Embedded EEG Feature Selection for Multi-Dimension Emotion Recognition via Local and Global Label Relevance.

Due to the problem of a small amount of EEG samples and relatively high dimensionality of electroencephalogram (EEG) features, feature selection plays an essential role in EEG-based emotion recognition. However, current EEG-based emotion recognition studies utilize a problem transformation approach to transform multi-dimension emotional labels into single-dimension labels, and then implement commonly used single-label feature selection methods to search feature subsets, which ignores the relations between different emotional dimensions. To tackle the problem, we propose an efficient EEG feature selection method for multi-dimension emotion recognition (EFSMDER) via local and global label relevance. First, to capture the local label correlations, EFSMDER implements orthogonal regression to map the original EEG feature space into a low-dimension space. Then, it employs the global label correlations in the original multi-dimension emotional label space to effectively construct the label information in the low-dimension space. With the aid of local and global relevance information, EFSMDER can conduct representational EEG feature subset selection. Three EEG emotional databases with multi-dimension emotional labels were used for performance comparison between EFSMDER and fourteen state-of-the-art methods, and the EFSMDER method achieves the best multi-dimension classification accuracies of 86.43, 84.80, and 97.86 percent on the DREAMER, DEAP, and HDED datasets, respectively.

A Multi-Source Transfer Joint Matching Method for Inter-Subject Motor Imagery Decoding.

Individual differences among different subjects pose a great challenge to motor imagery (MI) decoding. Multi-source transfer learning (MSTL) is one of the most promising ways to reduce individual differences, which can utilize rich information and align the data distribution among different subjects. However, most MSTL methods in MI-BCI combine all data in the source subjects into a single mixed domain, which will ignore the effect of important samples and the large differences in multiple source subjects. To address these issues, we introduce transfer joint matching and improve it to multi-source transfer joint matching (MSTJM) and weighted MSTJM (wMSTJM). Different from previous MSTL methods in MI, our methods align the data distribution for each pair of subjects, and then integrate the results by decision fusion. Besides that, we design an inter-subject MI decoding framework to verify the effectiveness of these two MSTL algorithms. It mainly consists of three modules: covariance matrix centroid alignment in the Riemannian space, source selection in the Euclidean space after tangent space mapping to reduce negative transfer and computation overhead, and further distribution alignment by MSTJM or wMSTJM. The superiority of this framework is verified on two common public MI datasets from BCI competition IV. The average classification accuracy of the MSTJM and wMSTJ methods outperformed other state-of-the-art methods by at least 4.24% and 2.62% respectively. It's promising to advance the practical applications of MI-BCI.

Position-independent gesture recognition using sEMG signals via canonical correlation analysis.

Gesture recognition based on surface electromyogram (sEMG) signals has drawn significant attention and obtained satisfactory achievement in the field of human-computer interaction. However, the same gesture performed with different arm positions tends not to generate the same sEMG signals. Traditional solutions can be divided into two types. One type treats the same gesture with different arm positions as the same type, leading to a relatively low classification rate. The other type adopts a gesture classifier followed by the position classifier, which will achieve a satisfactory classification accuracy but at the expenses of high training burdens. To address these issues, we propose a novel framework to explore the intrinsic position independent (PI) characteristics of sEMG signals generated from the same gesture with different arm positions by canonical correlation analysis (CCA), termed as PICCA. In this framework, with the bridge link of the predefined expert set, both the training set and the testing set can be mapped into a unified-style with CCA, and hence, the classification accuracy can be improved in both user-dependent and user-independent manners. Experimental results on 13 gestures with 3 arm positions indicate that the proposed PICCA can achieve higher classification rates than those without CCA (with 28.52% and 44.19% promotions during user-dependent and user-independent manners respectively) while maintaining acceptable training burdens. These improvements will facilitate the practical implementation of myoelectric interfaces.

Association of glutathione S-transferase T1, M1 and P1 polymorphisms in the breast cancer risk: a meta-analysis in Asian population.

BACKGROUND: Published data regarding the associations between glutathione S-transferase (GST) T1, M1 and P1 polymorphisms and breast cancer risk are inconclusive. The aim of this study is to comprehensively evaluate the genetic risk of GST genes for breast cancer. MATERIALS AND METHODS: A systematic literature search was carried out in Pubmed, Medline (Ovid), Embase, CBM, CNKI, Weipu, and Wanfang database, covering all publications (last search was performed on May 20, 2015). Statistical analysis was performed using Revman 5.2 and STATA 12.0 softwares. RESULTS: A total of 12,035 cases and 13,911 controls in 34 case-control studies were included in this meta-analysis. The results suggested that the GSTM1 and GSTP1 polymorphisms can obviously increase the risk of breast cancer in Asian population (odds ratio (OR) = 1.18, 95% confidence interval (CI) = 1.04-1.33, P = 0.008 and OR = 1.23, 95% CI = 1.07-1.41, P = 0.003, respectively), especially in East Asian (OR = 1.14, 95% CI = 1.01-1.27, P = 0.03 and OR = 1.15, 95% CI = 1.03-1.28, P = 0.01, respectively) and hospital-based case-control study (HCC) group (OR = 1.32, 95% CI = 1.11-1.56, P = 0.001 and OR = 1.38, 95% CI = 1.03-1.84, P = 0.03, respectively), while the association between GSTT1 null genotype and breast cancer risk is not significant (OR = 1.08, 95% CI = 0.93-1.25, P = 0.3). CONCLUSIONS: This meta-analysis indicated that the GSTM1 and GSTP1 polymorphisms might significantly contribute to breast cancer susceptibility in Asian population, especially in East Asian, while the GSTT1 polymorphism might not be associated with breast cancer.

The P275A Polymorphism in the Macrophage Scavenger Receptor 1 Gene and Prostate Cancer Risk: a Meta-Analysis.

BACKGROUND: Published data regarding associations between the P275A polymorphism in the macrophage scavenger receptor 1 (MSR1) gene and prostate cancer (PCa) risk are inconclusive. The aim of this study was to comprehensively evaluate the genetic risk of P275A polymorphism in MSR1 gene for PCa. MATERIALS AND METHODS: A systematic literature search was carried out in Pubmed, Medline (Ovid), Embase, CBM, CNKI, Weipu, and Wanfang databases, covering all available publications (last search was performed on Apr 27, 2015). Statistical analysis was performed using Revman 5.2 and STATA 10.1 software. RESULTS: A total of 5,017 cases and 4,869 controls in 12 case-control studies were included in this meta-analysis. When all groups were pooled, there was no evidence that the P275A polymorphism had a significant association with PCa under dominant (OR=0.93, 95%CI=0.81-1.06, and p=0.28), co-dominant (homogeneous OR=0.97, 95%CI=0.56-1.68, and p=0.92; heterogeneous OR=0.93, 95%CI=0.74-1.15, and p=0.49), recessive (OR=1.10, 95%CI=0.65-1.87, and p=0.73), over-dominant (OR=0.93, 95%CI=0.75-1.15, and p=0.50), and allelic (OR=0.95, 95%CI=0.77-1.16, and p=0.61) genetic models. For stratified analyses by ethnicity and study design, no significant associations were found in the white race, the yellow race, the black race and mixed ethnicity, and the population-based case-control (PCC) and hospital-based case-control (HCC) studies under all genetic models. CONCLUSIONS: Based on our meta-analysis, the P275A polymorphism in the MSR1 gene is unlikely to be a risk factor for PCa.

Effects of Ru and Ag cap layers on microstructure and magnetic properties of FePt ultrathin films.

The effects of Ru and Ag cap layers on the microstructure and magnetic properties of the FePt ultrathin films have been investigated. The results indicate that i) The Ag cap layer segregates from the FePt/Ag bilayer, lowers the FePt ordering temperature, promotes the FePt thin films to form island structure, and enhances the coercivity; ii) The Ru cap layer increases the FePt ordering temperature, helps to maintain smooth continuous structure film, and restrains the FePt (001) orientation and perpendicular magnetic anisotropy (PMA). The effects become more pronounced for the 3-nm-thick FePt thin films. The effects can be mainly attributed to the different melting point and thermal expansion stress between the cap layer and FePt thin films.

Quantitative analysis of the magnetic domain structure in polycrystalline La(0.7)Sr(0.3)MnO3 thin films by magnetic force microscopy.

The nanoscale magnetic domain structure of the polycrystalline La(0.7)Sr(0.3)MnO(3) granular thin films was imaged with a developed magnetic force microscopy technique by simultaneously detecting both the perpendicular and in-plane components of magnetic field gradients during the same scan of the tip oscillation. The characteristics of both the perpendicular and in-plane magnetic field gradient at the grain edges or the nonmagnetic grain boundary phase for LSMO films were demonstrated and can be used to evaluate the magnetic domain structure and magnetic isolation between neighboring grains. A two dimensional signal transformation algorithm to reconstruct the in-plane magnetization distribution of the polycrystalline LSMO thin films from the measured raw MFM images with the aid of the deconvolution technique was presented. The comparison between the experimental and simulated MFM images indicates that the magnetic grains or clusters are in the single domain (SD) or multi-domain (MD) state with the magnetic moments parallel or anti-parallel to the effective magnetic field of each grain, possibly due to the need for minimizing the total energy. The quantitative interpretation of the magnetic domain structure indicates that the large magnetoresistance in the studied LSMO films is mainly due to tunnel effect and scattering of conducted electrons at the nonmagnetic grain boundary phase related to the different configurations of magnetic domain states between neighboring grains.