Search

We report a simple and convenient N-terminal thiazolidine (Thz) deprotection strategy and its application in one-pot multisegment ligation. In this strategy, O -benzylhydroxylamine (O-BHA) is used to efficiently and rapidly convert Thz into N-terminal cysteine. O-BHA can be easily separated from the ligation buffer by organic solvent extraction, avoiding the degradation of the peptide thioester by O-BHA. The utility of the O-BHA-based one-pot ligation strategy has been demonstrated in the assembly of CC chemokine ligand-2.

Three p-terphenyl metabolites (1-3), three indole-diterpenoids (4-6), an herbicide sesquiterpene (7), a flavonoid (8), and five other small molecules containing nitrogen (9-13) were isolated from the medicinal insect (Periplaneta americana)-derived endophytic Aspergillus taichungensis SMU01. Their chemical structures were elucidated on the basis of spectroscopic data and quantum chemical computational methods. Biological activity of these isolates in the differentiation of mouse CD4 + T cell subsets was evaluated. Importantly, metabolites 2 targeting JAK-STAT signaling pathway could hold potential benefits in maintaining peripheral immune homeostasis and alleviating the progression of autoimmune diseases., Competing Interests: Declaration of competing interest The authors declare no competing financial interest., (Copyright © 2023. Published by Elsevier B.V.)

Bone proliferation is an important pathological feature of inflammatory rheumatic diseases. Although recent advance in high-resolution peripheral quantitative computed tomography (HR-pQCT) enables physicians to study microarchitectures, physicians' annotation of proliferation suffers from slice inconsistency and subjective variations. Also, there are only few effective automatic or semi-automatic tools for proliferation detection. In this study, by integrating pathological knowledge of proliferation formation with the advancement of statistical shape analysis theory, we present an unsupervised method, named Deformation-Controllable Elastic Shape model, for 3D bone Proliferation Analysis (DCES-PA). Unlike previous shape analysis methods that directly regularize the smoothness of the displacement field, DCES-PA regularizes the first and second-order derivative of the displacement field and decomposes these vector fields according to different deformations. For the first-order elastic metric, DCES-PA orthogonally decomposes the first-order derivative of the displacement field by shearing, scaling and bending deformation, and then penalize deformations triggering proliferation formation. For the second-order elastic metric, DCES-PA encodes both intrinsic and extrinsic surface curvatures into the second-order derivative of the displacement field to control the generation of high-curvature regions. By integrating the elastic shape metric with the varifold distances, DCES-PA achieves correspondence-free shape analysis. Extensive experiments on both simulated and real clinical datasets demonstrate that DCES-PA not only shows an improved accuracy than other state-of-the-art shape-based methods applied to proliferation analysis but also produces highly sensitive proliferation annotations to assist physicians in proliferation analysis., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2024 Elsevier Ltd. All rights reserved.)

This article proposes a survival model based on graph convolutional networks (GCNs) with geometric graphs directly constructed from high-dimensional features. First, we clarify that the graphs used in GCNs play an important role in processing the relational information of samples, and the graphs that align well with the underlying data structure could be beneficial for survival analysis. Second, we show that sparse geometric graphs derived from high-dimensional data are more favorable compared with dense graphs when used in GCNs for survival analysis. Third, from this insight, we propose a model for survival analysis based on GCNs. By using multiple sparse geometric graphs and a proposed sequential forward floating selection algorithm, the new model is able to simultaneously perform survival analysis and unveil the local neighborhoods of samples. The experimental results on real-world datasets show that the proposed survival analysis approach based on GCNs outperforms a variety of existing methods and indicate that geometric graphs can aid survival analysis of high-dimensional data.

DNA modifications add another layer of complexity to the eukaryotic genome to regulate gene expression, playing critical roles as epigenetic marks. In eukaryotes, the study of DNA epigenetic modifications has been confined to 5mC and its derivatives for decades. However, rapid developing approaches have witnessed the expansion of DNA modification reservoirs during the past several years, including the identification of 6mA, 5gmC, 4mC, and 4acC in diverse organisms. However, whether these DNA modifications function as epigenetic marks requires careful consideration. In this review, we try to present a panorama of all the DNA epigenetic modifications in eukaryotes, emphasizing recent breakthroughs in the identification of novel DNA modifications. The characterization of their roles in transcriptional regulation as potential epigenetic marks is summarized. More importantly, the pathways for generating or eliminating these DNA modifications, as well as the proteins involved are comprehensively dissected. Furthermore, we briefly discuss the potential challenges and perspectives, which should be taken into account while investigating novel DNA modifications., Competing Interests: Conflict of interest The authors declare that they have no conflicts of interest with the contents of this article., (Copyright © 2024 The Authors. Published by Elsevier Inc. All rights reserved.)

Metric learning aims to learn a distance metric such that semantically similar instances are pulled together while dissimilar instances are pushed away. Many existing methods consider maximizing or at least constraining a distance margin in the feature space that separates similar and dissimilar pairs of instances to guarantee their generalization ability. In this article, we advocate imposing an adversarial margin in the input space so as to improve the generalization and robustness of metric learning algorithms. We first show that the adversarial margin, defined as the distance between training instances and their closest adversarial examples in the input space, takes account of both the distance margin in the feature space and the correlation between the metric and triplet constraints. Next, to enhance robustness to instance perturbation, we propose to enlarge the adversarial margin through minimizing a derived novel loss function termed the perturbation loss. The proposed loss can be viewed as a data-dependent regularizer and easily plugged into any existing metric learning methods. Finally, we show that the enlarged margin is beneficial to the generalization ability by using the theoretical technique of algorithmic robustness. Experimental results on 16 datasets demonstrate the superiority of the proposed method over existing state-of-the-art methods in both discrimination accuracy and robustness against possible noise.

Background: Triglyceride-glucose (TyG) index values are a new surrogate marker for insulin resistance. This study aimed to explore the relationship between cumulative TyG index values and atrial fibrillation (AF) recurrence after radiofrequency catheter ablation (RFCA)., Methods: A total of 576 patients with AF who underwent RFCA at the Second Affiliated Hospital of Xi'an Jiaotong University were included in this study. The participants were grouped based on cumulative TyG index values tertiles within 3 months after ablation. Cox regression and restricted cubic spline analyses were used to determine the relationship between cumulative TyG index values and AF recurrence. The predictive value of all risk factors was assessed by receiver operating curve analysis., Results: There were 375 patients completed the study (age: 63.23 ± 10.73 years, 64.27% male). The risk of AF recurrence increased with increasing cumulative TyG index values tertiles. After adjusting for potential confounders, patients in the medium cumulative TyG index group [hazard ratio (HR) = 4.949, 95% CI: 1.778-13.778, P = 0.002] and the high cumulative TyG index group (HR = 8.716, 95% CI: 3.371-22.536, P < 0.001) had a higher risk of AF recurrence than those in the low cumulative TyG index group. The restricted cubic spline regression model also showed an increased risk of AF recurrence with increasing cumulative TyG index values. When considering cumulative TyG index values, left atrial diameter, and lactate dehydrogenase levels as a comprehensive factor, the model could effectively predict AF recurrence after RFCA [area under the curve (AUC) = 0.847, 95% CI: 0.797-0.897, P < 0.001]., Conclusions: Cumulative TyG index values were a risk factor for AF recurrence after RFCA. Monitoring longitudinal TyG index values may assist with optimized for risk stratification and outcome prediction for AF recurrence., (© 2024 JGC All rights reserved; www.jgc301.com.)

The introduction of trifluoromethyl groups into organic molecules is of paramount importance in modern synthetic chemistry and medicinal chemistry. While methods for constructing C(sp 2 )-CF 3 bonds have been well established, the advancement of practical and comprehensive approaches for forming C(sp 3 )-CF 3 bonds remains considerably restricted. In this work, we describe an efficient and site-specific deaminative trifluoromethylation reaction of aliphatic primary amines to afford the corresponding alkyl trifluoromethyl compounds. The reaction proceeds at room temperature with readily accessible N-anomeric amide (Levin's reagent) and bench-stable bpyCu(CF 3 ) 3 (Grushin's reagent, bpy=2,2'-bipyridine) under blue light. The protocol features mild reaction conditions, good functional group tolerance, and moderate to good yields. Remarkably, the method can be applied to the direct, late-stage trifluoromethylation of natural products and bioactive molecules. Experimental mechanistic studies were conducted, and a radical mechanism is proposed, wherein the dual roles of Grushin's reagent have been elucidated., (© 2024 Wiley-VCH GmbH.)

Recent methods in network pruning have indicated that a dense neural network involves a sparse subnetwork (called a winning ticket), which can achieve similar test accuracy to its dense counterpart with much fewer network parameters. Generally, these methods search for the winning tickets on well-labeled data. Unfortunately, in many real-world applications, the training data are unavoidably contaminated with noisy labels, thereby leading to performance deterioration of these methods. To address the above-mentioned problem, we propose a novel two-stream sample selection network (TS3-Net), which consists of a sparse subnetwork and a dense subnetwork, to effectively identify the winning ticket with noisy labels. The training of TS3-Net contains an iterative procedure that switches between training both subnetworks and pruning the smallest magnitude weights of the sparse subnetwork. In particular, we develop a multistage learning framework including a warm-up stage, a semisupervised alternate learning stage, and a label refinement stage, to progressively train the two subnetworks. In this way, the classification capability of the sparse subnetwork can be gradually improved at a high sparsity level. Extensive experimental results on both synthetic and real-world noisy datasets (including MNIST, CIFAR-10, CIFAR-100, ANIMAL-10N, Clothing1M, and WebVision) demonstrate that our proposed method achieves state-of-the-art performance with very small memory consumption for label noise learning. Code is available at https://github.com/Runqing-forMost/TS3-Net/tree/master.

Human emotions contain both basic and compound facial expressions. In many practical scenarios, it is difficult to access all the compound expression categories at one time. In this paper, we investigate comprehensive facial expression recognition (FER) in the class-incremental learning paradigm, where we define well-studied and easily-accessible basic expressions as initial classes and learn new compound expressions incrementally. To alleviate the stability-plasticity dilemma in our incremental task, we propose a novel Relationship-Guided Knowledge Transfer (RGKT) method for class-incremental FER. Specifically, we develop a multi-region feature learning (MFL) module to extract fine-grained features for capturing subtle differences in expressions. Based on the MFL module, we further design a basic expression-oriented knowledge transfer (BET) module and a compound expression-oriented knowledge transfer (CET) module, by effectively exploiting the relationship across expressions. The BET module initializes the new compound expression classifiers based on expression relevance between basic and compound expressions, improving the plasticity of our model to learn new classes. The CET module transfers expression-generic knowledge learned from new compound expressions to enrich the feature set of old expressions, facilitating the stability of our model against forgetting old classes. Extensive experiments on three facial expression databases show that our method achieves superior performance in comparison with several state-of-the-art methods.

Graph convolutional networks (GCN) have recently been studied to exploit the graph topology of the human body for skeleton-based action recognition. However, most of these methods unfortunately aggregate messages via an inflexible pattern for various action samples, lacking the awareness of intra-class variety and the suitableness for skeleton sequences, which often contain redundant or even detrimental connections. In this paper, we propose a novel Deformable Graph Convolutional Network (DeGCN) to adaptively capture the most informative joints. The proposed DeGCN learns the deformable sampling locations on both spatial and temporal graphs, enabling the model to perceive discriminative receptive fields. Notably, considering human action is inherently continuous, the corresponding temporal features are defined in a continuous latent space. Furthermore, we design an innovative multi-branch framework, which not only strikes a better trade-off between accuracy and model size, but also elevates the effect of ensemble between the joint and bone modalities remarkably. Extensive experiments show that our proposed method achieves state-of-the-art performances on three widely used datasets, NTU RGB+D, NTU RGB+D 120, and NW-UCLA.

Knowledge distillation (KD), which aims at transferring the knowledge from a complex network (a teacher) to a simpler and smaller network (a student), has received considerable attention in recent years. Typically, most existing KD methods work on well-labeled data. Unfortunately, real-world data often inevitably involve noisy labels, thus leading to performance deterioration of these methods. In this article, we study a little-explored but important issue, i.e., KD with noisy labels. To this end, we propose a novel KD method, called ambiguity-guided mutual label refinery KD (AML-KD), to train the student model in the presence of noisy labels. Specifically, based on the pretrained teacher model, a two-stage label refinery framework is innovatively introduced to refine labels gradually. In the first stage, we perform label propagation (LP) with small-loss selection guided by the teacher model, improving the learning capability of the student model. In the second stage, we perform mutual LP between the teacher and student models in a mutual-benefit way. During the label refinery, an ambiguity-aware weight estimation (AWE) module is developed to address the problem of ambiguous samples, avoiding overfitting these samples. One distinct advantage of AML-KD is that it is capable of learning a high-accuracy and low-cost student model with label noise. The experimental results on synthetic and real-world noisy datasets show the effectiveness of our AML-KD against state-of-the-art KD methods and label noise learning (LNL) methods. Code is available at https://github.com/Runqing-forMost/ AML-KD.

Person attribute recognition (PAR) aims to simultaneously predict multiple attributes of a person. Existing deep learning-based PAR methods have achieved impressive performance. Unfortunately, these methods usually ignore the fact that different attributes have an imbalance in the number of noisy-labeled samples in the PAR training datasets, thus leading to suboptimal performance. To address the above problem of imbalanced noisy-labeled samples, we propose a novel and effective loss called drop loss for PAR. In the drop loss, the attributes are treated differently in an easy-to-hard way. In particular, the noisy-labeled candidates, which are identified according to their gradient norms, are dropped with a higher drop rate for the harder attribute. Such a manner adaptively alleviates the adverse effect of imbalanced noisy-labeled samples on model learning. To illustrate the effectiveness of the proposed loss, we train a simple ResNet-50 model based on the drop loss and term it DropNet. Experimental results on two representative PAR tasks (including facial attribute recognition and pedestrian attribute recognition) demonstrate that the proposed DropNet achieves comparable or better performance in terms of both balanced accuracy and classification accuracy over several state-of-the-art PAR methods.

Rheumatoid arthritis (RA) is a chronic inflammatory disease. It leads to bone erosion in joints and other complications, which severely affect patients' quality of life. To accurately diagnose and monitor the progression of RA, quantitative imaging and analysis tools are desirable. High-resolution peripheral quantitative computed tomography (HR-pQCT) is such a promising tool for monitoring disease progression in RA. However, automatic erosion detection tools using HR-pQCT images are not yet available. Inspired by the consensus among radiologists on the erosions in HR-pQCT images, in this paper we define erosion as the significant concave regions on the cortical layer, and develop a model-based 3D automatic erosion detection method. It mainly consists of two steps: constructing closed cortical surface, and detecting erosion regions on the surface. In the first step, we propose an initialization-robust region competition methods for joint segmentation, and then fill the surface gaps by using joint bone separation and curvature-based surface alignment. In the second step, we analyze the curvature information of each voxel, and then aggregate the candidate voxels into concave surface regions and use the shape information of the regions to detect the erosions. We perform qualitative assessments of the new method using 59 well-annotated joint volumes. Our method has shown satisfactory and consistent performance compared with the annotations provided by medical experts., Competing Interests: Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2023. Published by Elsevier Ltd.)

Background: Since the coronavirus disease 2019 (COVID-19) has spread throughout the world, many studies on innate immunity in COVID-19 have been published, and great progress has been achieved, while bibliometric analysis on hotspots and research trends in this field remains lacking., Methods: On 17 November 2022, articles and reviews on innate immunity in COVID-19 were recruited from the Web of Science Core Collection (WoSCC) database after papers irrelevant to COVID-19 were further excluded. The number of annual publications and the average citations per paper were analyzed by Microsoft Excel. Bibliometric analysis and visualization of the most prolific contributors and hotspots in the field were performed by VOSviewer and CiteSpace software., Results: There were 1,280 publications that met the search strategy on innate immunity in COVID-19 and were published from 1 January 2020 to 31 October 2022. Nine hundred thirteen articles and reviews were included in the final analysis. The USA had the highest number of publications (Np) at 276 and number of citations without self-citations (Nc) at 7,085, as well as an H-index of 42, which contributed 30.23% of the total publications, followed by China (Np: 135, Nc: 4,798, and H-index: 23) with 14.79% contribution. Regarding Np for authors, Netea, Mihai G. (Np: 7) from the Netherlands was the most productive author, followed by Joosten, Leo A. B. (Np: 6) and Lu, Kuo-Cheng (Np: 6). The Udice French Research Universities had the most publications (Np: 31, Nc: 2,071, H-index: 13), with an average citation number (ACN) at 67. The journal Frontiers in Immunology possessed the most publications (Np: 89, Nc: 1,097, ACN: 12.52). "Evasion" (strength 1.76, 2021-2022), "neutralizing antibody" (strength 1.76, 2021-2022), "messenger RNA" (strength 1.76, 2021-2022), "mitochondrial DNA" (strength 1.51, 2021-2022), "respiratory infection" (strength 1.51, 2021-2022), and "toll-like receptors" (strength 1.51, 2021-2022) were the emerging keywords in this field., Conclusion: The study on innate immunity in COVID-19 is a hot topic. The USA was the most productive and influential country in this field, followed by China. The journal with the most publications was Frontiers in Immunology . "Messenger RNA," "mitochondrial DNA," and "toll-like receptors" are the current hotspots and potential targets in future research., Competing Interests: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2023 Lai, Xu, Xue, Zhang, Zhong, Liao.)

Objective: We sought to examine the independent correlates of long-term hospitalization in a sample of Chinese inpatients with schizophrenia (SCZ) from a gender-based perspective., Methods: This was a cross-sectional study that was carried out in a tertiary psychiatric hospital. All adult inpatients in this hospital were screened from January to March 2020, 251 of whom were identified as long-stay inpatients with SCZ (LSIS) and 224 as short-stay inpatients with SCZ (SSIS). Demographic and clinical information of the two groups was collected through medical records, scale assessments and interviews. Gender differences were analyzed, and independent correlates of long-stay between genders were explored by logistic regression analyses., Results: Compared to SSIS, greater proportions of LSIS patients were male (64.1%), single (82.1%), unemployed (81.7%) and had no family caregivers (54.2%). For LSIS per se, proportionally more males were single (88.8%), had no family caregiver (65.8%), had concomitant physical disease (65.2%) and had a history of hazardous behavior (27.3%) than their female counterparts. For females, the top independent risk factors for a long stay included poor functioning ( OR  = 5.9, 95% CI : 2.9-12.0), older age ( OR  = 4.3, 95% CI : 2.1-9.1) and being single ( OR  = 3.9, 95% CI : 1.8-8.4). Similar to women, both older age ( OR  = 5.3, 95% CI : 2.5-11.2) and poor functioning ( OR  = 4.0, 95% CI : 2.1-7.9) were also independent factors for long-term hospitalization of male patients; however, having no family caregiver ( OR  = 10.2, 95% CI : 4.6-22.6) was the primary risk factor for men., Conclusions: Both clinical and nonclinical factors play important roles in long-term hospitalization in Chinese patients with schizophrenia. There are overlaps and distinctions across genders with respect to the independent factors of long stays. These findings provide clues for developing better service strategies for this population, and highlight the importance of paying attention to gender differences in further research in this field., Competing Interests: The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (© 2023 The Authors.)

Electrophilic addition to alkenes is a textbook-taught reaction, yet it is not always possible to control the regioselectivity of addition to unsymmetrical 1,2-disubstituted substrates. We report the observation and applications of the β-boron effect that accounts for high regioselectivity in electrophilic addition reactions to allylic MIDA ( N -methyliminodiacetic acid) boronates. While the well-established β-silicon effect bears partial resemblance to the observed reactivity, the silyl group is typically lost during functionalization. In contrast, the boryl moiety is retained in the product when B(MIDA) is used as the nucleophilic stabilizer. Mechanistic studies elucidate the origin of this effect and demonstrate how σ(C-B) hyperconjugation helps stabilize the incipient carbocation. This transformation represents a rare example of the stereospecific hydrohalogenation of secondary allyl MIDA-boronates that proceeds in a syn-fashion.

As a typical non-Gaussian vector variable, a neutral vector variable contains nonnegative elements only, and its l 1 -norm equals one. In addition, its neutral properties make it significantly different from the commonly studied vector variables (e.g., the Gaussian vector variables). Due to the aforementioned properties, the conventionally applied linear transformation approaches [e.g., principal component analysis (PCA) and independent component analysis (ICA)] are not suitable for neutral vector variables, as PCA cannot transform a neutral vector variable, which is highly negatively correlated, into a set of mutually independent scalar variables and ICA cannot preserve the bounded property after transformation. In recent work, we proposed an efficient nonlinear transformation approach, i.e., the parallel nonlinear transformation (PNT), for decorrelating neutral vector variables. In this article, we extensively compare PNT with PCA and ICA through both theoretical analysis and experimental evaluations. The results of our investigations demonstrate the superiority of PNT for decorrelating the neutral vector variables.

GAN inversion aims to invert a given image back into the latent space of a pretrained GAN model so that the image can be faithfully reconstructed from the inverted code by the generator. As an emerging technique to bridge the real and fake image domains, GAN inversion plays an essential role in enabling pretrained GAN models, such as StyleGAN and BigGAN, for applications of real image editing. Moreover, GAN inversion interprets GAN's latent space and examines how realistic images can be generated. In this paper, we provide a survey of GAN inversion with a focus on its representative algorithms and its applications in image restoration and image manipulation. We further discuss the trends and challenges for future research. A curated list of GAN inversion methods, datasets, and other related information can be found at https://github.com/weihaox/awesome-gan-inversion.

Background: Exercise training significantly improves cardiorespiratory fitness (CRF) in heart failure with reduced ejection fraction (HFrEF) patients, but high-intensity interval training (HIIT) is not superior to moderate-intensity interval training (MIIT). Whether HIIT is more beneficial than MIIT in patients with heart failure with preserved ejection fraction (HFpEF) remains unclear., Methods: On August 29, 2021, we conducted a comprehensive computerized literature search of the Medline, EMBASE, Web of Science, and Cochrane databases using the following keywords: "HF or diastolic HF or HFpEF or HF with normal ejection fraction and exercise training or aerobic exercise or isometric exercises or physical activity or cardiac rehabilitation." Only randomized controlled trials (RCTs) reporting comparisons between HIIT and MIIT in HFpEF were included in the final analysis to maintain consistency and obtain robust pooled estimates. Methodological quality was assessed based on the ratings of individual biases. To generate an overall test statistic, the data were analyzed using the random-effects model for a generic inverse variance. Outcome measures were reported as an odds ratio, and confidence intervals (CIs) were set at 95%. The study followed PRISMA guidelines., Results: This meta-analysis included only RCTs comparing the efficacy of HIIT and MIIT in HFpEF patients. This study included 150 patients from 3 RCTs. In the current pooled data analysis, HIIT significantly improves diastolic function measured by E/A ratio (WMD, 0.13; 95% CI, 0.03-0.23, P = .009). However, no significant change was observed in the diastolic function measured by E/e' ratio (WMD, 0.39; 95% CI, -2.40 to 3.18, P = .78), and CRF evaluated by both VO2 (mL/kg per min; WMD, -0.86; 95%CI, -5.27 to 3.55, P = .70) and VE/CO2 slope (WMD, 0.15; 95% CI, -10.24 to 10.53, P = .98), and systolic function (EF-WMD, -2.39; 95% CI, -12.16% to 7.38%, P = .63) between HIIT and MIIT in patients with HFpEF., Conclusion: In HFpEF patients, HIIT may be superior to MIIT in improving diastolic function, measured by E/A, but not CRF and left ventricular systolic function., Competing Interests: The authors have no conflicts of interest to disclose., (Copyright © 2023 the Author(s). Published by Wolters Kluwer Health, Inc.)

The primary amino group has been seldom utilized as a transformable functionality in organic synthesis. Reported herein is a deaminative halogenation of primary amines using N -anomeric amide as the nitrogen-deletion reagent. Both aliphatic and aromatic amines are competent substrates for direct halogenations. The mildness and robustness of the protocol are evidenced by the successful reactions of several complex- and functional group-enriched bioactive compounds or drugs. Elaboration of the resulting products provides interesting analogues of drug molecules., Competing Interests: The authors declare no competing interests., (© 2023 The Author(s).)

Two new α -pyrones, micropyrones A ( 1 ) and B ( 2 ), along with four known γ -pyrones, nocapyrone D ( 3 ), nocapyrone A ( 4 ), marinactinone A ( 5 ), and nocapyrone H ( 6 ), were isolated from the culture extract of actinomycete Microbacterium sp. GJ312, which was isolated from Glycyrrhiza uralensis . The structures of these compounds were identified by analysis of spectral data. They are the first α - and γ -pyrones reported from the genus Microbacterium . The antibacterial activity of all compounds against Staphylococcus aureus and methicillin resistant S. aureus was evaluated. However, none of them showed significant activity. This study represents the first phytochemical example of a Glycyrrhiza -derived actinomycete.

In this paper, we reveal the discriminant capacity of orthogonal data projection onto the generalized difference subspace (GDS), both theoretically and experimentally. In our previous work, we demonstrated that the GDS projection works as a quasi-orthogonalization of class subspaces, which is an effective feature extraction for subspace based classifiers. Here, we further show that GDS projection also works as a discriminant feature extraction through a similar mechanism to the Fisher discriminant analysis (FDA). A direct proof of the connection between GDS projection and FDA is difficult due to the significant difference in their formulations. To circumvent the complication, we first introduce geometrical Fisher discriminant analysis (gFDA) based on a simplified Fisher criterion. It is derived from a heuristic yet practically plausible assumption: the direction of the sample mean vector of a class is largely aligned to the first principal component vector of the class, given that the principal component analysis (PCA) is applied without data centering. gFDA works stably even under few samples, bypassing the small sample size (SSS) problem of FDA. We then prove that gFDA is equivalent to GDS projection with a small correction term. This equivalence ensures GDS projection to inherit the discriminant ability from FDA via gFDA. Furthermore, we discuss two useful extensions of these methods, 1) a nonlinear extension by kernel trick, 2) a combination with CNN features. The equivalence and the effectiveness of the extensions have been verified through extensive experiments on the extended Yale B+, CMU face database, ALOI, ETH80, MNIST, and CIFAR10, mainly focusing on image recognition under small samples.

Noise annoyance has been often reported as one of the main adverse effects of noise exposure on human health, and there is consensus that it relates to several factors going beyond the mere energy content of the signal. Research has historically focused on a limited set of sound sources (e.g., transport and industrial noise); only more recently is attention being given to more holistic aspects of urban acoustic environments and the role they play in the noise annoyance perceptual construct. This is the main approach promoted in soundscape studies, looking at both wanted and unwanted sounds. In this study, three specific aspects were investigated, namely: (1) the effect of different sound sources combinations, (2) the number of sound sources present in the soundscape, and (3) the presence of individual sound source, on noise annoyance perception. For this purpose, a large-scale online experiment was carried out with 1.2k+ participants, using 2.8k+ audio recordings of complex urban acoustic environments to investigate how they would influence the perceived noise annoyance. Results showed that: (1) the combinations of different sound sources were not important, compared, instead, to the number of sound sources identified in the soundscape recording (regardless of sound sources type); (2) the annoyance ratings expressed a minimum when any two clearly distinguishable sound sources were present in a given urban soundscape; and (3) the presence (either in isolation or combination) of traffic-related sound sources increases noise annoyance, while the presence (either in isolation or combination) of nature-related sound sources decreases noise annoyance.

Objective: To explore the trajectory of disability in the dying elderly in China. Methods: Based on the activity of daily living (ADL) data from the 2002-2018 Chinese Longitudinal Healthy Longevity Survey, the longitudinal item response theory (LIRT) model was fitted with the difficulty threshold parameters to analyze the ADL loss in the elderly in China. Then, a mixed-effects model was fitted to analyze the trajectory of the disability level of the dying elderly. Results: A total of 5 817 old adults who entered the cohort in 2002 were included, in whom 41.81% were males, with a baseline age of (86.80±12.40) years and a follow-up time of 4 (3,8) years. The results of LIRT showed that the lowest difficulty threshold parameter in the basic activity of daily living (BADL) was partially disability on bathing (0.41±0.05), and the highest was entirely disability on indoor movement (6.19±0.16). In comparison, the lowest difficulty threshold parameter in instrumental activity of daily living (IADL) was partially disability on using public transportation (-3.01±0.07), and the highest was entirely disability on visiting neighbors (1.51±0.07). In the trajectory of disability, the average dependency in ADL was lower in dying men than in dying women ( P <0.001), in the elderly living alone than in the elderly living with family members ( P <0.001) and in the non-illiterate elderly than in the illiterate elderly ( P <0.001). The estimated value of both the linear change rate and quadratic coefficient of disability level development with time were 0.231 ( P <0.001) and 0.002 ( P <0.001). Conclusions: In China, the development of disability in the elderly in China has its characteristics, IADL disability might occurs earlier than BADL. Among the IADL/BADL items, the disability of lower limb-based items is more prone to occur compared with upper limb-based items, and the disability of complex items is more prone to occur compared with simple items, and the growth rate of the disability trajectory also accelerates over time. It is necessary to pay attention to old women, old people living with family members, old people with low education level and old people with poor cognitive function in the disability prevention.

Purpose: The accuracy of level of anti- severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) antibodies is a great concern. We aimed to compare the efficacy of anti-SARS-CoV-2 antibody detection kits from two manufacturers in evaluating the efficacy of SARS-CoV-2 vaccines., Methods: The immune responses and consistency of four anti-SARS-CoV-2 antibodies were evaluated using two manufacturers' antibody kits (A and B) in 61 subjects within 160 days after vaccination with the CoronaVac vaccine., Results: The total seropositivity rates of neutralizing antibodies and IgM antibodies detected by kit A were higher than those detected by kit B (P = 0.003 and P < 0.001, respectively). Conversely, the total seropositivity rates of total antibodies and IgG antibodies were higher in kit B than kit A (P < 0.001 and P < 0.001, respectively). The consistency rates showed less than 90% agreement between the kits for the detection of the four antibodies, and the κ score showed moderate or substantial consistency. The half-lives of neutralizing antibodies, total antibodies, and IgG antibodies within 160 days after vaccination, detected by kit A were 63.88 days, 80.50 days, and 63.70 days, respectively and by kit B were 97.06 days, 65.41 days, and 77.99 days, respectively., Conclusion: The efficacy of antibody detection differed between the two commercial anti-SARS-CoV-2 antibody kits, although there was moderate consistency, which may affect the clinical application and formulation of the vaccine strategy., Competing Interests: Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2022 Elsevier B.V. All rights reserved.)

Due to lack of data, overfitting ubiquitously exists in real-world applications of deep neural networks (DNNs). We propose advanced dropout, a model-free methodology, to mitigate overfitting and improve the performance of DNNs. The advanced dropout technique applies a model-free and easily implemented distribution with parametric prior, and adaptively adjusts dropout rate. Specifically, the distribution parameters are optimized by stochastic gradient variational Bayes in order to carry out an end-to-end training. We evaluate the effectiveness of the advanced dropout against nine dropout techniques on seven computer vision datasets (five small-scale datasets and two large-scale datasets) with various base models. The advanced dropout outperforms all the referred techniques on all the datasets. We further compare the effectiveness ratios and find that advanced dropout achieves the highest one on most cases. Next, we conduct a set of analysis of dropout rate characteristics, including convergence of the adaptive dropout rate, the learned distributions of dropout masks, and a comparison with dropout rate generation without an explicit distribution. In addition, the ability of overfitting prevention is evaluated and confirmed. Finally, we extend the application of the advanced dropout to uncertainty inference, network pruning, text classification, and regression. The proposed advanced dropout is also superior to the corresponding referred methods. Codes are available at https://github.com/PRIS-CV/AdvancedDropout.

Members of the genus Marinomonas are known for their environmental adaptation and metabolically versatility, with abundant proteins associated with antifreeze, osmotic pressure resistance, carbohydrase and multiple secondary metabolites. Comparative genomic analysis focusing on secondary metabolites and orthologue proteins was conducted with 30 reference genome sequences in the genus Marinomonas. In this study, a Gram-stain-negative, rod-shaped, non-flagellated and strictly aerobic bacterium, designated as strain E8 T , was isolated from the red algae (Gelidium amansii) in the coastal of Weihai, China. Optimal growth of the strain E8 T was observed at temperatures 25-30 °C, pH 6.5-8.0 and 1-3% (w/v) NaCl. The DNA G + C content was 42.8 mol%. The predominant isoprenoid quinone was Q-8 and the major fatty acids were C 16:0 , summed feature 3 and summed feature 8. The major polar lipids were phosphatidylglycerol (PG) and phosphatidylethanolamine (PE). Based on data obtained from this polyphasic taxonomic study, strain E8 T should be considered as a novel species of the genus Marinomonas, for which the name Marinomonas algarum is proposed. The type strain is E8 T (= KCTC 92201 T  = MCCC 1K07070 T )., (© 2022. The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature.)

Background: Sacubitril/valsartan has been approved for the treatment of heart failure (HF) patients with reduced ejection fraction; since then, it gradually became a new star drug in the therapy of HF. Nevertheless, the effectiveness of sacubitril/valsartan remains under investigation. Thus far, only a few bibliometric studies have systematically analyzed the application of sacubitril/valsartan., Methods: Publications on sacubitril/valsartan were retrieved from the Web of Science Core Collection on April 29, 2021. Data were analyzed using Microsoft Excel 2019 (Redmond, WA), VOS viewer (Redmond, WA), and Cite Space V (Drexel University, Philadelphia, PA)., Results: A total of 1309 publications on sacubitril/valsartan published from 1995 to 2021 were retrieved. The number of publications regarding sacubitril/valsartan increased sharply in the last 6 years (2015-2021), and American scholars authored >40% of those publications. Most were published in the European Journal of Heart Failure, the United States was the bellwether with a solid academic reputation in this area. Solomon published the highest number of related articles and was the most frequently cited author. "Heart failure" was the leading research hotspot. The keywords, "inflammation," "fibrosis," and "oxidative stress" appeared most recently as research fronts., Conclusions: Research attention should be focused on clinical trial outcomes. Considering its effectiveness in HF, the mechanisms and further applications of sacubitril/valsartan may become research hotspots in the future and should be closely examined., Competing Interests: The authors have no conflicts of interest to disclose., (Copyright © 2022 the Author(s). Published by Wolters Kluwer Health, Inc.)

Object detection has made enormous progress and has been widely used in many applications. However, it performs poorly when only limited training data is available for novel classes that the model has never seen before. Most existing approaches solve few-shot detection tasks implicitly without directly modeling the detectors for novel classes. In this article, we propose GenDet, a new meta-learning-based framework that can effectively generate object detectors for novel classes from few shots and, thus, conducts few-shot detection tasks explicitly. The detector generator is trained by numerous few-shot detection tasks sampled from base classes each with sufficient samples, and thus, it is expected to generalize well on novel classes. An adaptive pooling module is further introduced to suppress distracting samples and aggregate the detectors generated from multiple shots. Moreover, we propose to train a reference detector for each base class in the conventional way, with which to guide the training of the detector generator. The reference detectors and the detector generator can be trained simultaneously. Finally, the generated detectors of different classes are encouraged to be orthogonal to each other for better generalization. The proposed approach is extensively evaluated on the ImageNet, VOC, and COCO data sets under various few-shot detection settings, and it achieves new state-of-the-art results.

Context.—: Neutralizing antibody detection can assess the incidence of COVID-19 and the effectiveness of vaccines. However, commercial reagents for neutralizing antibodies were developed after the anti-SARS-CoV-2 immunoglobulin (Ig) G and IgM antibodies. Therefore, some laboratories did not perform neutralizing antibody testing services because of multiple factors., Objective.—: To find a fast, accurate, and economic alternative for the detection of neutralizing antibodies for the development of COVID-19 screening programs., Design.—: The response and correlation of 3 antibodies (anti-spike protein neutralizing antibody, total anti-receptor-binding domain [RBD] antibody, and anti-RBD IgG) were determined by observing the dynamics in 61 participants for 160 days after vaccination., Results.—: The levels of neutralizing and anti-RBD IgG antibodies reached their peak values on day 42 after vaccination (120.75 IU/mL and 14.38 signal-to-cutoff ratio [S/CO], respectively). The total antibody levels peaked at 138.47 S/CO on day 35 after vaccination. The strongest correlation was found between neutralizing and anti-RBD IgG antibody levels (r = 0.894, P < .001). The area under the receiver operating characteristic curve for total antibody levels for the prediction of seropositivity for neutralizing antibodies was 0.881 (P < .001), and that for anti-RBD IgG antibody levels was 0.937 (P < .001)., Conclusions.—: Neutralizing and anti-RBD IgG antibody levels were strongly correlated, and thus anti-RBD IgG antibody levels can be used for the accurate assessment of immunity following SARS-CoV-2 infection or vaccination.

Background: Determining what quarantine period and detection strategy are more effective and sustainable remains a challenge for further prevention and social stability., Methods: From October 2020 to December 2021, 290,547 inbound overseas travelers were subject to government quarantine in Xiamen, China. The detection rate of COVID-19 during different quarantine periods using dual or single nucleic acid testing reagents., Results: The COVID-19 positive rate was 1.79% (519/290,547). The detection rates during the 7-day, 14-day and 14+7-day quarantine periods using the dual reagents were 78.4%, 91.7%, and 100%, respectively. The detection rate of the 7-day, 14-day and 14+7-day quarantine periods were 73.99%, 86.51%, and 94.22%, respectively, using the Liferiver reagent and 72.25%, 84.59%, and 91.91%, respectively, using the Daan reagent. Based on the 14+7 day strategy, dual nucleic acid testing reagent strategy detected all imported cases, but 30 cases (5.78%) were not detected via Liferiver reagent and 42 (8.09%) cases not detected via Daan reagent., Conclusion: A 14+7-day quarantine period and dual nucleic acid testing reagent strategy are effective screening methods for COVID-19 among inbound overseas travelers. The superior detection rate of these strategies reduce the risk of secondary transmission of the SARS-CoV-2 virus., (Copyright © 2022 Elsevier B.V. All rights reserved.)

Background: The timely, rapid, and accurate near real-time observations are urgent to monitor the damage of corn armyworm, because the rapid expansion of armyworm would lead to severe yield losses. Therefore, the potential of machine learning algorithms for identifying the armyworm infected areas automatically and accurately by multispectral unmanned aerial vehicle (UAV) dataset is explored in this study. The study area is in Beicuizhuang Village, Langfang City, Hebei Province, which is the main corn-producing area in the North China Plain., Results: Firstly, we identified the optimal combination of image features by Gini-importance and the comparation of four kinds of machine learning methods including Random Forest (RF), Multilayer Perceptron (MLP), Naive Bayesian (NB) and Support Vector Machine (SVM) was done. And RF was proved to be the most potential with the highest Kappa and OA of 0.9709 and 0.9850, respectively. Secondly, the armyworm infected areas and healthy corn areas were predicted by an optimized RF model in the UAV dataset, and the armyworm incidence levels were classified subsequently. Thirdly, the relationship between the spectral characteristics of different bands and pest incidence levels within the Sentinel-2 and UAV images were analyzed, and the B3 in UAV images and the B6 in Sentinel-2 image were less sensitive for armyworm incidence levels. Therefore, the Sentinel-2 image was used to monitor armyworm in two towns., Conclusions: The optimized dataset and RF model are effective and reliable, which can be used for identifying the corn damage by armyworm using UAV images accurately and automatically in field-scale. © 2022 Society of Chemical Industry., (© 2022 Society of Chemical Industry.)

Background & Aims: Most patients with gastric cancer (GCa) are diagnosed at an advanced stage. We aimed to investigate novel fecal signatures for clinical application in early diagnosis of GCa., Methods: This was an observational study that included 1043 patients from 10 hospitals in China. In the discovery cohort, 16S ribosomal RNA gene analysis was performed in paired samples (tissues and feces) from patients with GCa and chronic gastritis (ChG) to determine differential abundant microbes. Their relative abundances were detected using quantitative real-time polymerase chain reaction to test them as bacterial candidates in the training cohort. Their diagnostic efficacy was validated in the validation cohort., Results: Significant enrichments of Streptococcus anginosus (Sa) and Streptococcus constellatus (Sc) in GCa tumor tissues (P < .05) and feces (P < .0001) were observed in patients with intraepithelial neoplasia, early and advanced GCa. Either the signature parallel test Sa∪Sc or single signature Sa/Sc demonstrated superior sensitivity (Sa: 75.6% vs 72.1%, P < .05; Sc: 84.4% vs 64.0%, P < .001; and Sa∪Sc: 91.1% vs 81.4%, P < .01) in detecting early GCa compared with advanced GCa (specificity: Sa: 84.0% vs 83.9%, Sc: 70.4% vs 82.3%, and Sa∪Sc: 64.0% vs 73.4%). Fecal signature Sa∪Sc outperformed Sa∪CEA/Sc∪CEA in the discrimination of advanced GCa (sensitivity: 81.4% vs 74.2% and 81.4% vs 72.3%, P < .01; specificity: 73.4% vs 81.0 % and 73.4% vs 81.0%). The performance of Sa∪Sc in the diagnosis of both early and advanced GCa was verified in the validation cohort., Conclusion: Fecal Sa and Sc are noninvasive, accurate, and sensitive signatures for early warning in GCa. (ClinicalTrials.gov, Number: NCT04638959)., (Copyright © 2022 AGA Institute. Published by Elsevier Inc. All rights reserved.)

A Gram-stain-negative, strictly aerobic, non-flagellated, oxidase- and catalase-positive, rod-shaped marine bacterium, designated strain DM8 T , was isolated from the intestine of Trichiurus japonicus in Weihai, China. The strain optimally grew at 25-35℃, with 1.0-4.0% (w/v) NaCl and at pH 7.0-8.0. Its colonies were circular, slightly yellow, non-transparent, smooth, and approximately 0.8-1.5 mm in diameter, after being cultured for 48 h on marine agar 2216. Based on the result of phylogenetic analysis of 16S rRNA gene sequence, strain DM8 T had close relationship with Oceanisphaera profunda SM1222 T (96.9%) and the type strain DSM 15406  T of the type species Oceanisphaera litoralis (94.7%), respectively. Genome sequencing revealed a genome size of 3,109,059 bp and a G + C content of 46.9 mol%. It had Q-8 as the sole respiratory quinone and possessed C 16:0 , summed features 3 (C 16:1 ω7c/C 16:1 ω6c) and summed features 8 (C 18:1 ω7c/C 18:1 ω6c) as major fatty acids. The major polar lipid profile was composed of phosphatidylglycerol and phosphatidylethanolamine. Based on the phenotypic, chemotaxonomic characterizations, phylogenetic properties and genome analysis, strain DM8 T should represent a novel species of the genus Oceanisphaera, for which the name Oceanisphaera pacifica sp. nov. is proposed. The type strain is DM8 T (= KCTC 82764  T  = MCCC 1K06133 T )., (© 2022. The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature.)

We propose a polarization-based probabilistic discriminative model for deriving a set of new sigmoid-transformed polarimetry feature parameters, which not only enables accurate and quantitative characterization of cancer cells at pixel level, but also accomplish the task with a simple and stable model. By taking advantages of polarization imaging techniques, these parameters enable a low-magnification and wide-field imaging system to separate the types of cells into more specific categories that previously were distinctive under high magnification. Instead of blindly choosing the model, the L0 regularization method is used to obtain the simplified and stable polarimetry feature parameter. We demonstrate the model viability by using the pathological tissues of breast cancer and liver cancer, in each of which there are two derived parameters that can characterize the cells and cancer cells respectively with satisfactory accuracy and sensitivity. The stability of the final model opens the possibility for physical interpretation and analysis. This technique may bypass the typically labor-intensive and subjective tumor evaluating system, and could be used as a blueprint for an objective and automated procedure for cancer cell screening., Competing Interests: The authors declare no conflicts of interest., (© 2022 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement.)

Background: Neuromyelitis Optica spectrum disorder (NMOSD) is severe relapsing and disabling autoimmune disease of the central nervous system. Its optimal first-line treatment to reduce relapse rate and ameliorate neurological disability remains unclear. We will conduct a prospective, multicenter, randomized, placebo-controlled clinical trial to study the safety and effectiveness of human umbilical cord mesenchymal stem cells (hUC-MSCs) in treating NMOSD., Methods: The trial is planned to recruit 430 AQP4-IgG seropositive NMOSD patients. It consists of three consecutive stages. The first stage will be carried out in the leading center only and aims to evaluate the safety of hUC-MSCs. Patients will be treated with three different doses of hUC-MSCs: 1, 2, or 5 × 10 6 MSC/kg·weight for the low-, medium-, and high-dose group, respectively. The second and third stages will be carried out in six centers. The second stage aims to find the optimal dosage. Patients will be 1:1:1:1 randomized into the low-, medium-, high-dose group and the controlled group. The third stage aims to evaluate the effectiveness. Patients will be 1:1 randomized into the optimal dose and the controlled group. The primary endpoint is the first recurrent time and secondary endpoints are the recurrent times, EDSS scores, MRI lesion numbers, OSIS scores, Hauser walking index, and SF-36 scores. Endpoint events and side effects will be evaluated every 3 months for 2 years., Discussion: Although hUC-MSC has shown promising treatment effects of NMOSD in preclinical studies, there is still a lack of well-designed clinical trials to evaluate the safety and effectiveness of hUC-MSC among NMOSD patients. As far as we know, this trial will be the first one to systematically demonstrate the clinical safety and efficacy of hUC-MSC in treating NMOSD and might be able to determine the optimal dose of hUC-MSC for NMOSD patients., Trial Registration: The study was registered with the Chinese Clinical Trial Registry (CHICTR.org.cn) on 2 March 2016 (registration No. ChiCTR-INR-16008037), and the revised trial protocol (Protocol version 1.2.1) was released on 16 March 2020., Competing Interests: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2022 Yao, Xie, Cai, Zhang, Deng, Gao, Wang, Xu, Ding, Wu, Zhao, Wang, Song, Wang, Xie, Li, Wan, Lin, Jin, Wang, Qiu, Zhuang, Zhou, Jin, Ni, Yan, Guo, Xue, Qian and Guan.)

Mitogen-activated protein kinase kinase (MAPKK) was the hub component of the Mitogen-activated protein kinase (MAPK) signaling pathway and played an important role in the cellular response to environmental stress. In this study, we identified five MmMAPKK genes in hard clam Mercenaria mercenaria and found that all MmMAPKK genes contain a conserved protein kinase domain. The MmMAPKK genes derived from dispersed duplication were unevenly distributed in three chromosomes. Although the genome size was highly variable among different bivalve mollusks, the number of MAPKK genes was relatively stable. Phylogenetic analysis showed that bivalve MAPKK was divided into five clades, and amino acid sequences of MAPKK from the same clade consisted of similar conserved motifs. The syntenic analysis demonstrated that MmMAPKKs had the highest number of homologous gene pairs with Cyclina sinensis. MmMAPKKs were ubiquitously expressed in all examined tissues, and all MmMAPKK genes were highly expressed in the ovary. MmMAPKK genes showed stress-specific expression under envirionmental stress. MmMAPKK7 showed an upregulated in heat and heat plus hypoxia stress while MmMAPKK1 showed an upregulated in hypoxic stress groups. Dynamic changes of MmMAPKK7, MmMAPKK6 and MmMAPKK1 in hemocytes were observed in response to air exposure. MmMAPKK4 significantly downregulated after air exposure for five days. MmMAPKK7 and MmMAPKK6 might participate in adaptation to low salinity stress. Our results provided useful information about MAPKK and laid a foundation for further studies on MAPKK evolution in the bivalve., (Copyright © 2022. Published by Elsevier Ltd.)

The success of deep convolutional networks (ConvNets) generally relies on a massive amount of well-labeled data, which is labor-intensive and time-consuming to collect and annotate in many scenarios. To eliminate such limitation, self-supervised learning (SSL) is recently proposed. Specifically, by solving a pre-designed proxy task, SSL is capable of capturing general-purpose features without requiring human supervision. Existing efforts focus obsessively on designing a particular proxy task but ignore the semanticity of samples that are advantageous to downstream tasks, resulting in the inherent limitation that the learned features are specific to the proxy task, namely the proxy task-specificity of features. In this work, to improve the generalizability of features learned by existing SSL methods, we present a novel self-supervised framework SSL++ to incorporate the proxy task-independent semanticity of samples into the representation learning process. Technically, SSL++ aims to leverage the complementarity, between the low-level generic features learned by a proxy task and the high-level semantic features newly learned by the generated semantic pseudo-labels, to mitigate the task-specificity and improve the generalizability of features. Extensive experiments show that SSL++ performs favorably against the state-of-the-art approaches on the established and latest SSL benchmarks.

Different from the object motion blur, the defocus blur is caused by the limitation of the cameras' depth of field. The defocus amount can be characterized by the parameter of point spread function and thus forms a defocus map. In this paper, we propose a new network architecture called Defocus Image Deblurring Auxiliary Learning Net (DID-ANet), which is specifically designed for single image defocus deblurring by using defocus map estimation as auxiliary task to improve the deblurring result. To facilitate the training of the network, we build a novel and large-scale dataset for single image defocus deblurring, which contains the defocus images, the defocus maps and the all-sharp images. To the best of our knowledge, the new dataset is the first large-scale defocus deblurring dataset for training deep networks. Moreover, the experimental results demonstrate that the proposed DID-ANet outperforms the state-of-the-art methods for both tasks of defocus image deblurring and defocus map estimation, both quantitatively and qualitatively. The dataset, code, and model is available on GitHub: https://github.com/xytmhy/DID-ANet-Defocus-Deblurring.

Background: A vaccine against coronavirus disease 2019 (COVID-19) with highly effective protection is urgently needed. The anti-severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) antibody response and duration after vaccination are crucial predictive indicators., Objectives: To evaluate the response and duration for 5 subsets of anti-SARS-CoV-2 antibodies after vaccination and their predictive value for protection., Methods: We determined the response and duration for 5 subsets of anti-SARS-CoV-2 antibodies (neutralizing antibody, anti-RBD total antibody, anti-Spike IgG, anti-Spike IgM, and anti-Spike IgA) in 61 volunteers within 160 days after the CoronaVac vaccine. A logistic regression model was used to determine the predictors of the persistence of neutralizing antibody persistence., Results: The seropositivity rates of neutralizing antibody, anti-RBD total antibody, anti-Spike IgG, anti-Spike IgM, and anti-Spike IgA were only 4.92%, 27.87%, 21.31%, 3.28% and 0.00%, respectively, at the end of the first dose (28 days). After the second dose, the seropositivity rates reached peaks of 95.08%, 100.00%, 100.00%, 59.02% and 31.15% in two weeks (42 days). Their decay was obvious and the seropositivity rate remained at 19.67%, 54.10%, 50.82%, 3.28% and 0.00% on day 160, respectively. The level of neutralizing antibody reached a peak of 149.40 (101.00-244.60) IU/mL two weeks after the second dose (42 days) and dropped to 14.23 (7.62-30.73) IU/mL at 160 days, with a half-life of 35.61(95% CI, 32.68 to 39.12) days. Younger participants (≤31 years) had 6.179 times more persistent neutralizing antibodies than older participants (>31 years) ( P <0.05). Participants with anti-Spike IgA seropositivity had 4.314 times greater persistence of neutralizing antibodies than participants without anti-Spike IgA seroconversion ( P <0.05)., Conclusions: Antibody response for the CoronaVac vaccine was intense and comprehensive with 95.08% neutralizing seropositivity rate, while decay was also obvious after 160 days. Therefore, booster doses should be considered in the vaccine strategies., Competing Interests: Authors Z-JJ, MJ-W, and Y-YL are employed by Xiamen Boson Biotech Co., Ltd., and author W-LL is employed by Autobio Diagnostic Co., Ltd. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2021 Xu, Xue, Xiao, Jia, Wu, Liu, Li, Liang and Yang.)

Polarization images encode high resolution microstructural information even at low resolution. We propose a framework combining polarization imaging and traditional microscopy imaging, constructing a dual-modality machine learning framework that is not only accurate but also generalizable and interpretable. We demonstrate the viability of our proposed framework using the cervical intraepithelial neoplasia grading task, providing a polarimetry feature parameter to quantitatively characterize microstructural variations with lesion progression in hematoxylin-eosin-stained pathological sections of cervical precancerous tissues. By taking advantages of polarization imaging techniques and machine learning methods, the model enables interpretable and quantitative diagnosis of cervical precancerous lesion cases with improved sensitivity and accuracy in a low-resolution and wide-field system. The proposed framework applies routine image-analysis technology to identify the macro-structure and segment the target region in H&E-stained pathological images, and then employs emerging polarization method to extract the micro-structure information of the target region, which intends to expand the boundary of the current image-heavy digital pathology, bringing new possibilities for quantitative medical diagnosis.

Background and Aim: Fusobacterium nucleatum is increasingly being recognized as an important risk factor in colorectal cancer and colorectal adenoma. Endoscopic polypectomy is associated with a decreased incidence of colorectal cancer; however, patients still suffer from a risk of metachronous adenoma. Currently, there are few effective non-invasive factors that may predict metachronous colorectal adenoma. Here, we evaluated the performance of F. nucleatum in predicting metachronous adenoma., Methods: Fecal samples and clinical information of patients before endoscopic polypectomy were collected from 367 patients in a retrospective cohort, and 238 patients in a prospective cohort. The abundance of fecal F. nucleatum was measured via quantitative polymerase chain reaction. Surveillance colonoscopies were conducted between 1 and 3 years after polypectomy (average follow-up 27.07 months for the retrospective cohort & 22.57 months for the prospective cohort) to identify metachronous adenoma. Candidate predictive factors and cut-off value of F. nucleatum abundance were identified from the retrospective cohort and then validated in the prospective cohort., Results: A high abundance of fecal F. nucleatum was found to be an independent risk factor for metachronous adenomas (odds ratio, 6.38; P < 0.001) in the retrospective cohort and was validated in the prospective cohort with a specificity of 65.00%, and a sensitivity of 73.04%, and an overall performance with the area under the curve of 0.73., Conclusion: Fecal abundance of F. nucleatum may be a reliable predictor for metachronous adenoma after endoscopic polypectomy., (© 2021 Journal of Gastroenterology and Hepatology Foundation and John Wiley & Sons Australia, Ltd.)

In deep face recognition, the commonly used softmax loss and its newly proposed variations are not yet sufficiently effective to handle the class imbalance and softmax saturation issues during the training process while extracting discriminative features. In this brief, to address both issues, we propose a class-variant margin (CVM) normalized softmax loss, by introducing a true-class margin and a false-class margin into the cosine space of the angle between the feature vector and the class-weight vector. The true-class margin alleviates the class imbalance problem, and the false-class margin postpones the early individual saturation of softmax. With negligible computational complexity increment during training, the new loss function is easy to implement in the common deep learning frameworks. Comprehensive experiments on the LFW, YTF, and MegaFace protocols demonstrate the effectiveness of the proposed CVM loss function.

Background: Hepatitis B cirrhosis with hyperalphafetoproteinemia is the intermediate stage of liver cirrhosis progressing to hepatocellular carcinoma (HCC), there is no effective way to treat precancerous lesions of liver in modern medicine. In recent decades, clinical and experimental evidence shows that Chinese medicine (CM) has a certain beneficial effect on Hepatitis B Cirrhosis. Therefore, this trial aims to evaluate the efficacy and safety of a CM erzhu jiedu recipe (EZJDR) for the treatment of Hepatitis B Cirrhosis with Hyperalphafetoproteinemia., Methods: We designed a randomized, double blind, placebo-controlled clinical trial. A total of 72 patients of Hepatitis B Cirrhosis with hyperalphafetoproteinemia were randomized in 2 parallel groups. Patients in the control group received placebo granules similar to the EZJDR. In the EZJDR group, patients received EZJDR twice a day, after meals, for 48 weeks. The primary efficacy measures were changes in serum alpha-fetoprotein (AFP) and alpha-fetoprotein alloplasm (AFP-L3); The secondary indicators of efficacy are changes in liver function indicators, HBV-DNA level; Liver stiffness measurement (LSM); Hepatic portal vein diameter; T lymphocyte subgroup indexes during treatment. All data will be recorded in case report forms and analyzed by Statistical Analysis System software. Adverse events will also be evaluated., Results: The results showed that EZJDR can significantly inhibit the levels of AFP and AFP-L3 in patients with hepatitis B cirrhosis and hyperalphafetoproteinemia and have good security., Ethics and Dissemination: The study protocol was approved by the Medical Ethics Committee of Shuguang Hospital, affiliated with University of Traditional Chinese Medicine, Shanghai (NO.2018-579-08-01)., Trial Registration: This trial was registered on Chinese Clinical Trial Center (NO.ChiCTR1800017165)., Competing Interests: The authors have no conflicts of interests to disclose., (Copyright © 2021 the Author(s). Published by Wolters Kluwer Health, Inc.)

Three new phlorizin derivatives, 6"- O -vanilloylphlorizin ( 1 ), 6"- O -(4-hydroxybenzoyl)phlorizin ( 2 ), 6"- O -feruloylphlorizin ( 3 ), along with four known dihydrochalcones, phlorizin ( 4 ), 3-hydroxyphlorizin, trilobatin, and 6"- O -acetylphlorizin were isolated from the leaves of Lithocarpus litseifolius . Their structures were established by analysis of extensive spectroscopic data. The new compounds were shown to be non-cytotoxic when tested against A549, HeLa, HepG2, and MCF-7 cell lines.

Background and Aims: Although ticagrelor exerts an antibacterial activity, its effect on infections in patients with ST-segment elevation myocardial infarction (STEMI) undergoing percutaneous coronary intervention (PCI) is unclear. We aimed to assess whether ticagrelor and clopidogrel affect infections in these patients during hospitalization., Methods: A total of 2116 consecutive patients with STEMI undergoing PCI were divided into the ticagrelor (n = 388) and clopidogrel (n = 1728) groups. The primary outcome was infection onset. Secondary outcomes were in-hospital all-cause death and major adverse cardiovascular and cerebrovascular events (MACCE). Propensity score analyses were conducted to test the robustness of the results., Results: Infections developed in 327 (15.4%) patients. There was no significant difference in infection between both groups (ticagrelor vs. clopidogrel: 13.1% vs. 16.0%, p = 0.164). Patients in the ticagrelor group had lower rates of in-hospital all-cause death and MACCE than patients in the clopidogrel group. Multivariate logistic regression analysis determined that ticagrelor and clopidogrel had a similar preventive effect on infections during hospitalization (adjusted odds ratio [OR] = 1.20; 95% confidence interval [CI] = 0.80-1.78, p = 0.380). Compared to the patients treated with clopidogrel, patients treated with ticagrelor had a slightly lower risk of other outcomes, but no statistical difference. Propensity score analyses demonstrated similar results for infections and other outcomes., Conclusions: Compared with clopidogrel treatment, ticagrelor treatment did not significantly alter the risk of infections during hospitalization among STEMI patients undergoing PCI, but was associated with a slightly lower risk of in-hospital all-cause death and MACCE., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2021 Elsevier B.V. All rights reserved.)

Elastic weight consolidation (EWC) has been successfully applied for general incremental learning to overcome the catastrophic forgetting issue. It adaptively constrains each parameter of the new model not to deviate much from its counterpart in the old model during fine-tuning on new class data sets, according to its importance weight for old tasks. However, the previous study demonstrates that it still suffers from catastrophic forgetting when directly used in object detection. In this article, we show EWC is effective for incremental object detection if with critical adaptations. First, we conduct controlled experiments to identify two core issues why EWC fails if trivially applied to incremental detection: 1) the absence of old class annotations in new class images makes EWC misclassify objects of old classes in these images as background and 2) the quadratic regularization loss in EWC easily leads to gradient explosion when balancing old and new classes. Then, based on the abovementioned findings, we propose the corresponding solutions to tackle these issues: 1) utilize pseudobounding box annotations of old classes on new data sets to compensate for the absence of old class annotations and 2) adopt a novel Huber regularization instead of the original quadratic loss to prevent from unstable training. Finally, we propose a general EWC-based incremental object detection framework and implement it under both Fast R-CNN and Faster R-CNN, showing its flexibility and versatility. In terms of either the final performance or the performance drop with respect to the upper bound of joint training on all seen classes, evaluations on the PASCAL VOC and COCO data sets show that our method achieves a new state of the art.