Event-related potential (ERP)-based P300 spellers are commonly used in the field of brain-computer interfaces as an alternative channel of communication for people with severe neuro-muscular diseases. This study introduces a novel P300 based brain-computer interface (BCI) stimulus paradigm using a random set presentation pattern and exploiting the effects of face familiarity. The effect of face familiarity is widely studied in the cognitive neurosciences and has recently been addressed for the purpose of BCI. In this study we compare P300-based BCI performances of a conventional row-column (RC)-based paradigm with our approach that combines a random set presentation paradigm with (non-) self-face stimuli. Our experimental results indicate stronger deflections of the ERPs in response to face stimuli, which are further enhanced when using the self-face images, and thereby improving P300-based spelling performance. This lead to a significant reduction of stimulus sequences required for correct character classification. These findings demonstrate a promising new approach for improving the speed and thus fluency of BCI-enhanced communication with the widely used P300-based BCI setup.
How brain dynamics change across conscious states, including reliable signatures of the transitions between unconsciousness and consciousness, remains unclear. In this work, we addressed the changes in functional brain networks during self-titrated midazolam sedation using high-density electroencephalography (EEG) in ten subjects. We were particularly interested in the underlying network alterations, identified with graph theory, associated with transitions between states of consciousness. The weighted Phase Lag Index (wPLI) was used as the connectivity estimator between two signals. Based on wPLI, we calculated network properties such as characteristic path length, clustering coefficient, and small-worldness for measuring the integration and segregation of the brain network. We found significant changes in power and wPLI at different levels of consciousness. During unconsciousness, wPLI over the parietal region was higher in the delta band (1-4Hz). The frontal-parietal interaction in the delta band was also stronger during the transition from consciousness to unconsciousness. There was the significant difference of wPLI over the frontal region between consciousness and unconsciousness in the sigma band (12-15Hz). The topological properties across conscious states were significantly changed in the delta band and sigma band. Our results showed parietal brain dynamics is associated with consciousness. Our data also suggest that reversible changes in delta power and connectivity underlie changes in conscious state.
Electroencephalogram (EEG) measurement could help to distinguish the level of consciousness in an individual without requiring a behavioral response, which could be useful as a diagnostic aid in patients with disorders of consciousness. In this study, we explored the EEG-evoked perturbation and analyzed consciousness using event-related spectral perturbation and convolutional neural network. We observed a novel EEG neurophysiological signature that can be used to monitor brain activity during unconsciousness. Also, the performance accuracy in the parietal region was higher than in the frontal region. The sensitivity for conscious experience was 90.9%, whereas sensitivity for unconscious experience was at the chance level in the parietal region. These results could be evidence for the importance of the posterior hot zone and could help shed light on the internal neural dynamics related to conscious experience.
Despite of the remarkable performance, modern deep neural networks are inevitably accompanied with a significant amount of computational cost for learning and deployment, which may be incompatible with their usage on edge devices. Recent efforts to reduce these overheads involves pruning and decomposing the parameters of various layers without performance deterioration. Inspired by several decomposition studies, in this paper, we propose a novel energy-aware pruning method that quantifies the importance of each filter in the network using nuclear-norm (NN). Proposed energy-aware pruning leads to state-of-the art performance for Top-1 accuracy, FLOPs, and parameter reduction across a wide range of scenarios with multiple network architectures on CIFAR-10 and ImageNet after fine-grained classification tasks. On toy experiment, despite of no fine-tuning, we can visually observe that NN not only has little change in decision boundaries across classes, but also clearly outperforms previous popular criteria. We achieve competitive results with 40.4/49.8% of FLOPs and 45.9/52.9% of parameter reduction with 94.13/94.61% in the Top-1 accuracy with ResNet-56/110 on CIFAR-10, respectively. In addition, our observations are consistent for a variety of different pruning setting in terms of data size as well as data quality which can be emphasized in the stability of the acceleration and compression with negligible accuracy loss. Our code is available at https:// github.com/ nota-github/ nota-pruning-rank.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.