The recent advances in pervasive sensing technologies have enabled us to monitor and analyze the multi-channel electroencephalogram (EEG) signals of epilepsy patients to prevent serious outcomes caused by epileptic seizures. To avoid manual visual inspection from long-term EEG readings, automatic EEG seizure detection has garnered increasing attention among researchers. In this paper, we present a unified multi-view deep learning framework to capture brain abnormalities associated with seizures based on multi-channel scalp EEG signals. The proposed approach is an end-to-end model that is able to jointly learn multi-view features from both unsupervised multi-channel EEG reconstruction and supervised seizure detection via spectrogram representation. We construct a new autoencoder-based multi-view learning model by incorporating both inter and intra correlations of EEG channels to unleash the power of multi-channel information. By adding a channel-wise competition mechanism in the training phase, we propose a channel-aware seizure detection module to guide our multi-view structure to focus on important and relevant EEG channels. To validate the effectiveness of the proposed framework, extensive experiments against nine baselines, including both traditional handcrafted feature extraction and conventional deep learning methods, are carried out on a benchmark scalp EEG dataset. Experimental results show that the proposed model is able to achieve higher average accuracy and f1-score at 94.37% and 85.34%, respectively, using 5-fold subject-independent cross validation, demonstrating a powerful and effective method in the task of EEG seizure detection.

Predicting patients' risk of developing certain diseases is an important research topic in healthcare. Accurately identifying and ranking the similarity among patients based on their historical records is a key step in personalized healthcare. The electric health records (EHRs), which are irregularly sampled and have varied patient visit lengths, cannot be directly used to measure patient similarity due to the lack of an appropriate representation. Moreover, there needs an effective approach to measure patient similarity on EHRs. In this paper, we propose two novel deep similarity learning frameworks which simultaneously learn patient representations and measure pairwise similarity. We use a convolutional neural network (CNN) to capture local important information in EHRs and then feed the learned representation into triplet loss or softmax cross entropy loss. After training, we can obtain pairwise distances and similarity scores. Utilizing the similarity information, we then perform disease predictions and patient clustering. Experimental results show that CNN can better represent the longitudinal EHR sequences, and our proposed frameworks outperform state-of-the-art distance metric learning methods.

Background: Sleep is a complex and dynamic biological process characterized by different sleep patterns. Comprehensive sleep monitoring and analysis using multivariate polysomnography (PSG) records has achieved significant efforts to prevent sleep-related disorders. To alleviate the time consumption caused by manual visual inspection of PSG, automatic multivariate sleep stage classification has become an important research topic in medical and bioinformatics. Results: We present a unified hybrid self-attention deep learning framework, namely HybridAtt, to automatically classify sleep stages by capturing channel and temporal correlations from multivariate PSG records. We construct a new multi-view convolutional representation module to learn channel-specific and global view features from the heterogeneous PSG inputs. The hybrid attention mechanism is designed to further fuse the multi-view features by inferring their dependencies without any additional supervision. The learned attentional representation is subsequently fed through a softmax layer to train an end-to-end deep learning model. Conclusions: We empirically evaluate our proposed HybridAtt model on a benchmark PSG dataset in two feature domains, referred to as the time and frequency domains. Experimental results show that HybridAtt consistently outperforms ten baseline methods in both feature spaces, demonstrating the effectiveness of HybridAtt in the task of sleep stage classification.

Recently, pervasive sensing technologies have been widely applied to comprehensive patient monitoring in order to improve clinical treatment. Various types of biomedical signals collected by different sensing channels provide different aspects of patient health information. However, due to the uncertainty and variability in clinical observation, not all the channels are relevant and important to the target task. Thus, in order to extract informative representations from multi-channel biosignals, channel awareness has become a key enabler for deep learning in biosignal processing and has attracted increasing research interest in health informatics. Towards this end, we propose FusionAtt—a deep fusional attention network that can learn channel-aware representations of multi-channel biosignals, while preserving complex correlations among all the channels. FusionAtt is able to dynamically quantify the importance of each biomedical channel, and relies on more informative ones to enhance feature representation in an end-to-end manner. We empirically evaluated FusionAtt in two clinical tasks: multi-channel seizure detection and multivariate sleep stage classification. Experimental results showed that FusionAtt consistently outperformed the state-of-the-art models in four different evaluation measurements, demonstrating the effectiveness of the proposed fusional attention mechanism.

Epilepsy is a neurological disease characterized by unprovoked seizures in the brain. The recent advances in sensor technologies allow researchers to analyze the collected biological records to improve the treatment of epilepsy. Electroencephalogram (EEG) is the most commonly used biological measurement to effectively capture the abnormalities of different brain areas during the EEG seizures. To avoid manual visual inspection from long-term EEG readings, automatic epileptic EEG seizure detection has become an important research issue in bioinformatics. We present a multi-context learning approach to automatically detect EEG seizures by incorporating a feature fusion strategy. We generate EEG scalogram sequences from the EEG records by utilizing waveform transform to describe the frequency content over time. We propose a multi-stage unsupervised model that integrates the features extracted from the global handcrafted engineering, channel-wise deep learning, and EEG embeddings, respectively. The learned multi-context features are subsequently merged to train a seizure detector. To validate the effectiveness of the proposed approach, extensive experiments against several baseline methods are carried out on two benchmark biological datasets. The experimental results demonstrate that the representative context features from multiple perspectives can be learned by the proposed model, and further improve the performance for the task of EEG seizure detection.

With the recent advances in wireless technologies, high frequency radio has become the primary medium for long-distance communication. Among various types of modulation in signal transmission, morse code stands out due to its simplicity and efficiency that can be transmitted in a small bandwidth while carrying rich information. In practice, however, it is extremely laborious for receivers to locate morse signals in a blind scenario without prior information on transmission frequency. To avoid manual retrieval from such underdetermined radio data, automatic blind morse signal detection has become a challenging task in wireless morse communications, and new solutions could be derived from the latest machine learning techniques. Towards this end, we propose a deep learning framework, namely DeepMorse, to blindly detect morse signals in wideband spectrum data. In particular, we first develop a multi-signal sensing module to retrieve signal candidates from wideband spectrum without prior knowledge. Then, we construct a CNN-based module to extract informative features from the located candidates, in order to distinguish morse signal from other types of modulation. To evaluate the proposed DeepMorse model, we set up a testbed using commercialized long-distance wireless communication devices. Experimental results show that DeepMorse can effectively detect morse signals and outperform the state-of-the-art methods on four real-world datasets.