Search

The rapid serial visual presentation-based brain–computer interface (RSVP-BCI) system achieves the recognition of target images by extracting event-related potential (ERP) features from electroencephalogram (EEG) signals and then building target classification models. Currently, how to reduce the training and calibration time for classification models across different subjects is a crucial issue in the practical application of RSVP. To address this issue, a zero-calibration (ZC) method termed Attention-ProNet, which involves meta-learning with a prototype network integrating multiple attention mechanisms, was proposed in this study. In particular, multiscale attention mechanisms were used for efficient EEG feature extraction. Furthermore, a hybrid attention mechanism was introduced to enhance model generalization, and attempts were made to incorporate suitable data augmentation and channel selection methods to develop an innovative and high-performance ZC RSVP-BCI decoding model algorithm. The experimental results demonstrated that our method achieved a balance accuracy (BA) of 86.33% in the decoding task for new subjects. Moreover, appropriate channel selection and data augmentation methods further enhanced the performance of the network by affording an additional 2.3% increase in BA. The model generated by the meta-learning prototype network Attention-ProNet, which incorporates multiple attention mechanisms, allows for the efficient and accurate decoding of new subjects without the need for recalibration or retraining. [ABSTRACT FROM AUTHOR]

Directly applying brain signals to operate a mobile manned platform, such as a vehicle, may help people with neuromuscular disorders regain their driving ability. In this paper, we developed a novel electroencephalogram (EEG) signal-based driver–vehicle interface (DVI) for the continuous and asynchronous control of brain-controlled vehicles. The proposed DVI consists of the user interface, the command decoding algorithm, and the control model. The user interface is designed to present the control commands and induce the corresponding brain patterns. The command decoding algorithm is developed to decode the control command. The control model is built to convert the decoded commands to control signals. Offline experimental results show that the developed DVI can generate a motion control command with an accuracy of 83.59% and a detection time of about 2 s, while it has a recognition accuracy of 90.06% in idle states. A real-time brain-controlled simulated vehicle based on the DVI was developed and tested on a U-turn road. Experimental results show the feasibility of the DVI for continuously and asynchronously controlling a vehicle. This work not only advances the research on brain-controlled vehicles but also provides valuable insights into driver–vehicle interfaces, multimodal interaction, and intelligent vehicles. [ABSTRACT FROM AUTHOR]

Exploring the effective signal features of electroencephalogram (EEG) signals is an important issue in the research of brain–computer interface (BCI), and the results can reveal the motor intentions that trigger electrical changes in the brain, which has broad research prospects for feature extraction from EEG data. In contrast to previous EEG decoding methods that are based solely on a convolutional neural network, the traditional convolutional classification algorithm is optimized by combining a transformer mechanism with a constructed end-to-end EEG signal decoding algorithm based on swarm intelligence theory and virtual adversarial training. The use of a self-attention mechanism is studied to expand the receptive field of EEG signals to global dependence and train the neural network by optimizing the global parameters in the model. The proposed model is evaluated on a real-world public dataset and achieves the highest average accuracy of 63.56% in cross-subject experiments, which is significantly higher than that found for recently published algorithms. Additionally, good performance is achieved in decoding motor intentions. The experimental results show that the proposed classification framework promotes the global connection and optimization of EEG signals, which can be further applied to other BCI tasks. [ABSTRACT FROM AUTHOR]

The identification of drug–drug and chemical–protein interactions is essential for understanding unpredictable changes in the pharmacological effects of drugs and mechanisms of diseases and developing therapeutic drugs. In this study, we extract drug-related interactions from the DDI (Drug–Drug Interaction) Extraction-2013 Shared Task dataset and the BioCreative ChemProt (Chemical–Protein) dataset using various transfer transformers. We propose BERTGAT that uses a graph attention network (GAT) to take into account the local structure of sentences and embedding features of nodes under the self-attention scheme and investigate whether incorporating syntactic structure can help relation extraction. In addition, we suggest T5slim_dec, which adapts the autoregressive generation task of the T5 (text-to-text transfer transformer) to the relation classification problem by removing the self-attention layer in the decoder block. Furthermore, we evaluated the potential of biomedical relation extraction of GPT-3 (Generative Pre-trained Transformer) using GPT-3 variant models. As a result, T5slim_dec, which is a model with a tailored decoder designed for classification problems within the T5 architecture, demonstrated very promising performances for both tasks. We achieved an accuracy of 91.15% in the DDI dataset and an accuracy of 94.29% for the CPR (Chemical–Protein Relation) class group in ChemProt dataset. However, BERTGAT did not show a significant performance improvement in the aspect of relation extraction. We demonstrated that transformer-based approaches focused only on relationships between words are implicitly eligible to understand language well without additional knowledge such as structural information. [ABSTRACT FROM AUTHOR]

In recent years, deep learning has achieved good results in the semantic segmentation of medical images. A typical architecture for segmentation networks is an encoder–decoder structure. However, the design of the segmentation networks is fragmented and lacks a mathematical explanation. Consequently, segmentation networks are inefficient and less generalizable across different organs. To solve these problems, we reconstructed the segmentation network based on mathematical methods. We introduced the dynamical systems view into semantic segmentation and proposed a novel segmentation network based on Runge–Kutta methods, referred to hereafter as the Runge–Kutta segmentation network (RKSeg). RKSegs were evaluated on ten organ image datasets from the Medical Segmentation Decathlon. The experimental results show that RKSegs far outperform other segmentation networks. RKSegs use few parameters and short inference time, yet they can achieve competitive or even better segmentation results compared to other models. RKSegs pioneer a new architectural design pattern for segmentation networks. [ABSTRACT FROM AUTHOR]

Cervical cancer is one of the most common cancers that threaten women's lives, and its early screening is of great significance for the prevention and treatment of cervical diseases. Pathologically, the accurate segmentation of cervical cells plays a crucial role in the diagnosis of cervical cancer. However, the frequent presence of adherent or overlapping cervical cells in Pap smear images makes separating them individually a difficult task. Currently, there are few studies on the segmentation of adherent cervical cells, and the existing methods commonly suffer from low segmentation accuracy and complex design processes. To address the above problems, we propose a novel star-convex polygon-based convolutional neural network with an encoder-decoder structure, called SPCNet. The model accomplishes the segmentation of adherent cells relying on three steps: automatic feature extraction, star-convex polygon detection, and non-maximal suppression (NMS). Concretely, a new residual-based attentional embedding (RAE) block is suggested for image feature extraction. It fuses the deep features from the attention-based convolutional layers with the shallow features from the original image through the residual connection, enhancing the network's ability to extract the abundant image features. And then, a polygon-based adaptive NMS (PA-NMS) algorithm is adopted to screen the generated polygon proposals and further achieve the accurate detection of adherent cells, thus allowing the network to completely segment the cell instances in Pap smear images. Finally, the effectiveness of our method is evaluated on three independent datasets. Extensive experimental results demonstrate that the method obtains superior segmentation performance compared to other well-established algorithms. [ABSTRACT FROM AUTHOR]