Search

Brain decoding is to predict the external stimulus information from the collected brain response activities, and visual information is one of the most important sources of external stimulus information. Decoding functional magnetic resonance imaging (fMRI) based on visual stimulation is helpful in understanding the working mechanism of the brain visual function regions. Traditional brain decoding algorithms cannot accurately extract stimuli features from fMRI. To address these shortcomings, this paper proposed a brain decoding algorithm based on a graph convolution network (GCN). Firstly, 11 regions of interest (ROI) were selected according to the human brain visual function regions, which can avoid the noise interference of the non-visual regions of the human brain; then, a deep three-dimensional convolution neural network was specially designed to extract the features of these 11 regions; next, the GCN was used to extract the functional correlation features between the different human brain visual regions. Furthermore, to avoid the problem of gradient disappearance when there were too many layers of graph convolutional neural network, the residual connections were adopted in our algorithm, which helped to integrate different levels of features in order to improve the accuracy of the proposed GCN. The proposed algorithm was tested on the public dataset, and the recognition accuracy reached 98.67%. Compared with the other state-of-the-art algorithms, the proposed algorithm performed the best. [ABSTRACT FROM AUTHOR]

Sign language recognition and translation can address the communication problem between hearing-impaired and general population, and can break the sign language boundariesy between different countries and different languages. Traditional sign language recognition and translation algorithms use Convolutional Neural Networks (CNNs) to extract spatial features and Recurrent Neural Networks (RNNs) to extract temporal features. However, these methods cannot model the complex spatiotemporal features of sign language. Moreover, RNN and its variant algorithms find it difficult to learn long-term dependencies. This paper proposes a novel and effective network based on Transformer and Graph Convolutional Network (GCN), which can be divided into three parts: a multi-view spatiotemporal embedding network (MSTEN), a continuous sign language recognition network (CSLRN), and a sign language translation network (SLTN). MSTEN can extract the spatiotemporal features of RGB data and skeleton data. CSLRN can recognize sign language glosses and obtain intermediate features from multi-view input sign data. SLTN can translate intermediate features into spoken sentences. The entire network was designed as end-to-end. Our method was tested on three public sign language datasets (SLR-100, RWTH, and CSL-daily) and the results demonstrated that our method achieved excellent performance on these datasets. [ABSTRACT FROM AUTHOR]

Wearable devices are increasingly popular for health monitoring via electrocardiograms (ECGs) as they can portably monitor heart conditions over a long time. However, so far there are no publicly available ECG data sets collected from wearable devices. Most ECG analysis algorithms target ECG data collected by hospital equipment. In the present study, we used the IREALCARE2.0 Flexible Cardiac Monitor Patch as the wearable device to collect ECG signals and formed ECG data sets. Wearable ECG data tended to contain more interference and be large in size. This article proposed a deep CNN approach, named time–spatial convolutional neural networks (TSCNNs), for the automatic classification and analysis of ECG signals from wearable devices. First, the original long-term ECG signals were divided into separate heartbeats and input into the TSCNN. Second, we applied convolution over time and spatial filtering for each heartbeat to extract abundant features. Finally, the cascaded small-scale kernel convolution was applied to improve classification performance and reduce the number of network parameters. To avoid overfitting, some regularized methods such as dropout and batch normalization were adopted. In the experiments, the method proposed in this letter is compared with other eight ECG classification algorithms. Our method attained the highest classification accuracy. The experimental results indicated that the proposed method can achieve better performance for wearable ECG data and can effectively monitor whether the wearer has an abnormal ECG. [ABSTRACT FROM AUTHOR]

Electrical status epilepticus during sleep (ESES) is an epileptic syndrome in which neurons in the brain continue to discharge during the sleep phase and is common in mid-childhood. Affected patients often experience a decline in cognition, learning ability, memory, and expressive language skills. Therefore, timely and accurate diagnosis can effectively protect the health of a patient. Currently, the identification and analysis of ESES activities mainly rely on manual detection or traditional matching learning algorithms, such as morphology and template matching. These algorithms are time-consuming or have low accuracy. Therefore, in this paper, we propose a graph convolutional neural network that can automatically and accurately identify ESES activity from non-ESES activity. We divide the whole EEG signal into small segments, each of which covers one second of the EEG data. Then, we construct a graph according to each segment of the EEG data and train a graph convolutional neural network to classify the graph into two categories: ESES or non-ESES. Compared with other state-of-the-art algorithms, for the proposed algorithm, the accuracy, F1-Score, Area Under Curve(AUC) and sensitivity reaches 91.2%, 95.0%, 96.5%, and 91.3%, respectively, and outperforms the other algorithms. [ABSTRACT FROM AUTHOR]

The liver is an essential metabolic organ of the human body, and malignant liver tumors seriously affect and threaten human life. The segmentation algorithm for liver and liver tumors is one of the essential branches of computer-aided diagnosis. This paper proposed a two-stage liver and tumor segmentation algorithm based on the convolutional neural network (CNN). In the present study, we used two stages to segment the liver and tumors: liver localization and tumor segmentation. In the liver localization stage, the network segments the liver region, adopts the encoding–decoding structure and long-distance feature fusion operation, and utilizes the shallow features' spatial information to improve liver identification. In the tumor segmentation stage, based on the liver segmentation results of the first two steps, a CNN model was designed to accurately identify the liver tumors by using the 2D image features and 3D spatial features of the CT image slices. At the same time, we use the attention mechanism to improve the segmentation performance of small liver tumors. The proposed algorithm was tested on the public data set Liver Tumor Segmentation Challenge (LiTS). The Dice coefficient of liver segmentation was 0.967, and the Dice coefficient of tumor segmentation was 0.725. The proposed algorithm can accurately segment the liver and liver tumors in CT images. Compared with other state-of-the-art algorithms, the segmentation results of the proposed algorithm rank the highest in the Dice coefficient. [ABSTRACT FROM AUTHOR]