Search

Drowsiness detection is a test designed to detect a person's reaction ability and speed while in a state of fatigue. This type of drowsiness detection has many practical applications in areas like driver monitoring, transportation industry, and workplace safety. In this paper, a new machine learning drowsiness detection algorithm based on hybrid brain network feature extraction is proposed. First, the raw electroencephalogram signals are segmented into 30 s epochs, which are decomposed with wavelet packet transform into 7 layers to extract 30 s sleep characteristic subbands and the 7-th layer subbands. After that, 30 s subbands are divided into 5 s subbands with 50% overlap. At the same time, 30 s subbands are reconstructed by phase space reconstruction (PSR), whose parameters are optimized through energy valley optimizer. Then, based on mutual information and horizontal visual graph, a hybrid brain network is constructed with the 7-th layer subbands, 5 s subbands and PSR subbands, respectively. Finally, network features and time–frequency features are extracted and input into random forest classifier. The performance of the algorithm is evaluated through two tests: subject-independent test and subject-non-independent test. The results show that in Sleep-EDF dataset, accuracy, κ coefficient, recall, precision and F1-score obtained by the proposed algorithm can reach 94.19%, 88.39%, 95.37%, 92.53% and 93.93%, respectively. In Sleep-EDF Expanded dataset, accuracy, κ coefficient, recall, precision and F1-score can reach 93.65%, 87.65%, 94.53%, 92.11%, and 92.89%, respectively. [ABSTRACT FROM AUTHOR]

CNNs with strong learning abilities are widely chosen to resolve super-resolution problem. However, CNNs depend on deeper network architectures to improve performance of image super-resolution, which may increase computational cost in general. In this paper, we present an enhanced super-resolution group CNN (ESRGCNN) with a shallow architecture by fully fusing deep and wide channel features to extract more accurate low-frequency information in terms of correlations of different channels in single image super-resolution (SISR). Also, a signal enhancement operation in the ESRGCNN is useful to inherit more long-distance contextual information for resolving long-term dependency. An adaptive up-sampling operation is gathered into a CNN to obtain an image super-resolution model with low-resolution images of different sizes. Extensive experiments report that our ESRGCNN surpasses the state-of-the-arts in terms of SISR performance, complexity, execution speed, image quality evaluation and visual effect in SISR. Code is found at https://github.com/hellloxiaotian/ESRGCNN. [ABSTRACT FROM AUTHOR]

• A fast finite-time criterion for deterministic systems is established. • We establish a fast finite-time control criterion for stochastic systems by revamping the fast finite time method. • The control strategy can effectively improve the control speed compared to the existing results in this paper. This paper elucidates the finite-time stability in probability problem for a class of p-norm stochastic nonlinear systems. Among them, from the point of view of convergence speed, finite-time control does have optimal properties. Consequently, a fast finite-time criterion for deterministic systems is established. However, although there exist a significant amount of results on fast finite time stability, there is no article involves p-norm stochastic nonlinear systems case available so far. There are random disturbance terms that made the traditional fast finite-time control methods inapplicable. To overcome this problem, we establish a fast finite-time control criterion for stochastic systems by revamping the fast finite time method of deterministic systems. Then, with the aid of sign function and adding a power integrator technology, a non-Lipschitz continuous state feedback controller is established successfully. Next, the control strategy can effectively improve the control speed compared to the existing results in this paper. Finally, numerical and dynamic suspension simulations are offered to augment our theoretical analysis results, and comparison plots with conventional control strategies are shown. [ABSTRACT FROM AUTHOR]

Domain generalization-based fault diagnosis (DGFD) presents significant prospects for recognizing faults without the accessibility of the target domain. Previous DGFD methods have achieved significant progress; however, there are some limitations. First, most DGFG methods statistically model the dependence between time-series data and labels, and they are superficial descriptions to the actual data-generating process. Second, most of the existing DGFD methods are only verified on vibrational time-series datasets, which is insufficient to show the potential of domain generalization in the fault diagnosis area. In response to the above issues, this paper first proposes a DGFD method named Causal Disentanglement Domain Generalization (CDDG), which can reestablish the data-generating process by disentangling time-series data into the causal factors (fault-related representation) and no-casual factors (domain-related representation) with a structural causal model. Specifically, in CDDG, causal aggregation loss is designed to separate the unobservable causal and non-causal factors. Meanwhile, the reconstruction loss is proposed to ensure the information completeness of the disentangled factors. We also introduce a redundancy reduction loss to learn efficient features. The proposed CDDG is verified on five cross-machine vibrational fault diagnosis cases and three cross-environment acoustical anomaly detection cases by comparing it with eight state-of-the-art (SOTA) DGFD methods. We publicize the open-source time-series DGFD Benchmark containing CDDG and the eight SOTA methods. The code repository will be available at https://github.com/ShaneSpace/DGFDBenchmark. [ABSTRACT FROM AUTHOR]

With the development of human society, environmental problems caused by the use of traditional fossil energy sources are increasing. Therefore, environment-friendly, renewable new energy technologies need to be urgently developed. The conversion of solar energy into other forms of clean energy sources, such as hydrogen gas, can alleviate environmental and energy problems to some extent. Due to their special physical and chemical characteristics, semiconductor materials can convert solar energy to other forms of energy sources, enabling the cyclic utilization of energy. As an important semiconductor material, InGaN can be used in photoelectrochemical (PEC) water splitting to produce hydrogen because of its tunable bandgap and high electron mobility. This paper reviews the strategy of PEC water splitting into hydrogen by InGaN nanorods, summarizes the methods to enhance PEC water splitting efficiency, and provides an outlook on how to enhance the efficiency and stability of this process. • The strategy of PEC water splitting into hydrogen by InGaN nanorods is reviewed. • The methods to enhance PEC water splitting efficiency are summarized. • An outlook on enhancing the efficiency and stability of InGaN nanorods are provided. [ABSTRACT FROM AUTHOR]

Fault diagnosis of the bearing is vital for the safe and reliable operation of rotating machines in the manufacturing industry. Convolutional neural networks (CNNs) have been popular in bearing fault diagnosis by right of robust and reliable feature extraction ability. However, the collected vibrational signals from machines are usually corrupted by unrelated noises due to complicated transfer path modulations and component coupling. As traditional CNN lacks the denoising structure, its capability of extracting features from vibrational features is restrained by noise disturbances. In response to the above issue, this paper first proposes a simple but efficient Gramian-based noise reduction strategy called Gramian Noise Reduction (GNR) based on the periodic self-similarity of vibrational signals. Second, for the problem of lacking denoising structure in traditional CNNs, a novel end-to-end GNR-based CNN model, termed as Gramian Time Frequency Enhancement Network (GTFE-Net), is presented for bearing fault diagnosis. The GTFE-Net has three branches to parallelly process the raw original signal, the GNR denoised signal, and the frequency spectrums, respectively. GNR is integrated into the GTFE-Net, prompting the network to pay more attention to feature extraction rather than noise suppression. Three case studies using test rig and real engineering datasets are performed to verify the effectiveness of the proposed method for bearing fault diagnosis. The experimental results show that the GTFE-Net can reduce the useless noises in vibrational signals and deliver a remarkable improvement in classification performance compared with the six state-of-the-art methods. The source code is available at [ABSTRACT FROM AUTHOR]

The scarcity of data on rare diseases poses a significant challenge to the development of diagnostic systems. While few-shot learning (FSL) offers promise in low-data regimes, it often struggles in open-set scenarios, failing to identify unknown diseases. In this paper, we introduce a Dynamic Attribute-guided Few-shot Open-set Network (DAFON), representing the first effort to simultaneously address closed-set classification and open-set recognition in rare disease diagnosis. To alleviate incomprehensive category knowledge stemming from data scarcity, we propose a Global Attribute Generator to create attributes and produce image attribute activations for closed-set data as auxiliary information. An attribute space is then constructed and employed to generate the pseudo open-set attribute activations using a designed Open-set Data Sampler. By incorporating closed-set and open-set attribute activations as conditions, we propose a Dynamic Attribute Guided Alignment module to align feature space with attribute space, so as to derive feature space with intra-class compactness for closed-set and open-set classes. Our DAFON achieves state-of-the-art performance on two public medical image datasets, demonstrating its effectiveness for FSOSR in medical image diagnosis. • Attributes can be generated for specific dataset even without auxiliary data provided. • Attribute information enhances category knowledge estimation in low-data regimes. • The idea of attribution composition helps generate desired open-set data. • Integrating open-set data enhances intra-class feature space compactness. • Task-specific knowledge is beneficial to classification tasks with scarce data. [ABSTRACT FROM AUTHOR]

West syndrome is an epileptic disease that seriously affects the normal growth and development of infants in early childhood. Based on the methods of brain topological network and graph theory, this article focuses on three clinical states of patients before and after treatment. In addition to discussing bidirectional and unidirectional global networks from the perspective of computational principles, a more in-depth analysis of local intra-network and inter-network characteristics of multi-partitioned networks is also performed. The spatial feature distribution based on feature path length is introduced for the first time. The results show that the bidirectional network has better significant differentiation. The rhythmic feature change trend and spatial characteristic distribution of this network can be used as a measure of the impact on global information processing in the brain after treatment. And localized brain regions variability in features and differences in the ability to interact with information between brain regions have potential as biomarkers for medication assessment in WEST syndrome. The above shows specific conclusions on the interaction relationship and consistency of macro-network and micro-network, which may have a positive effect on patients' treatment and prognosis management. • Evaluate the brain network changes in three clinical states of patients before and after treatment. • Construct multi-partition local network for intra-network and inter-network feature analysis. • The bidirectional network has more significant differences than the unidirectional network. • Localized brain regions variability is effective for medication assessment in WEST syndrome. [ABSTRACT FROM AUTHOR]

• We propose a SFDA framework with Fourier style mining for medical image segmentation, composing of a generation stage and an adaptation stage. To the best of our knowledge, this represents the first effort to solve unsupervised domain adaptation with only the source model on medical image segmentation. • In the generation stage, we propose a Fourier Style Mining (FSM) generator to produce high-quality source-like images from the well-trained source segmentation model through BN constraints and mutual Fourier Transform. FSM represents the first attempt to generate source-like images from a segmentation model to deal with domain shift in SFDA. • In the adaptation stage, CDD and CADC modules are proposed to achieve featurelevel and output-level alignment based on paired source-like and target images. The CADC introduces shape compactness metric to filter out noisy pseudo labels for medical image segmentation. • Comprehensive experiments on different domain scenarios (including modalities, vendors and clinical cites), demonstrate the effectiveness of our style mining guided SFDA framework, which outperforms state -of-the-art domain adaptation methods. [Display omitted] Unsupervised domain adaptation (UDA) aims to exploit the knowledge learned from a labeled source dataset to solve similar tasks in a new unlabeled target domain. Existing UDA techniques typically assume that samples from source and target domains are freely accessible during the training. However, it may be impractical to access source images due to privacy concerns, especially in medical imaging scenarios with the patient information. To tackle this issue, we devise a novel source free domain adaptation framework with fourier style mining, where only a well-trained source segmentation model is available for the adaptation to the target domain. Our framework is composed of two stages: a generation stage and an adaptation stage. In the generation stage, we design a Fourier Style Mining (FSM) generator to inverse source-like images through statistic information of the pretrained source model and mutual Fourier Transform. These generated source-like images can provide source data distribution and benefit the domain alignment. In the adaptation stage, we design a Contrastive Domain Distillation (CDD) module to achieve feature-level adaptation, including a domain distillation loss to transfer relation knowledge and a domain contrastive loss to narrow down the domain gap by a self-supervised paradigm. Besides, a Compact-Aware Domain Consistency (CADC) module is proposed to enhance consistency learning by filtering out noisy pseudo labels with shape compactness metric, thus achieving output-level adaptation. Extensive experiments on cross-device and cross-centre datasets are conducted for polyp and prostate segmentation, and our method delivers impressive performance compared with state-of-the-art domain adaptation methods. The source code is available at https://github.com/CityU-AIM-Group/SFDA-FSM. [ABSTRACT FROM AUTHOR]

• We present an Instance Importance-aware Graph Convolutional Network (I 2 GCN) to achieve 3D medical diagnosis with merely patient-level supervision, which enables the multi-instance learning (MIL) framework to exploit the relationship among instances comprehensively. • We evaluate the diagnostic importance of each instance by revisiting the channelwise contribution of embeddings, which can be utilized to perform the refined diagnosis. To the best of our knowledge, this work represents the first attempt to estimate the instance importance from the supervised knowledge in MIL. • In the refined diagnosis branch, we propose the Instance Importance-aware Graph Convolutional Layer (I2GCLayer) to explore the complementary information with the importance-based and feature-based topologies under the orthogonal constraint. Besides, the importance-based Sub-Graph Augmentation (SGA) is devised to enhance the graph-based learning compatible with MIL. • We perform extensive experiments of 3D medical diagnosis on publicly available lung CT and prostate MRI datasets to validate the effectiveness of the proposed I 2 GCN and instance importance calculation thoroughly. The proposed I 2 GCN outperforms state-of-the-art methods by a large margin. [Display omitted] Automatic diagnosis of 3D medical data is a significant goal of intelligent healthcare. By exploiting the abundant pathological information of 3D data, human experts and algorithms can provide accurate predictions for patients. Considering the high cost of collecting exhaustive annotations for 3D data, a sustainable alternative is to develop diagnosis algorithms with merely patient-level labels. Motivated by the fact that 2D slices of 3D data hold explicit diagnostic efficacy, we propose the Instance Importance-aware Graph Convolutional Network (I 2 GCN) under the multi-instance learning (MIL). Specifically, we first calculate the instance importance of each slice towards diagnosis using a preliminary MIL classifier, which is further utilized to promote the refined diagnosis branch. In the refined diagnosis branch, we devise the Instance Importance-aware Graph Convolutional Layer (I 2 GCLayer) to exploit complementary features in both importance-based and feature-based topologies. Moreover, to alleviate the deficient supervision of 3D dataset, we propose the importance-based Sub-Graph Augmentation (SGA) to effectively regularize the framework training. Extensive experiments confirm the effectiveness of our method with different organs and modals on the CC-CCII and PROSTATEx datasets, which outperforms state-of-the-art methods by a large margin. The source code is available at https://github.com/CityU-AIM-Group/I2GCN. [ABSTRACT FROM AUTHOR]

• A novel MLMix data augmentation method enlarges the limited annotated training dataset, in order to promote the generalization ability of the optimized segmentation model. • MLMix utilizes meta-learning strategy to obtain compatible augmented image- label pairs in a data-driven manner. • A CAR strategy adopts the easy-to-hard gradual learning scheme in both image and pixel levels to enable the robust training of polyp segmentation model. • The class prior knowledge is integrated in the training procedure to rebalance the data distribution, so as to mitigate the class imbalance problem. • Comprehensive experiments demonstrate the effectiveness and generality of MLMix and CAR strategy. [Display omitted] Automatic segmentation of polyp regions in endoscope images is essential for the early diagnosis and surgical planning of colorectal cancer. Recently, deep learning-based approaches have achieved remarkable progress for polyp segmentation, but they are at the expense of laborious large-scale pixel-wise annotations. In addition, these models treat samples equally, which may cause unstable training due to polyp variability. To address these issues, we propose a novel Meta-Learning Mixup (MLMix) data augmentation method and a Confidence-Aware Resampling (CAR) strategy for polyp segmentation. MLMix adaptively learns the interpolation policy for mixup data in a data-driven way, thereby transferring the original soft mixup label to a reliable hard label and enriching the limited training dataset. Considering the difficulty of polyp image variability in segmentation, the CAR strategy is proposed to progressively select relatively confident images and pixels to facilitate the representation ability of model and ensure the stability of the training procedure. Moreover, the CAR strategy leverages class distribution prior knowledge and assigns different penalty coefficients for polyp and normal classes to rebalance the selected data distribution. The effectiveness of the proposed MLMix data augmentation method and CAR strategy is demonstrated through comprehensive experiments, and our proposed model achieves state-of-the-art performance with 87.450% dice on the EndoScene test set and 86.453% dice on the wireless capsule endoscopy (WCE) polyp dataset. [ABSTRACT FROM AUTHOR]

• A consolidated domain adaptive framework to detect and lo- cate polyps from colonoscopic images collected from different devices or hospitals. • A novel training-free image style transfer method to unify the style of colonoscopic images without distorting their contents. • A hierarchical unsupervised domain adaptive detection net- work utilizes adversarial training to align colonoscopic image features. • A Centre Besiegement loss for better optimization of the pre- dicted polyp locations. • Comprehensive studies demonstrate the superiority of the pro- posed consolidated domain adaptive framework in automatic polyp detection and localization. [Display omitted] Automatic polyp detection has been proven to be crucial in improving the diagnosis accuracy and reducing colorectal cancer mortality during the precancerous stage. However, the performance of deep neural networks may degrade severely when being deployed to polyp data in a distinct domain. This domain distinction can be caused by different scanners, hospitals, or imaging protocols. In this paper, we propose a consolidated domain adaptive detection and localization framework to bridge the domain gap between different colonosopic datasets effectively, consisting of two parts: the pixel-level adaptation and the hierarchical feature-level adaptation. For the pixel-level adaptation part, we propose a Gaussian Fourier Domain Adaptation (GFDA) method to sample the matched source and target image pairs from Gaussian distributions then unify their styles via the low-level spectrum replacement, which can reduce the domain discrepancy of the cross-device polyp datasets in appearance level without distorting their contents. The hierarchical feature-level adaptation part comprising a Hierarchical Attentive Adaptation (HAA) module to minimize the domain discrepancy in high-level semantics and an Iconic Concentrative Adaptation (ICA) module to perform reliable instance alignment. These two modules are regularized by a Generalized Consistency Regularizer (GCR) for maintaining the consistency of their domain predictions. We further extend our framework to the polyp localization task and present a Centre Besiegement (CB) loss for better location optimization. Experimental results show that our framework outperforms other domain adaptation detectors by a large margin in the detection task meanwhile achieves the state-of-the-art recall rate of 87.5% in the localization task. The source code is available at https://github.com/CityU-AIM-Group/ConsolidatedPolypDA. [ABSTRACT FROM AUTHOR]