proposed in this paper. First, the CC algorithm is used to identify similar discontinuous local temporal patterns. Then, the bicluster membership matrix using a new similarity measurement is constructed to reduce the dimensionality. At last, the ELM classifier is adopted to discriminate between epileptic seizure and seizure-free EEG signals. With extensive comparative studies and evaluations on the publicly available Bonn epileptic EEG dataset, it indicates that the proposed framework could not only automatically detect or predict an epilepsy seizure with high performances with respect to accuracy, robustness and efficiency, but also implicitly provide valuable knowledge for studying the mechanisms of epilepsy.

BN-SiO2-SiAlON composite ceramics were successfully prepared by the means of hot pressed sintering. Xe plasma flow generated by Hall Thruster was used for sputtering the surface of the samples in order to evaluate the plasma erosion resistance. XRD, TEM, SEM, and LSCM were used to characterize the phase composition and morphologies of as-made composite ceramics before and after Xe plasma erosion. The ceramics were composed of h-BN, fused silica, and SiAlON, which maintained structural stability during the process of Xe plasma sputtering. In conclusion, comparing with BN-SiO2 composite ceramics, the plasma erosion rate of BN-SiO2-SiAlON composite ceramics decreases significantly at first then rises with the increase of AlN addition. Erosion pits can be observed by using SEM on the surface after plasma sputtering, which demonstrates that the BN grains have dropped off the surface. In addition, mechanical denudation by high-speed Xe ions is recognized as the injury mechanism for the BN-matrix composite materials.

Ablation mechanism and properties of in-situ SiAlON reinforced BN–SiO 2 ceramic composites have been carefully studied under different ablation time. Results showed that the ablation properties depend on the ablation time. When the ablation time increases from 10 s to 60 s, the mass ablation rate increases from 0.20 mg/s to 6.24 mg/s and the linear ablation rate increases from 9.0 μm/s to 28.1 μm/s. The phases of ablation products mainly consist of amorphous SiO 2 , Mullite, Corundum and B 2 O. Four time-dependent ablation behaviors contributed to the various ablation regions: oxidation of matrix which includes BN and Si 2 Al 3 O 7 N; evaporation of B 2 O 3 ( l ) and softening of SiO 2 ( s ); scouring away of SiO 2 ( l ) and Al 2 O 3 ( l ) by high-speed gas stream; evaporation of SiO 2 ( l ) and Al 2 O 3 ( l ).

Automatic detection of epileptic seizure plays an important role in the diagnosis of epilepsy for it can obtain invisible information of epileptic electroencephalogram (EEG) signals exactly and reduce the heavy burdens of doctors efficiently. Current automatic detection technologies are almost shallow learning models that are insufficient to learn the complex and non-stationary epileptic EEG signals. Moreover, most of their feature extraction or feature selection methods are supervised and depend on domain-specific expertise. To solve these problems, we proposed a novel framework for the automatic detection of epileptic EEG by using stacked sparse autoencoder (SSAE) with a softmax classifier. The proposed framework firstly learns the sparse and high level representations from the preprocessed data via SSAE, and then send these representations into softmax classifier for training and classification. To verify the performance of this framework, we adopted the epileptic EEG datasets to conduct experiments. The simulation results with an average accuracy of 96 % illustrated the effectiveness of the proposed framework.

Large Average Sub matrices (LAS) is one of the biclustering algorithms, which can capture large average sub matrices within a high dimensional data matrix. It has gained increasing popularity in many fields such as biological data analysis and financial forecasting. However, due to urgent requirements for high performance in large scale data processing applications, high performance parallel solutions for LAS biclustering are highly desirable. In this paper, we propose an efficient parallel large average submatrices biclustering based on MapReduce. Furthermore, we boost the search efficiency of LAS by using heap sort. Experimental results demonstrate that the presented parallel algorithm has advantages of both high speedup and good scalability.