Fan Li

In diagnosing sleep disorders, sleep stage classification is a very essential yet time-consuming process. Most of the existing state-of-the-art approaches rely on hand-crafted features and multi-modality polysomnography (PSG) data, where prior knowledge is compulsory and high computation cost can be expected. Besides, few studies are able to obtain high accuracy sleep staging using raw single-channel electroencephalogram (EEG). To overcome these shortcomings, this paper proposes an end-to-end framework with a deep neural network, namely SingleChannelNet, for automatic sleep stage classification based on raw single-channel EEG. The proposed model utilizes a 90s epoch as the textual input and employs two multi-convolution blocks and several max-average pooling layers to learn different scales of feature representations. To demonstrate the efficiency of the proposed model, we evaluate our model using different raw single-channel EEGs (C4/A1 and Fpz-Cz) on two different datasets (CC-SHS and Sleep-EDF datasets). Experimental results show that the proposed architecture can achieve better over-all accuracy and Cohen’s kappa (CCSHS: 90.2%-86.5%, Sleep-EDF: 86.1%-80.5%) compared with state-of-the-art approaches. Additionally, the proposed model can learn features automatically for sleep stage classification using different single-channel EEGs with distinct sampling rates from different datasets without using any hand-engineered features.

Heart sounds play an important role in the diagnosis of cardiac conditions. Due to the low signal-to-noise ratio (SNR), it is problematic and time-consuming for experts to discriminate different kinds of heart sounds. Thus, objective classification of heart sounds is essential. In this study, we combined a conventional feature engineering method with deep learning algorithms to automatically classify normal and abnormal heart sounds. First, 497 features were extracted from eight domains. Then, we fed these features into the designed convolutional neural network (CNN), in which the fully connected layers that are usually used before the classification layer were replaced with a global average pooling layer to obtain global information about the feature maps and avoid overfitting. Considering the class imbalance, the class weights were set in the loss function during the training process to improve the classification algorithm’s performance. Stratified five-fold cross-validation was used to evaluate the performance of the proposed method. The mean accuracy, sensitivity, specificity and Matthews correlation coefficient observed on the PhysioNet/CinC Challenge 2016 dataset were 86.8%, 87%, 86.6% and 72.1% respectively. The proposed algorithm’s performance achieves an appropriate trade-off between sensitivity and specificity.

Information classification is imperative for machine identification in artificial intelligence. Current information classification approaches only satisfy the mechanization of machines for human services. An essential question is still unresolved, which motivates people to find a way to make machine work for human with thinking. This thesis first discusses information classification in a broad perspective, using a method different from traditional ones which only pay attention to technology and ignore integrity. Then it proposes the three dimension information model of "physical entities - information entities - observers". In this paper, we divide information into "natural information" and "social information". And the process of out-structuring information is divided into "generalized communication" and "generalized transmission", which puts forward new ideas for the generalization and systematization of information research.