Glaucoma is one of the leading causes of irreversible vision loss. Many approaches have recently been proposed for automatic glaucoma detection based on fundus images. However, none of the existing approaches can efficiently remove high redundancy in fundus images for glaucoma detection, which may reduce the reliability and accuracy of glaucoma detection. To avoid this disadvantage, this paper proposes an attention-based convolutional neural network (CNN) for glaucoma detection, called AG-CNN. Specifically, we first establish a large-scale attention-based glaucoma (LAG) database, which includes 11,760 fundus images labeled as either positive glaucoma (4,878) or negative glaucoma (6,882). Among the 11,760 fundus images, the attention maps of 5,824 images are further obtained from ophthalmologists through a simulated eye-tracking experiment. Then, a new structure of AG-CNN is designed, including an attention prediction subnet, a pathological area localization subnet and a glaucoma classification subnet. The attention maps are predicted in the attention prediction subnet to highlight the salient regions for glaucoma detection, under a weakly supervised training manner. In contrast to other attention-based CNN methods, the features are also visualized as the localized pathological area, which are further added in our AG-CNN structure to enhance the glaucoma detection performance. Finally, the experiment results from testing over our LAG database and another public glaucoma database show that the proposed AG-CNN approach significantly advances the state-of-the-art in glaucoma detection.

The intelligent recognition of epileptic electroencephalogram (EEG) signals is a valuable tool for the epileptic seizure detection. Recent deep learning models fail to fully consider both spectral and temporal domain representations simultaneously, which may lead to omitting the nonstationary or nonlinear property in epileptic EEGs and further produce a suboptimal recognition performance consequently. In this paper, an end-to-end EEG seizure detection framework is proposed by using a novel channel-embedding spectral-temporal squeeze-andexcitation network (CE-stSENet) with a maximum mean discrepancy-based information maximizing loss. Specifically, the CE-stSENet firstly integrates both multi-level spectral and multiscale temporal analysis simultaneously. Hierarchical multidomain representations are then captured in a unified manner with a variant of squeeze-and-excitation block. The classification net is finally implemented for epileptic EEG recognition based on features extracted in previous subnetworks. Particularly, to address the fact that the scarcity of seizure events results in finite data distribution and the severe overfitting problem in seizure detection, the CE-stSENet is coordinated with a maximum mean discrepancy-based information maximizing loss for mitigating the overfitting problem. Competitive experimental results on three EEG datasets against the state-of-the-art methods demonstrate the effectiveness of the proposed framework in recognizing epileptic EEGs, indicating its powerful capability in the automatic seizure detection.

Motor imagery electroencephalography (EEG) decoding is an essential part in brain-computer interfaces (BCIs) which help motor-disabled patients to communicate with the outside world by external devices. Recently, deep learning algorithms using decomposed spectrums of EEG as inputs may omitted important spatial dependencies and different temporal scale information, thus generated the poor decoding performance. In this paper, we propose an end-to-end EEG decoding framework, which employs raw multi-channel EEG as inputs, to boost decoding accuracy by the channel-projection mixed-scale convolutional neural network (CP-MixedNet) aided by amplitudeperturbation data augmentation. Specifically, the first block in CP-MixedNet is designed to learn primary spatial and temporal representations from EEG signals. The mixed-scale convolutional block is then used to capture mixed-scale temporal information, which effectively reduces the number of training parameters when expanding reception fields of the network. Finally, based on the features extracted in previous blocks, the classification block is constructed to classify EEG tasks. Experiments are implemented on two public EEG datasets (BCI competition IV 2a and High gamma dataset) to validate the effectiveness of the proposed approach comparing to the state-of-the-art methods. The competitive results demonstrate that our proposed method is a promising solution to improve the decoding performance of motor imagery BCIs.

Epileptic seizure detection is commonly implemented by expert clinicians with visual observation of electroencephalography (EEG) signals, which tends to be time consuming and sensitive to bias. The epileptic detection in most previous research suffers from low power and unsuitability for processing large datasets. Therefore, a computerized epileptic seizure detection method is highly required to eradicate the aforementioned problems, expedite epilepsy research and aid medical professionals. In this work, we propose an automatic epilepsy diagnosis framework based on the combination of multi-domain feature extraction and nonlinear analysis of EEG signals. Firstly, EEG signals are pre-processed by using the wavelet threshold method to remove the artifacts. We then extract representative features in the time domain, frequency domain, time-frequency domain and nonlinear analysis features based on the information theory. These features are further extracted in five frequency sub-bands based on the clinical interest, and the dimension of the original feature space is then reduced by using both a principal component analysis and an analysis of variance. Furthermore, the optimal combination of the extracted features is identified and evaluated via different classifiers for the epileptic seizure detection of EEG signals. Finally, the performance of the proposed method is investigated by using a public EEG database at the University Hospital Bonn, Germany. Experimental results demonstrate that the proposed epileptic seizure detection method can achieve a high average accuracy of 99.25%, indicating a powerful method in the detection and classification of epileptic seizures. The proposed seizure detection scheme is thus hoped to eliminate the burden of expert clinicians when they are processing a large number of data by visual observation and to speed-up the epilepsy diagnosis.

Graph theory analysis using electroencephalogram (EEG) signals is currently an advanced technique for seizure prediction. Recent deep learning approaches, which fail to fully explore both the characterizations in EEGs themselves and correlations among different electrodes simultaneously, generally neglect the spatial or temporal dependencies in an epileptic brain and, thus, produce suboptimal seizure prediction performance consequently. To tackle this issue, in this article, a patient-specific EEG seizure predictor is proposed by using a novel spatio-temporal-spectral hierarchical graph convolutional network with an active preictal interval learning scheme (STS-HGCN-AL). Specifically, since the epileptic activities in different brain regions may be of different frequencies, the proposed STS-HGCN-AL framework first infers a hierarchical graph to concurrently characterize an epileptic cortex under different rhythms, whose temporal dependencies and spatial couplings are extracted by a spectral-temporal convolutional neural network and a variant self-gating mechanism, respectively. Critical intrarhythm spatiotemporal properties are then captured and integrated jointly and further mapped to the final recognition results by using a hierarchical graph convolutional network. Particularly, since the preictal transition may be diverse from seconds to hours prior to a seizure onset among different patients, our STS-HGCN-AL scheme estimates an optimal preictal interval patient dependently via a semisupervised active learning strategy, which further enhances the robustness of the proposed patient-specific EEG seizure predictor. Competitive experimental results validate the efficacy of the proposed method in extracting critical preictal biomarkers, indicating its promising abilities in automatic seizure prediction.

Retinal vessel segmentation with deep learning technology is a crucial auxiliary method for clinicians to diagnose fundus diseases. However, the deep learning approaches inevitably lose the edge information, which contains spatial features of vessels while performing down-sampling, leading to the limited segmentation performance of fine blood vessels. Furthermore, the existing methods ignore the dynamic topological correlations among feature maps in the deep learning framework, resulting in the inefficient capture of the channel characterization. To address these limitations, we propose a novel dual encoder-based dynamic-channel graph convolutional network with edge enhancement (DE-DCGCN-EE) for retinal vessel segmentation. Specifically, we first design an edge detection-based dual encoder to preserve the edge of vessels in down-sampling. Secondly, we investigate a dynamic-channel graph convolutional network to map the image channels to the topological space and synthesize the features of each channel on the topological map, which solves the limitation of insufficient channel information utilization. Finally, we study an edge enhancement block, aiming to fuse the edge and spatial features in the dual encoder, which is beneficial to improve the accuracy of fine blood vessel segmentation. Competitive experimental results on five retinal image datasets validate the efficacy of the proposed DE-DCGCN-EE, which achieves more remarkable segmentation results against the other state-of-the-art methods, indicating its potential clinical application.