Search

The identification of plant disease leaves based on deep learning is the key to control the development and spread of plant diseases. In this paper, the existing problems of traditional classification and recognition of plant disease leaves and the limitations of deep learning-based plant disease leaf training are analysed. An enhanced GAN model network based on the Wasserstein GAN loss function has been developed to address the limited training images of plant disease leaves. The self-attention layer is added into the self-encoding structure of the generating network. The effectiveness of data generated by the encoder is increased after the self-attention layer is added after the convolution. Finally, the model's training process is stabilised using the depth gradient punishment method. Three types of corn disease photos and 100 health images from the PlantVillage dataset were used as data sets in the experiment. An AWGAN model was applied to generate around 3000 images. Several data improvement techniques were applied to augment the same datasets. Comparative tests are conducted using AlexNet, VGG16, and ResNet18. The results indicate that the proposed AWGAN model is capable of generating sufficient images of maize leaf diseases with apparent lesions, making it a viable solution for data augmentation of plant disease images. The training model's recognition accuracy is significantly increased. The proposed awGAN-based image identification method for plant leaf disease efficiently resolves the over-fitting problem in the small sample training set. The model recognition accuracy in the ResNet18 network achieves 98.4%.

The aim of this study to evaluate the maxillary and mandibular arch expansion efficiency of clear aligner. A total of 15 patients aged between 8–11 who were admitted from January 2020 to December 2021. The 3ship digital oral scanner was used to establish a digital oral model for the patients before and after the correction. The three-dimensional model before and after correction was established by ClinCheck software to measure and record the width of each tooth position and evaluate the efficiency of arch expansion. The dental arch width was greater after correction of the width between the maxillary and maxillary canines, the width between the first premolars, the width between the second premolars, and the width between the first molars, showing statistically significant differences (P the maxillary first premolar efficiency > the maxillary second premolar efficiency > the maxillary first molar efficiency. Through the linear regression equation, the preset expansion amount of each tooth position was reasonably predicted.Taken together our data suggested that the clear aligner efficiency can be improved by adding accessories and torque in the base bone range when necessary.

This paper is concerned with existence and multiplicity results for the semilinear subelliptic equation with free perturbation term. By using the degenerate Rellich-Kondrachov compact embedding theorem, precise lower bound estimates of Dirichlet eigenvalues for the finitely degenerate elliptic operator and minimax method, we obtain the existence and multiplicity of weak solutions for the problem.

The fibroblast growth factor receptor (FGFR) belongs to the tyrosine kinase family consisting of four members (FGFR1-4). This study involved identification and characterization of FGFR1 and FGFR3 from mud crab Scylla paramamosain for the first time. The obtained cDNAs of SpFGFR1 and SpFGFR3 were 2,380 bp and 2,982 bp in length with a 1,503 bp and 2,310 bp open reading frame, respectively. The predicted SpFGFR1 protein included three immunoglobulin domains and a transmembrane region, while SpFGFR3 protein possessed a typical TyrKc (Tyrosine kinase, catalytic) domain. Real-time PCR analysis showed that SpFGFR1 and SpFGFR3 were highly expressed in the hepatopancreas. Furthermore, the expression levels of SpFGFR1 and SpFGFR3 in the hepatopancreas were enhanced following challenges with Vibro alginolyticus, Staphylococcus aureus, Poly (I:C) and White spot syndrome virus, which shows the involvement of SpFGFR1 and SpFGFR3 in innate immune response to infections from bacteria and virus. There was significant suppression of six antimicrobial peptide genes (SpALF1-5 and SpCrustin) and three NF-κB members (SpDorsal, SpIKK and SpRelish) when SpFGFR1 and SpFGFR3 was interfered in vivo. Also, treatment of the hemocytes with specific inhibitor of SpFGFR for 24 h consistently down-regulated SpDorsal, SpRelish and AMPs. These results suggested that SpFGFR1 and SpFGFR3 played important roles in regulating the Toll signaling pathway and immune deficiency (IMD) pathway through NF-κB signaling pathway. These findings may provide new insights into the role of FGFRs in the innate immune function of crustaceans.

In recent years, ubiquitous semantic Metaverse has been studied to revolutionize immersive cyber-virtual experiences for augmented reality (AR) and virtual reality (VR) users, which leverages advanced semantic understanding and representation to enable seamless, context-aware interactions within mixed-reality environments. This survey focuses on the intelligence and spatio-temporal characteristics of four fundamental system components in ubiquitous semantic Metaverse, i.e., artificial intelligence (AI), spatio-temporal data representation (STDR), semantic Internet of Things (SIoT), and semantic-enhanced digital twin (SDT). We thoroughly survey the representative techniques of the four fundamental system components that enable intelligent, personalized, and context-aware interactions with typical use cases of the ubiquitous semantic Metaverse, such as remote education, work and collaboration, entertainment and socialization, healthcare, and e-commerce marketing. Furthermore, we outline the opportunities for constructing the future ubiquitous semantic Metaverse, including scalability and interoperability, privacy and security, performance measurement and standardization, as well as ethical considerations and responsible AI. Addressing those challenges is important for creating a robust, secure, and ethically sound system environment that offers engaging immersive experiences for the users and AR/VR applications., 18 pages, 7 figures, 3 tables

We consider the problem of video snapshot compressive imaging (SCI), where sequential high-speed frames are modulated by different masks and captured by a single measurement. The underlying principle of reconstructing multi-frame images from only one single measurement is to solve an ill-posed problem. By combining optimization algorithms and neural networks, deep unfolding networks (DUNs) score tremendous achievements in solving inverse problems. In this paper, our proposed model is under the DUN framework and we propose a 3D Convolution-Transformer Mixture (CTM) module with a 3D efficient and scalable attention model plugged in, which helps fully learn the correlation between temporal and spatial dimensions by virtue of Transformer. To our best knowledge, this is the first time that Transformer is employed to video SCI reconstruction. Besides, to further investigate the high-frequency information during the reconstruction process which are neglected in previous studies, we introduce variance estimation characterizing the uncertainty on a pixel-by-pixel basis. Extensive experimental results demonstrate that our proposed method achieves state-of-the-art (SOTA) (with a 1.2dB gain in PSNR over previous SOTA algorithm) results. We will release the code.

Snapshot compressive imaging emerges as a promising technology for acquiring real-world hyperspectral signals. It uses an optical encoder and compressively produces the 2D measurement, followed by which the 3D hyperspectral data can be retrieved via training a deep reconstruction network. Existing reconstruction models are trained with a single hardware instance, whose performance is vulnerable to hardware perturbation or replacement, demonstrating an overfitting issue to the physical configuration. This defect limits the deployment of pre-trained models since they would suffer from large performance degradation when are assembled to unseen hardware. To better facilitate the reconstruction model with new hardware, previous efforts resort to centralized training by collecting multi-hardware and data, which is impractical when dealing with proprietary assets among institutions. In light of this, federated learning (FL) has become a feasible solution to enable cross-hardware cooperation without breaking privacy. However, the naive FedAvg is subject to client drift upon data heterogeneity owning to the hardware inconsistency. In this work, we tackle this challenge by marrying prompt tuning with FL to snapshot compressive imaging for the first time and propose an federated hardware-prompt learning (FedHP) method. Rather than mitigating the client drift by rectifying the gradients, which only takes effect on the learning manifold but fails to touch the heterogeneity rooted in the input data space, the proposed FedHP globally learns a hardware-conditioned prompter to align the data distribution, which serves as an indicator of the data inconsistency stemming from different pre-defined coded apertures. Extensive experiments demonstrate that the proposed method well coordinates the pre-trained model to indeterminate hardware configurations., 11 figures, 4 tables

The epidermal growth factor receptor (EGFR) is a pleiotropic glycoprotein which plays a role in regulating cell proliferation, migration and differentiation. However, the genetic diversity of EGFR in crustaceans as well as its function, such as whether it is involved in immune regulation, remains obscure. In this study, two EGFR genes, including EGFR1 and EGFR2, and three transcripts were identified and characterized in Scylla Paramamosain for the first time. To our knowledge, this is the first time that more than one EGFR gene was identified in a single species. The complete open reading frames (ORFs) of SpEGFR1, SpEGFR2a and SpEGFR2b were 4377 bp, 4404 bp and 4341 bp encoding deduced proteins of 1458 amino acids (aa), 1467 aa and 1446 aa, respectively. All EGFR had a signal peptide region and two Recep_L_domain region, followed by a transmembrane region and a conserved tyrosine kinase domain (TyrKc), and phylogenetic analysis demonstrated three SpEGFRs clustered together with invertebrate EGFR branch. Tissue specific expression analysis depicted that all SpEGFRs presented similar transcription patterns. The expression levels of SpEGFR1 and SpEGFR2s in hepatopancreas and gills were significantly altered after the stimulation of bacterial and viral pathogens including Staphylococcus aureus, Vibrio alginolyticus, White spot syndromre virus and Polycytidylinic acid. The in vivo RNA interference assays demonstrated that expression levels of SpIKK, two members of NF-κB (SpRelish and SpDorsal) and six antimicrobial peptide (AMP) genes (SpCrustin and SpALF1-5) were significantly reduced when SpEGFR1 or SpEGFR2 was silenced, respectively. The transcription patterns of SpIKK, SpRelish, SpDorsal and AMPs exhibited similar down- or up-regulation trend when the primary cultured hemocytes were treated with EGFR antagonist or agonist for 24 h. These results suggested that SpEGFR might play an important role in innate immune responses to bacterial and viral infections by regulating the NF-κB pathway. It also provided a better understanding of the origin or evolution of EGFR in crustaceans and even invertebrates.

Streams and rivers are important sources of nitrous oxide (N

Mechanical allodynia impinges on the life quality of patients. Hen Egg Lysozyme (HEL) is a substance extracted from eggs that is commonly used to inhibit bacterial activity. The role of HEL in regulating and treating pain is unclear. Here, we find that HEL selectively attenuates static mechanical allodynia of mice induced by complete Freund's adjuvant (CFA), spinal nerve ligation (SNL) and chemotherapeutic agent. RNA-seq screening reveals that CFA significantly reduces the expression of Parkin in dorsal root ganglion (DRG) neurons of mice, while pre-administration of HEL increases the expression of Parkin and remits the static mechanical allodynia induced by Parkin-siRNA. Moreover, HEL increases the interaction between nuclear respiratory factor 1 (NRF1) and histone acetyltransferase P300 and then enhances the NRF1 mediated histone acetylation in prkn promoter region in DRGs of mice. Further, Parkin interacts with mechanotransducing ion channel TACAN (Tmem120a) and knockdown of Parkin significantly increases the membrane trafficking of TACAN in sensory neurons of mice. While pre-administration of HEL inhibits the increased membrane trafficking of TACAN in sensory neurons of mice induced by Parkin-siRNA. In addition, pre-given of HEL also significantly attenuates the static mechanical allodynia induced by overexpression of TACAN in mice, and the effect of HEL can be blocked by Parkin-siRNA. This indicates that HEL increases the expression of Parkin through epigenetic mechanisms and then decreases TACAN membrane trafficking in sensory neurons to relieve static mechanical hypersensitivity. Therefore, we reveal a novel function of HEL, which is a potential substance for the treatment of static mechanical pain.

The problem of finite-time adaptive tracking control against event-trigger error is investigated in this article for a type of uncertain nonlinear systems. By fusing the techniques of command filter backstepping technical and event-triggered control (ETC), an adaptive event-triggered design method is proposed to construct the controller, under which the effect of event-triggered error can be compensated completely. Moreover, the proposed controller can increase robustness against uncertainties and event error in the backstepping design framework. In particular, we establish the finite-time convergence condition under which the tracking error asymptotically converges to zero in finite time with the aid of a scaling function. Detailed and rigorous stability proofs are given by making use of the improved finite time stability criterion. Two simulation examples are provided to exhibit the validity of the designed adaptive ETC approach.

Smads, part of signaling cascades that represent downstream pathways of the TGF-β super family proteins, are pleiotropic cytokines with important role in mediating cell proliferation, homeostasis, differentiation, apoptosis and immune response. However, the specific functions of Smads remain unknown in crustaceans. In the present study, the drosophila mothers against decapentaplegic protein gene 1 (Smad1) was firstly identified and characterized from the Red Swamp Crayfish Procambarus clarkii. The obtained cDNA sequence of pcSmad1was 2, 503 bp long with a 1, 488 bp open reading fame, which encoded a putative protein of 496 amino acids. Furthermore, pcSmad1 responded to both Aeromonas hydrophila and WSSV challenge, suggesting the involvement of pcSmad1 in innate immune responses. Knockdown of pcSmad1 in vivo dramatically increased the expressions of NF-κB signaling genes and anti-lipopolysaccharide factor genes. Additionally, overexpression of pcSmad1 in HEK293T cells could markedly activate NF-κB signaling. Taken together, these results indicated that pcSmad1 played an immune-regulatory role in crayfish's innate immunity, which may provide a better understanding of TGF-β superfamily members in crustacean.

High-grade neuroendocrine tumors (NETs) of the lung consist of small-cell lung cancer (SCLC) and large-cell neuroendocrine carcinoma (LCNEC). Both exhibit aggressive malignancy with poor prognosis. The transformation of lung adenocarcinoma (ADC) to SCLC or LCNEC also contributes to acquired resistance to epidermal growth factor receptor (EGFR)-tyrosine kinase inhibitors (TKIs). Despite initially being responsive to chemotherapy, high-grade NET patients inevitably develop drug resistance; thus, novel therapeutic targets are urgently needed for these patients. Our study reported that VGF (nerve growth factor inducible), a factor mainly expressed in neurons during neural development, is highly expressed in SCLC and LCNEC as well as in a subset of ADCs, whereas targeting VGF attenuates cancer cell growth and tumor formation. High VGF expression was associated with advanced stage SCLC and predicted poor prognosis in lung ADC. In addition, EGFR-TKI selection enriched VGF expression in TKI-resistant ADC under epigenetic control. The VGF locus possessed the HDAC1 binding site, and treatment of ADC cells with the HDAC1 inhibitor induced VGF expression. High VGF expression was associated with chemoresistance, and silencing VGF induced BMF and BCL2L11 expression and rendered lung cancer cells sensitive to chemotherapy drugs. These findings suggested the potential of VGF as a prognostic factor and therapeutic target in lung cancers with neuroendocrine feature.

In this paper, we propose a novel complex convolutional neural network (CNN) CSI enhancer for integrated sensing and communications (ISAC), which exploits the correlation between the sensing parameters (such as angle-of-arrival and range) and the channel state information (CSI) to significantly improve the CSI estimation accuracy and further enhance the sensing accuracy. Within the CNN CSI enhancer, we use the complex-valued computation layers to form the CNN, which maintains the phase information of CSI. We also transform the CSI into the sparse angle-delay domain, leading to heatmap images with prominent peaks that can be efficiently processed by CNN. Based on the enhanced CSI outputs, we further propose a novel biased fast Fourier transform (FFT)-based sensing scheme for improving the range sensing accuracy, by artificially introducing phase biasing terms. Extensive simulation results show that the ISAC complex CNN CSI enhancer can converge within 30 training epochs. The normalized mean square error (NMSE) of its CSI estimates is about 17 dB lower than that of the linear minimum mean square error (LMMSE) estimator, and the bit error rate (BER) of demodulation using the enhanced CSI estimation approaches that with perfect CSI. Finally, the range estimation MSE of the proposed biased FFT-based sensing method approaches that of the subspace-based sensing method, at a much lower complexity., 13 pages, 15 figures, submitted to IEEE Journal of Selected Topics in Signal Processing

Diabetic retinopathy (DR) is a leading cause of irreversible vision loss in working-age populations. FTO is an N6-methyladenosine (m6A) demethylase that participates in various biological events, while its role in DR remains elusive. Herein, we detected elevated FTO expression in retinal proliferative membranes of DR patients. FTO promoted endothelial cell (EC) cell cycle progression and tip cell formation to facilitate angiogenesisin vitro, in mice and in zebrafish. FTO also regulated EC-pericyte crosstalk to trigger diabetic microvascular leakage, and mediated EC-microglia interactions to induce retinal inflammation and neurodegenerationin vivoandin vitro. Mechanistically, FTO affected EC features via modulatingCDK2mRNA stability in an m6A-YTHDF2-dependent manner. FTO up-regulation under diabetic conditions was driven by lactate mediated histone lactylation. FB23-2, an inhibitor to FTO’s m6A demethylase activity, suppressed angiogenic phenotypesin vivoandin vitro. Noteworthy, we developed a nanoplatform encapsulating FB23-2 for systemic administration, and confirmed its targeting and therapeutic efficiencies in mice. Collectively, our study demonstrated that FTO coordinates EC biology and retinal homeostasis in DR, providing a promising nanotherapeutic approach for DR.

Existing deep learning models for hyperspectral image (HSI) reconstruction achieve good performance but require powerful hardwares with enormous memory and computational resources. Consequently, these methods can hardly be deployed on resource-limited mobile devices. In this paper, we propose a novel method, Binarized Spectral-Redistribution Network (BiSRNet), for efficient and practical HSI restoration from compressed measurement in snapshot compressive imaging (SCI) systems. Firstly, we redesign a compact and easy-to-deploy base model to be binarized. Then we present the basic unit, Binarized Spectral-Redistribution Convolution (BiSR-Conv). BiSR-Conv can adaptively redistribute the HSI representations before binarizing activation and uses a scalable hyperbolic tangent function to closer approximate the Sign function in backpropagation. Based on our BiSR-Conv, we customize four binarized convolutional modules to address the dimension mismatch and propagate full-precision information throughout the whole network. Finally, our BiSRNet is derived by using the proposed techniques to binarize the base model. Comprehensive quantitative and qualitative experiments manifest that our proposed BiSRNet outperforms state-of-the-art binarization methods and achieves comparable performance with full-precision algorithms. Code and models will be released at https://github.com/caiyuanhao1998/BiSCI and https://github.com/caiyuanhao1998/MST, The first work to study binarized spectral compressive imaging reconstruction problem

Machine Learning (ML) models contain private information, and implementing the right to be forgotten is a challenging privacy issue in many data applications. Machine unlearning has emerged as an alternative to remove sensitive data from a trained model, but completely retraining ML models is often not feasible. This survey provides a concise appraisal of Machine Unlearning techniques, encompassing both exact and approximate methods, probable attacks, and verification approaches. The survey compares the merits and limitations each method and evaluates their performance using the Deltagrad exact machine unlearning method. The survey also highlights challenges like the pressing need for a robust model for non-IID deletion to mitigate fairness issues. Overall, the survey provides a thorough synopsis of machine unlearning techniques and applications, noting future research directions in this evolving field. The survey aims to be a valuable resource for researchers and practitioners seeking to provide privacy and equity in ML systems., 7pages, 5 figures, 1 table

The mechanism of deleted in lymphocytic leukemia 2 (DLEU2)-long non-coding RNA in tumors has become a major point of interest in recent research related to the occurrence and development of a variety of tumors. Recent studies have shown that the long non-coding RNA DLEU2 (lncRNA-DLEU2) can cause abnormal gene or protein expression by acting on downstream targets in cancers. At present, most lncRNA-DLEU2 play the role of oncogenes in different tumors, which are mostly associated with tumor characteristics, such as proliferation, migration, invasion, and apoptosis. The data thus far show that because lncRNA-DLEU2 plays an important role in most tumors, targeting abnormal lncRNA-DLEU2 may be an effective treatment strategy for early diagnosis and improving the prognosis of patients. In this review, we integrated lncRNA-DLEU2 expression in tumors, its biological functions, molecular mechanisms, and the utility of DLEU2 as an effective diagnostic and prognostic marker of tumors. This study aimed to provide a potential direction for the diagnosis, prognosis, and treatment of tumors using lncRNA-DLEU2 as a biomarker and therapeutic target.

Introduction: We aimed to evaluate the influence of 1,25-dihydroxyvitamin D (1,25(OH)2D) on metabolic dysfunction and elucidate its underlying mechanism using a rat model of polycystic ovary syndrome (PCOS).Methods: Twenty-four Sprague-Dawley rats were randomly divided into four groups: control group (CON, 2 ml/kg of oral 0.5% CMC), 1,25VD group (oral 0.5% CMC and 2.5 ug/kg intraperitoneal 1,25(OH)2D), PCOS group (1 mg/kg oral letrozole), PCOS+1,25VD group (1 mg/kg oral letrozole orally 2.5 ug/kg intraperitoneal 1,25(OH)2D). The treatments were administered for 8 weeks. Body weight, estrus cycle, insulin tolerance, and oral glucose tolerance of the rats in the different groups were assessed. The rats were euthanized at the 8th weeks, and plasma, ovarian, and liver samples were collected and analyzed. The hepatic lipid profile was characterized using HPLC/MRM.Results: Letrozole-induced PCOS rats exhibited increased weight, insulin resistance, postprandial glucose abnormalities, and dyslipidemia. Compared with the PCOS group rats, the PCOS+1,25VD group rats showed reduced body weight, increased sensitivity to insulin, decreased postprandial glucose, and elevated levels of high-density lipoprotein cholesterol. Moreover, abnormally increased liver concentrations of total diacylglycerol (DG) and DG species in the PCOS rats were reversed by treatment with 1,25(OH)2D. Additionally, hepatic DG and insulin sensitivity were correlated.Conclusion: 1,25(OH)2D inhibited hepatic DG accumulation and ameliorated metabolic dysfunction in PCOS rat models.

XLS file - 204K, Genomic regions enriched in MLLN, AF4C, H3K79me2 and the common of the three ChIP-seq in RS4;11 cells

XLS file - 330K, Genes included in BCR-ABL1, E2A-PBX1 or MLLr DNA methylation signatures vs. normal pre-B cells

XLS file - 176K, DNA methylation and gene expression signature of MLLr ALL

XLS file - 136K, DNA methylation and gene expression signature of E2A-PBX1-positive B-ALL

XLS file - 60K, Characteristics of the 215 E2993 B-ALL patients and the 12 normal bone marrow samples (excel file)