Emotion recognition based on group phase locking value using convolutional neural network.

Electroencephalography (EEG)-based emotion recognition is an important technology for human-computer interactions. In the field of neuromarketing, emotion recognition based on group EEG can be used to analyze the emotional states of multiple users. Previous emotion recognition experiments have been based on individual EEGs; therefore, it is difficult to use them for estimating the emotional states of multiple users. The purpose of this study is to find a data processing method that can improve the efficiency of emotion recognition. In this study, the DEAP dataset was used, which comprises EEG signals of 32 participants that were recorded as they watched 40 videos with different emotional themes. This study compared emotion recognition accuracy based on individual and group EEGs using the proposed convolutional neural network model. Based on this study, we can see that the differences of phase locking value (PLV) exist in different EEG frequency bands when subjects are in different emotional states. The results showed that an emotion recognition accuracy of up to 85% can be obtained for group EEG data by using the proposed model. It means that using group EEG data can effectively improve the efficiency of emotion recognition. Moreover, the significant emotion recognition accuracy for multiple users achieved in this study can contribute to research on handling group human emotional states.

Comprehensive Steroid Assay with Non-Targeted Analysis Using Liquid Chromatography Ion Mobility Mass Spectrometry.

Aldosterone-producing adenomas (APAs) have different steroid profiles in serum, depending on the causative genetic mutation. Ion mobility is a separation technique for gas-phase ions based on their m/z values, shapes, and sizes. Human serum (100 µL) was purified by liquid-liquid extraction using tert-butyl methyl ether/ethyl acetate at 1/1 (v/v) and mixed with deuterium-labeled steroids as the internal standard. The separated supernatant was dried, re-dissolved in water containing 20% methanol, and injected into a liquid chromatography-ion mobility-mass spectrometer (LC/IM/MS). We established a highly sensitive assay system by separating 20 steroids based on their retention time, m/z value, and drift time. Twenty steroids were measured in the serum of patients with primary aldosteronism, essential hypertension, and healthy subjects and were clearly classified using principal component analysis. This method was also able to detect phosphatidylcholine and phosphatidylethanolamine, which were not targeted. LC/IM/MS has a high selectivity for known compounds and has the potential to provide information on unknown compounds. This analytical method has the potential to elucidate the pathogenesis of APA and identify unknown steroids that could serve as biomarkers for APA with different genetic mutations.

Classification of grazing-incidence small-angle X-ray scattering patterns by convolutional neural network.

Grazing-incidence small-angle X-ray scattering (GISAXS) patterns have multiple superimposed contributions from the shape of the nanoscale structure, the coupling between the particles, the partial pair correlation, and the layer geometry. Therefore, it is not easy to identify the model manually from the huge amounts of combinations. The convolutional neural network (CNN), which is one of the artificial neural networks, can find regularities to classify patterns from large amounts of combinations. CNN was applied to classify GISAXS patterns, focusing on the shape of the nanoparticles. The network found regularities from the GISAXS patterns and showed a success rate of about 90% for the classification. This method can efficiently classify a large amount of experimental GISAXS patterns according to a set of model shapes and their combinations.

A study on EEG quality in physical movements with Steady-State Visual Evoked Potentials.

In this paper, we investigated the quality of ElectroEncephaloGraphic (EEG) signals during performing physical movements. By using a portable EEG device, the Steady-State Visual Evoked Potential (SSVEP) was recorded on parietal and occipital locations. The SSVEP induced by flickering stimuli was successfully observed in the self-paced mimic walking conditions as well as in the sitting conditions. To see the dependence of temporal and spatial filters on the potential performance of Brain-Computer Interfaces (BCI) we applied the signal processing of Principal Component Analysis and Linear Discriminant Analysis. The pattern recognition performances in inferring the subject's eye gaze directions from the EEG signals could be perfect even in the self-paced mimic walking conditions. It was found that three electrodes on parieto-occipital and occipital locations were essential in order to have perfect performances. From these results, we conclude that the applications using SSVEP-based BCI can be realized even in the physically moving context.

Non-target photo images in oddball paradigm improve EEG-based personal identification rates.

A research on biometry based on human brain activities has lately been emerging. In this study, we investigate the feasibility of personal identification using one-channel electroencephalogram during photo retrieval in oddball paradigm. The use of non-target photo images was examined to improve the identification performances. Nine photo images were randomly presented one after another to five subjects. The Principal Component Analysis and the Linear Discriminant Analysis were applied for the signal processing. With EEG activities both during target and non-target photo retrieval, the algorithm successfully improved the identification rates. The rates were 87.2, 95.0, and 97.6% using 5, 10, and 20-time averaging, respectively. The performances with EEG only during target photo retrieval were lower by 5-13%. This study reveals a future possibility of photo retrieval tasks to realize the personal identification using human brain activities, which will yield rich controls of machine for the users of brain computer-interface.

Extracting alpha band modulation during visual spatial attention without flickering stimuli using common spatial pattern.

In this paper, alpha band modulation during visual spatial attention without visual stimuli was focused. Visual spatial attention has been expected to provide a new channel of non-invasive independent brain computer interface (BCI), but little work has been done on the new interfacing method. The flickering stimuli used in previous work cause a decline of independency and have difficulties in a practical use. Therefore we investigated whether visual spatial attention could be detected without such stimuli. Further, the common spatial patterns (CSP) were for the first time applied to the brain states during visual spatial attention. The performance evaluation was based on three brain states of left, right and center direction attention. The 30-channel scalp electroencephalographic (EEG) signals over occipital cortex were recorded for five subjects. Without CSP, the analyses made 66.44 (range 55.42 to 72.27) % of average classification performance in discriminating left and right attention classes. With CSP, the averaged classification accuracy was 75.39 (range 63.75 to 86.13) %. It is suggested that CSP is useful in the context of visual spatial attention, and the alpha band modulation during visual spatial attention without flickering stimuli has the possibility of a new channel for independent BCI as well as motor imagery.