A Spatial-Spectral Relation-Guided Fusion Network for Multisource Optical RS Image Classification.

Multisource optical remote sensing (RS) image classification has obtained extensive research interest with demonstrated superiority. Existing approaches mainly improve classification performance by exploiting complementary information from multisource data. However, these approaches are insufficient in effectively extracting data features and utilizing correlations of multisource optical RS images. For this purpose, this article proposes a generalized spatial-spectral relation-guided fusion network ( S2 RGF-Net) for multisource optical RS image classification. First, we elaborate on spatial-and spectral-domain-specific feature encoders based on data characteristics to explore the rich feature information of optical RS data deeply. Subsequently, two relation-guided fusion strategies are proposed at the dual-level (intradomain and interdomain) to integrate multisource image information effectively. In the intradomain feature fusion, an adaptive de-redundancy fusion module (ADRF) is introduced to eliminate redundancy so that the spatial and spectral features are complete and compact, respectively. In interdomain feature fusion, we construct a spatial-spectral joint attention module (SSJA) based on interdomain relationships to sufficiently enhance the complementary features, so as to facilitate later fusion. Experiments on various multisource optical RS datasets demonstrate that S2 RGF-Net outperforms other state-of-the-art (SOTA) methods.

Open-ICL: Open-Set Modulation Classification via Incremental Contrastive Learning.

Open-set modulation classification (OMC) of signals is a challenging task for handling "unknown" modulation types that are not included in the training dataset. This article proposes an incremental contrastive learning method for OMC, called Open-ICL, to accurately identify unknown modulation types of signals. First, a dual-path 1-D network (DONet) with a classification path (CLP) and a contrast path (COP) is designed to learn discriminative signal features cooperatively. In the COP, the deep features of the input signal are compared with the semantic feature centers (SFCs) of known classes calculated from the network, to infer its signal novelty. An unknown signal bank (USB) is defined to store unknown signals, and a novel moving intersection algorithm (MIA) is proposed to dynamically select reliable unknown signals for the USB. The "unknown" instances, together with SFCs, are continuously optimized and updated, facilitating the process of incremental learning. Furthermore, a dynamic adaptive threshold (DAT) strategy is proposed to enable Open-ICL to adaptively learn changing signal distributions. Extensive experiments are performed on two benchmark datasets, and the results demonstrate the effectiveness of Open-ICL for OMC.

Sacrificial hydrogen bonds enhance the performance of covalently crosslinked composite films derived from soy protein isolate and dialdehyde starch.

Soy protein films have the advantage of being eco-friendly and renewable, but their practical applications are hindered by the mechanical properties. The exceptional tensile strength and fracture toughness of natural silk stem from sacrificial hydrogen bonds it contains that effectively dissipates energy. In this study, we draw inspiration from silk's structural principles to create biodegradable films based on soy protein isolate (SPI). Notably, composite films containing sodium lignosulfonate (LS) demonstrate exceptional strain at break (up to 153%) due to the augmentation of reversible hydrogen bonding, contrasted to films with the addition of solely dialdehyde starch (DAS). The enhancement of tensile strength is realized through a combination of Schiff base cross-linking and sacrificial hydrogen bonding. Furthermore, the incorporation of LS markedly improves the films' ultraviolet (UV) blocking capabilities and hydrophobicity. This innovative design strategy holds great promise for advancing the production of eco-friendly SPI-based films that combine strength and toughness.

Association of smoking cessation with airflow obstruction in workers with silicosis: A cohort study.

BACKGROUND: Studies in general population reported a positive association between tobacco smoking and airflow obstruction (AFO), a hallmark of chronic obstructive pulmonary disease (COPD). However, this attempt was less addressed in silica dust-exposed workers. METHODS: This retrospective cohort study consisted of 4481 silicotic workers attending the Pneumoconiosis Clinic during 1981-2019. The lifelong work history and smoking habits of these workers were extracted from medical records. Spirometry was carried out at the diagnosis of silicosis (n = 4177) and reperformed after an average of 9.4 years of follow-up (n = 2648). AFO was defined as forced expiratory volume in one second (FEV1)/force vital capacity (FVC) less than lower limit of normal (LLN). The association of AFO with smoking status was determined using multivariate logistics regression, and the effect of smoking cessation on the development of AFO was evaluated Cox regression. RESULTS: Smoking was significantly associated with AFO (current smokers: OR = 1.92, 95% CI 1.51-2.44; former smokers: OR = 2.09, 95% CI 1.65-2.66). The risk of AFO significantly increased in the first 3 years of quitting smoking (OR = 1.23, 95% CI 1.02-1.47) but decreased afterwards with increasing years of cessation. Smoking cessation reduced the risk of developing AFO no matter before or after the confirmation of silicosis (pre-silicosis cessation: HR = 0.58, 95% CI 0.46-0.74; post-silicosis cessation: HR = 0.62, 95% CI 0.48-0.79). CONCLUSIONS: Smoking cessation significantly reduced the risk of AFO in the workers with silicosis, although the health benefit was not observed until 3 years of abstinence. These findings highlight the importance of early and long-term smoking cessation among silicotic or silica dust-exposed workers.

Structural annotation of unknown molecules in a miniaturized mass spectrometer based on a transformer enabled fragment tree method.

Structural annotation of small molecules in tandem mass spectrometry has always been a central challenge in mass spectrometry analysis, especially using a miniaturized mass spectrometer for on-site testing. Here, we propose the Transformer enabled Fragment Tree (TeFT) method, which combines various types of fragmentation tree models and a deep learning Transformer module. It is aimed to generate the specific structure of molecules de novo solely from mass spectrometry spectra. The evaluation results on different open-source databases indicated that the proposed model achieved remarkable results in that the majority of molecular structures of compounds in the test can be successfully recognized. Also, the TeFT has been validated on a miniaturized mass spectrometer with low-resolution spectra for 16 flavonoid alcohols, achieving complete structure prediction for 8 substances. Finally, TeFT confirmed the structure of the compound contained in a Chinese medicine substance called the Anweiyang capsule. These results indicate that the TeFT method is suitable for annotating fragmentation peaks with clear fragmentation rules, particularly when applied to on-site mass spectrometry with lower mass resolution.

Brain-Inspired Learning, Perception, and Cognition: A Comprehensive Review.

The progress of brain cognition and learning mechanisms has provided new inspiration for the next generation of artificial intelligence (AI) and provided the biological basis for the establishment of new models and methods. Brain science can effectively improve the intelligence of existing models and systems. Compared with other reviews, this article provides a comprehensive review of brain-inspired deep learning algorithms for learning, perception, and cognition from microscopic, mesoscopic, macroscopic, and super-macroscopic perspectives. First, this article introduces the brain cognition mechanism. Then, it summarizes the existing studies on brain-inspired learning and modeling from the perspectives of neural structure, cognitive module, learning mechanism, and behavioral characteristics. Next, this article introduces the potential learning directions of brain-inspired learning from four aspects: perception, cognition, understanding, and decision-making. Finally, the top-ten open problems that brain-inspired learning, perception, and cognition currently face are summarized, and the next generation of AI technology has been prospected. This work intends to provide a quick overview of the research on brain-inspired AI algorithms and to motivate future research by illuminating the latest developments in brain science.

Multiscale Deep Learning for Detection and Recognition: A Comprehensive Survey.

Recently, the multiscale problem in computer vision has gradually attracted people's attention. This article focuses on multiscale representation for object detection and recognition, comprehensively introduces the development of multiscale deep learning, and constructs an easy-to-understand, but powerful knowledge structure. First, we give the definition of scale, explain the multiscale mechanism of human vision, and then lead to the multiscale problem discussed in computer vision. Second, advanced multiscale representation methods are introduced, including pyramid representation, scale-space representation, and multiscale geometric representation. Third, the theory of multiscale deep learning is presented, which mainly discusses the multiscale modeling in convolutional neural networks (CNNs) and Vision Transformers (ViTs). Fourth, we compare the performance of multiple multiscale methods on different tasks, illustrating the effectiveness of different multiscale structural designs. Finally, based on the in-depth understanding of the existing methods, we point out several open issues and future directions for multiscale deep learning.

Molecularly Thin 2D Organic Single Crystals: A New Platform for High-Performance Polarization-Sensitive Phototransistors.

The inherent linear dichroism (LD), high absorption, and solution processability of organic semiconductors hold immense potential to revolutionize polarized light detection. However, the disordered molecular packing inherent to polycrystalline thin films obscures their intrinsic diattenuation, resulting in diminished polarization sensitivity. In this study, we develop filter-free organic polarization-sensitive phototransistors (PSPs) with both a high linear dichroic ratio (LDR) and exceptional photosensitivity utilizing molecularly thin dithieno[3,2-b:2',3'-d]thiophene derivatives (DTT-8) two-dimensional molecular crystals (2DMCs) as the active layer. The orderly molecular packing in 2DMCs amplifies the inherent LD, and their molecular-scale thickness enables complete channel depletion, significantly reducing the dark current. As a result, PSPs with an impressive LDR of 3.15 and a photosensitivity reaching 3.02 × 106 are obtained. These findings present a practical demonstration of using the polarization angle as an encryption key in optical communication, showcasing the potential of 2DMCs as a viable and promising category of semiconductors for filter-free, polarization-sensitive photodetectors.

Overcoming the Unfavorable Effects of "Boltzmann Tyranny:" Ultra-Low Subthreshold Swing in Organic Phototransistors via One-Transistor-One-Memristor Architecture.

Organic phototransistors (OPTs), as photosensitive organic field-effect transistors (OFETs), have gained significant attention due to their pivotal roles in imaging, optical communication, and night vision. However, their performance is fundamentally limited by the Boltzmann distribution of charge carriers, which constrains the average subthreshold swing (SSave) to a minimum of 60 mV/decade at room temperature. In this study, an innovative one-transistor-one-memristor (1T1R) architecture is proposed to overcome the Boltzmann limit in conventional OFETs. By replacing the source electrode in an OFET with a memristor, the 1T1R device exploits the memristor's sharp resistance state transitions to achieve an ultra-low SSave of 18 mV/decade. Consequently, the 1T1R devices demonstrate remarkable sensitivity to photo illumination, with a high specific detectivity of 3.9 × 109 cm W-1Hz1/2, outperforming conventional OPTs (4.9 × 104 cm W-1Hz1/2) by more than four orders of magnitude. The 1T1R architecture presents a potentially universal solution for overcoming the detrimental effects of "Boltzmann tyranny," setting the stage for the development of ultra-low SSave devices in various optoelectronic applications.

CoT: Contourlet Transformer for Hierarchical Semantic Segmentation.

The Transformer-convolutional neural network (CNN) hybrid learning approach is gaining traction for balancing deep and shallow image features for hierarchical semantic segmentation. However, they are still confronted with a contradiction between comprehensive semantic understanding and meticulous detail extraction. To solve this problem, this article proposes a novel Transformer-CNN hybrid hierarchical network, dubbed contourlet transformer (CoT). In the CoT framework, the semantic representation process of the Transformer is unavoidably peppered with sparsely distributed points that, while not desired, demand finer detail. Therefore, we design a deep detail representation (DDR) structure to investigate their fine-grained features. First, through contourlet transform (CT), we distill the high-frequency directional components from the raw image, yielding localized features that accommodate the inductive bias of CNN. Second, a CNN deep sparse learning (DSL) module takes them as input to represent the underlying detailed features. This memory-and energy-efficient learning method can keep the same sparse pattern between input and output. Finally, the decoder hierarchically fuses the detailed features with the semantic features via an image reconstruction-like fashion. Experiments demonstrate that CoT achieves competitive performance on three benchmark datasets: PASCAL Context 57.21% mean intersection over union (mIoU), ADE20K (54.16% mIoU), and Cityscapes (84.23% mIoU). Furthermore, we conducted robustness studies to validate its resistance against various sorts of corruption. Our code is available at: https://github.com/yilinshao/CoT-Contourlet-Transformer.

Whole-genome variants dataset of 209 local chickens from China.

Compared to commercial chickens, local breeds exhibit better in meat quality and flavour, but the productivity (e.g., growth rate, body weight) of local chicken breeds is rather low. Genetic analysis based on whole-genome sequencing contributes to elucidating the genetic markers or putative candidate genes related to some economic traits, facilitating the improvement of production performance, the acceleration of breeding progress, and the conservation of genetic resources. Here, a total of 209 local chickens from 13 breeds were investigated, and the observation of approximately 91.4% high-quality sequences (Q30 > 90%) and a mapping rate over 99% for each individual indicated good results of this study, as confirmed by a genome coverage of 97.6%. Over 19 million single nucleotide polymorphisms (SNPs) and 1.98 million insertion-deletions (InDels) were identified using the reference genome (GRCg7b), further contributing to the public database. This dataset provides valuable resources for studying genetic diversity and adaptation and for the cultivation of new chicken breeds/lines.

Learning Cross-Domain Features With Dual-Path Signal Transformer.

The past decade has witnessed the rapid development of deep neural networks (DNNs) for automatic modulation classification (AMC). However, most of the available works learn signal features from only a single domain via DNNs, which is not reliable enough to work in uncertain and complex electromagnetic environments. In this brief, a new cross-domain signal transformer (CDSiT) is proposed for AMC, to explore the latent association between different domains of signals. By constructing a signal fusion bottleneck (SFB), CDSiT can implicitly fuse and classify signal features with complementary structures in different domains. Extensive experiments are performed on RadioML2016.10A and RadioML2018.01A, and the results show that CDSiT outperforms its counterparts, particularly for some modulation modes that are difficult to classify before. Through ablation experiences, we also verify the effectiveness of each module in CDSiT.

Association of preserved ratio impaired spirometry with mortality: a systematic review and meta-analysis.

BACKGROUND: Preserved ratio impaired spirometry (PRISm) is prevalent within the general population. Increased mortality has been reported among subjects with PRISm, but the evidence has never been summarised. This systematic review aims to synthesise evidence on the association between PRISm and the risk of all-cause, cardiovascular and respiratory-related mortality. METHODS: We systematically searched MEDLINE, Embase and Web of Science for population-based cohort studies from inception to April 2023 using the terms related to impaired spirometry and mortality. Titles and abstracts were screened to identify eligible studies that reported mortality estimates for individuals with PRISm. We excluded studies that adopted other definitions of impaired spirometry, had a specific study setting (e.g. HIV patients), had an insufficient follow-up period (<1 year) or reported duplicated data. Random-effects meta-analysis was used to produce pooled hazard ratio (HR) with 95% confidence intervals. Between-study heterogeneity was assessed with I2. RESULTS: Eight studies met the inclusion criteria involving 40 699 individuals with PRISm. All included studies reported increased risk of all-cause mortality among adults with PRISm. Meta-analysis showed that PRISm was associated with an increased risk of all-cause mortality (pooled HR 1.71, 95% CI 1.51-1.93; I2=64%), cardiovascular mortality (pooled HR 1.57, 95% CI 1.44-1.72; I2=35%) and respiratory-related mortality (pooled HR 1.97, 95% CI 1.55-2.49; I2=0%). CONCLUSIONS: Individuals with PRISm have a significantly increased risk of mortality compared with those with normal spirometry.

Safety, tolerability, and immunogenicity of a CpG/Alum adjuvanted SARS-CoV-2 recombinant protein vaccine (ZR202-CoV) in healthy adults: Preliminary report of a phase 1, randomized, double-blind, placebo-controlled, dose-escalation trial.

This was a phase 1 dose-escalation study of ZR202-CoV, a recombinant protein vaccine candidate containing a pre-fusion format of the spike (S)-protein (S-trimer) combined with the dual-adjuvant system of Alum/CpG. A total of 230 participants were screened and 72 healthy adults aged 18-59 years were enrolled and randomized to receive two doses at a 28-day interval of three different ZR202-CoV formulations or normal saline. We assessed the safety for 28 days after each vaccination and collected blood samples for immunogenicity evaluation. All formulations of ZR202-CoV were well-tolerated, with no observed solicited adverse events ≥ Grade 3 within 7 days after vaccination. No unsolicited adverse events ≥ Grade 3, or serious adverse events related to vaccination occurred as determined by the investigator. After the first dose, detectable immune responses were observed in all subjects. All subjects that received ZR202-CoV seroconverted at 14 days after the second dose by S-binding IgG antibody, pseudovirus and live-virus based neutralizing antibody assays. S-binding response (GMCs: 2708.7 ~ 4050.0 BAU/mL) and neutralizing activity by pseudovirus (GMCs: 363.1 ~ 627.0 IU/mL) and live virus SARS-CoV-2 (GMT: 101.7 ~ 175.0) peaked at 14 days after the second dose of ZR202-CoV. The magnitudes of immune responses compared favorably with COVID-19 vaccines with reported protective efficacy.

Association of spirometric restriction with mortality in the silicotics: a cohort study.

BACKGROUND: Restrictive spirometry pattern (RSP), defined as reduced forced vital capacity (FVC) in absence of airflow obstruction (AFO), is associated with increased risk of mortality in general population. However, evidence in the patients with silicosis is limited. This study was aimed to investigate the relationship between RSP and the risk of death in a silicotic cohort. METHOD: This retrospective cohort study used data from the Pneumoconiosis Clinic, Hong Kong Department of Health that containing 4315 patients aged 18-80 years and diagnosed with silicosis during 1981-2019, with a follow-up till 31 December 2019. Spirometry was carried out at the diagnostic examination of silicosis. Lung function categories were classified as normal spirometry (FEV1/FVC ≥ 0.7, FVC ≥ 80% predicted), RSP only (FEV1/FVC ≥ 0.7, FVC < 80% predicted), AFO only (FEV1/FVC < 0.7, FVC ≥ 80% predicted), and RSP&AFO mixed (FEV1/FVC < 0.7, FVC < 80% predicted). The hazard ratio (HR) and 95% confidence intervals (95% CI) were computed using a Cox proportional hazards model adjusting for age, body mass index, history of tuberculosis, smoking status, pack-years, and radiographic characteristics of silicotic nodules. RESULTS: Among the 4315 patients enrolled in the study, the prevalence of RSP was 24.1% (n = 1038), including 11.0% (n = 473) with RSP only and 13.1% (n = 565) with mixed RSP and AFO. During the follow-up period, a total of 2399 (55.6%) deaths were observed. Compared with the silicotics with normal spirometry, those with RSP only had significantly increased risk of all-cause mortality (HR = 1.63, 95% CI 1.44-1.85) and respiratory-related mortality (HR = 1.56, 95% CI 1.31-1.85). Notably, a higher risk of mortality was observed in silicotics with mixed ventilatory defects of both RSP and AFO (all-cause mortality: HR = 2.22, 95% CI 1.95-2.52; respiratory-related mortality: HR = 2.59, 95% CI 2.18-3.07) than in those with RSP only. CONCLUSION: RSP is significantly associated with increased risk of all-cause and respiratory-related mortality in the silicotics, and patients with mixed restrictive and obstructive ventilatory defect have higher risk of mortality than those with single RSP or AFO. These findings emphasize the importance of recognizing RSP in the occupational settings, especially for the silicotic patients with mixed ventilatory defect.

Multiresolution Interpretable Contourlet Graph Network for Image Classification.

Modeling contextual relationships in images as graph inference is an interesting and promising research topic. However, existing approaches only perform graph modeling of entities, ignoring the intrinsic geometric features of images. To overcome this problem, a novel multiresolution interpretable contourlet graph network (MICGNet) is proposed in this article. MICGNet delicately balances graph representation learning with the multiscale and multidirectional features of images, where contourlet is used to capture the hyperplanar directional singularities of images and multilevel sparse contourlet coefficients are encoded into graph for further graph representation learning. This process provides interpretable theoretical support for optimizing the model structure. Specifically, first, the superpixel-based region graph is constructed. Then, the region graph is applied to code the nonsubsampled contourlet transform (NSCT) coefficients of the image, which are considered as node features. Considering the statistical properties of the NSCT coefficients, we calculate the node similarity, i.e., the adjacency matrix, using Mahalanobis distance. Next, graph convolutional networks (GCNs) are employed to further learn more abstract multilevel NSCT-enhanced graph representations. Finally, the learnable graph assignment matrix is designed to get the geometric association representations, which accomplish the assignment of graph representations to grid feature maps. We conduct comparative experiments on six publicly available datasets, and the experimental analysis shows that MICGNet is significantly more effective and efficient than other algorithms of recent years.

Integration of Metabolomic and Transcriptomic Provides Insights into Anti-Inflammatory Response to trans-10-Hydroxy-2-decenoic Acid on LPS-Stimulated RAW 264.7 Cells.

Trans-10-hydroxy-2-decenoic acid (10-HDA) is a unique fatty acid found in royal jelly that possesses potential health benefits such as anti-inflammatory. However, further research is needed to fully understand its mechanisms of action and therapeutic potential for inflammation-associated diseases. In this present study, liquid chromatography-tandem mass spectrometry (LC-MS/MS) and RNA-seq analyses were conducted to comprehensively analyze the in vitro anti-inflammatory effects of 10-HDA on lipopolysaccharide (LPS)-stimulated RAW 264.7 cells. Our results demonstrated that 128 differentially expressed metabolites and 1721 differentially expressed genes were identified in the 10-HDA-treated groups compared to the LPS groups. Metabolites were significantly enriched in amino acid metabolism pathways, including methionine metabolism, glycine and serine metabolism, and tryptophan metabolism. The differentially expressed genes enrichment analysis indicated that antigen processing and presentation, NOD-like receptor signaling pathway, and arginine biosynthesis were enriched with the administration of 10-had. The correlation analysis revealed that glycerophospholipid metabolism and s-adenosylmethionine-dependent methylation processes might be involved in the response to the 10-HDA treatment. Overall, the findings from this study showed that 10-HDA might involve the modulation of certain signaling pathways involved in the inflammatory response, but further research is needed to determine the safety and efficacy as a therapeutic agent.

A Fast Evolutionary Knowledge Transfer Search for Multiscale Deep Neural Architecture.

The emergence of neural architecture search (NAS) algorithms has removed the constraints on manually designed neural network architectures, so that neural network development no longer requires extensive professional knowledge, trial and error. However, the extremely high computational cost limits the development of NAS algorithms. In this article, in order to reduce computational costs and to improve the efficiency and effectiveness of evolutionary NAS (ENAS) is investigated. In this article, we present a fast ENAS framework for multiscale convolutional networks based on evolutionary knowledge transfer search (EKTS). This framework is novel, in that it combines global optimization methods with local optimization methods for search, and searches a multiscale network architecture. In this article, evolutionary computation is used as a global optimization algorithm with high robustness and wide applicability for searching neural architectures. At the same time, for fast search, we combine knowledge transfer and local fast learning to improve the search speed. In addition, we explore a multiscale gray-box structure. This gray box structure combines the Bandelet transform with convolution to improve network approximation, learning, and generalization. Finally, we compare the architectures with more than 40 different neural architectures, and the results confirmed its effectiveness.

A Complex-Former Tracker With Dynamic Polar Spatio-Temporal Encoding.

Recently, the excellent performance of transformer has attracted the attention of the visual community. Visual transformer models usually reshape images into sequence format and encode them sequentially. However, it is difficult to explicitly represent the relative relationship in distance and direction of visual data with typical 2-D spatial structures. Also, the temporal motion properties of consecutive frames are hardly exploited when it comes to dynamic video tasks like tracking. Therefore, we propose a novel dynamic polar spatio-temporal encoding for video scenes. We use spiral functions in polar space to fully exploit the spatial dependences of distance and direction in real scenes. We then design a dynamic relative encoding mode for continuous frames to capture the continuous spatio-temporal motion characteristics among video frames. Finally, we construct a complex-former framework with the proposed encoding applied to video-tracking tasks, where the complex fusion mode (CFM) realizes the effective fusion of scenes and positions for consecutive frames. The theoretical analysis demonstrates the feasibility and effectiveness of our proposed method. The experimental results on multiple datasets validate that our method can improve tracker performance in various video scenarios.