Sleep scoring involves the inspection of multimodal recordings of sleep data to detect potential sleep disorders. Given that symptoms of sleep disorders may be correlated with specific sleep stages, the diagnosis is typically supported by the simultaneous identification of a sleep stage and a sleep disorder. This paper investigates the automatic recognition of sleep stages and disorders from multimodal sensory data (EEG, ECG, and EMG). We propose a new distributed multimodal and multilabel decision-making system (MML-DMS). It comprises several interconnected classifier modules, including deep convolutional neural networks (CNNs) and shallow perceptron neural networks (NNs). Each module works with a different data modality and data label. The flow of information between the MML-DMS modules provides the final identification of the sleep stage and sleep disorder. We show that the fused multilabel and multimodal method improves the diagnostic performance compared to single-label and single-modality approaches. We tested the proposed MML-DMS on the PhysioNet CAP Sleep Database, with VGG16 CNN structures, achieving an average classification accuracy of 94.34% and score of 0.92 for sleep stage detection (six stages) and an average classification accuracy of 99.09% and score of 0.99 for sleep disorder detection (eight disorders). A comparison with related studies indicates that the proposed approach significantly improves upon the existing state-of-the-art approaches.
19 July 2022 in IEEE Transactions on Industry Applications
Resonant mode dual active bridge (DAB) bi-directional DC–DC converters using either an inductive-capacitive-inductive (LCL) or a capacitive-inductive-capacitive (CLC) AC-port network offer several performance benefits compared to more conventional non-resonant filter DAB topologies. This includes a reduction in the circulating power between the two bridges, which consequently enables the conduction loss to be minimized and the physical converter volume to be reduced. However, these gains also require that soft-switching is maintained, which is very challenging to achieve across all operating conditions, particularly at light loads. This paper investigates the zero-voltage-switching (ZVS) capability of resonant mode DAB converters, applying Fourier analysis to the converter modulation patterns to quantify the ZVS boundaries analytically. The resulting analytic formulations also quantify the DAB real and reactive power flows, which subsequently enables the identification of modulation conditions that improve the converter's soft-switching and reactive circulating power flow behaviour. The analysis is verified using detailed time-domain PSIM simulations matched to experimental DAB converters equipped with both LCL and CLC filters.
The use of Stereoscopic 3D (S3D) videos has been popular in commercial markets with ongoing developments in the field of visual entertainment in recent years. A wide variety of projection methods of 3D video content is currently available, such as projection to a panoramic screen and projection of omnidirectional video content from head mounted displays using Virtual Reality (VR) technology. This paper investigates the Quality of Experience (QoE) and associated Visually Induced Motion Sickness (VIMS) caused by the viewing of S3D videos. The investigations used three different projection screens: a 3D flat screen, a 3D panoramic screen in a hemispherical shaped room and a VR headset. Several assessment methods including a Simulator Sickness Questionnaire (SSQ), ElectroEncephaloGraphy (EEG), and measurement tools for eye blink rate detection were applied to measure the QoE experienced by viewers. The SSQ scores were also compared with the behavioral data such as attention and meditation levels and enjoyment ratings acquired from different video content and projection screens. The results indicate that the projection screen is a key factor affecting the level of visual fatigue, VIMS and QoE assessments, which are discussed in-depth in the paper.
Existing automatic sleep stage detection methods predominantly use convolutional neural network classifiers (CNNs) trained on features extracted from single-modality signals such as electroencephalograms (EEG). On the other hand, multimodal approaches propose very complexly stacked network structures with multiple CNN branches merged by a fully connected layer. It leads to very high computational and data requirements. This study proposes replacing a stacked network with a distributed neural network system for multimodal sleep stage detection. It has relatively low computational and training data requirements while providing highly competitive results. The proposed multimodal classification and decision-making system (MM-DMS) method applies a fully connected shallow neural network, arbitrating between classification outcomes given by an assembly of independent convolutional neural networks (CNNs), each using a different single-modality signal. Experiments conducted on the CAP Sleep Database data, including the EEG-, ECG-, and EMG modalities representing six stages of sleep, show that the MM-DMS significantly outperforms each single-modality CNN. The fully-connected shallow network arbitration included in the MM-DMS outperforms the traditional majority voting-, average probability-, and maximum probability decision-making methods.
This paper investigates the problem of human pose estimation (HPE) from single 2-dimensional (2D) still images using a convolutional neural network (CNN). The aim was to train the CNN to analyze a 2D input image of a person to determine the person’s pose. The CNN output was given in the form of a tree-structured graph of interconnected nodes representing 2D image coordinates of the person’s body joints. A new data-driven tree-based model for HPE was validated and compared to the traditional anatomy-based tree-based structures. The effect of the number of nodes in anatomy-based tree-based structures on the accuracy of HPE was examined. The tree-based techniques were compared with non-tree-based methods using a common HPE framework and a benchmark dataset. As a result of this investigation, a new hybrid two-stage approach to the HPE estimation was proposed. In the first stage, a non-tree-based network was used to generate approximate results that were then passed for further refinement to the second, tree-based stage. Experimental results showed that both of the proposed methods, the data-driven tree-based model (TD_26) and the hybrid model (H_26_2B) lead to very similar results, obtaining 1% higher HPE accuracy compared to the benchmark anatomy-based model (TA_26) and 3% higher accuracy compared to the non-tree-based benchmark (NT_26_A). The best overall HPE results were obtained using the anatomy-based benchmark with the number of nodes increased from 26 to 50, which also significantly increased the computational cost.