Intrinsic Consistency Preservation With Adaptively Reliable Samples for Source-Free Domain Adaptation

Unsupervised domain adaptation (UDA) aims to alleviate the domain shift by transferring knowledge learned from a labeled source dataset to an unlabeled target domain. Although UDA has seen promising progress recently, it requires access to data from both domains, making it problematic in source data-absent scenarios. In this article, we investigate a practical task source-free domain adaptation (SFDA) that alleviates the limitations of the widely studied UDA in simultaneously acquiring source and target data. In addition, we further study the imbalanced SFDA (ISFDA) problem, which addresses the intra-domain class imbalance and inter-domain label shift in SFDA. We observe two key issues in SFDA that: 1) target data form clusters in the representation space regardless of whether the target data points are aligned with the source classifier and 2) target samples with higher classification confidence are more reliable and have less variation in their classification confidence during adaptation. Motivated by these observations, we propose a unified method, named intrinsic consistency preservation with adaptively reliable samples (ICPR), to jointly cope with SFDA and ISFDA. Specifically, ICPR first encourages the intrinsic consistency in the predictions of neighbors for unlabeled samples with weak augmentation (standard flip-and-shift), regardless of their reliability. ICPR then generates strongly augmented views specifically for adaptively selected reliable samples and is trained to fix the intrinsic consistency between weakly and strongly augmented views of the same image concerning predictions of neighbors and their own. Additionally, we propose to use a prototype-like classifier to avoid the classification confusion caused by severe intra-domain class imbalance and inter-domain label shift. We demonstrate the effectiveness and general applicability of ICPR on six benchmarks of both SFDA and ISFDA tasks. The reproducible code of our proposed ICPR method is available at https://github.com/CFM-MSG/Code_ICPR.

Avialan-like brain morphology in Sinovenator (Troodontidae, Theropoda)

Many modifications to the skull and brain anatomy occurred along the lineage encompassing non-avialan theropod dinosaurs and modern birds. Anatomical changes to the endocranium include an enlarged endocranial cavity, relatively larger optic lobes that imply elevated visual acuity, and proportionately smaller olfactory bulbs that suggest reduced olfactory capacity. Here, we use micro-computed tomographic (μCT) imaging to reconstruct the endocranium and its neuroanatomical features from an exceptionally well-preserved skull of Sinovenator changii (Troodontidae, Theropoda). While its overall morphology resembles the typical endocranium of other troodontids, Sinovenator also exhibits unique endocranial features that are similar to other paravian taxa and non-maniraptoran theropods. Landmark-based geometric morphometric analysis on endocranial shape of non-avialan and avialan dinosaurs points to the overall brain morphology of Sinovenator most closely resembling that of Archaeopteryx, thus indicating acquisition of avialan-grade brain morphology in troodontids and wide existence of such architecture in Maniraptora.

A Systematic Investigation of Proteoforms with N-Terminal Glycine and Their Dynamics Reveals Its Impacts on Protein Stability

The N-termini of proteins can regulate their degradation, and the same protein with different N-termini may have distinct dynamics. Recently, it was found that N-terminal glycine can serve as a degron recognized by two E3 ligases, but N-terminal glycine was also reported to stabilize proteins. Here we developed a chemoenzymatic method for selective enrichment of proteoforms with N-terminal glycine and integrated dual protease cleavage to further improve the enrichment specificity. Over 2000 unique peptides with protein N-terminal glycine were analyzed from >1000 proteins, and most of them are previously unknown, indicating the effectiveness of the current method to capture low-abundance proteoforms with N-terminal glycine. The degradation rates of proteoforms with N-terminal glycine were quantified along with those of proteins from the whole proteome. Bioinformatic analyses reveal that proteoforms with N-terminal glycine with the fastest and slowest degradation rates have different functions and localizations. Membrane proteins with N-terminal glycine and proteins with N-terminal glycine from the N-terminal methionine excision degrade more rapidly. Furthermore, the secondary structures, adjacent amino acid residues, and protease specificities for N-terminal glycine are also vital for protein degradation. The results advance our understanding of the effects of N-terminal glycine on protein properties and functions.

Learning Relationship-Enhanced Semantic Graph for Fine-Grained Image-Text Matching

Image-text matching of natural scenes has been a popular research topic in both computer vision and natural language processing communities. Recently, fine-grained image-text matching has shown its significant advance in inferring the high-level semantic correspondence by aggregating pairwise region-word similarity, but it remains challenging mainly due to insufficient representation of high-order semantic concepts and their explicit connections in one modality as its matched in another modality. To tackle this issue, we propose a relationship-enhanced semantic graph (ReSG) model, which can improve the image-text representations by learning their locally discriminative semantic concepts and then organizing their relationships in a contextual order. To be specific, two tailored graph encoders, visual relationship-enhanced graph (VReG) and textual relationship-enhanced graph (TReG), are respectively exploited to encode the high-level semantic concepts of corresponding instances and their semantic relationships. Meanwhile, the representations of each graph node are optimized by aggregating semantically contextual information to enhance the node-level semantic correspondence. Further, the hard-negative triplet ranking loss, center hinge loss, and positive-negative margin loss are jointly leveraged to learn the fine-grained correspondence between the ReSG representations of image and text, whereby the discriminative cross-modal embeddings can be explicitly obtained to benefit various image-text matching tasks in a more interpretable way. Extensive experiments verify the advantages of the proposed fine-grained graph matching approach, by achieving the state-of-the-art image-text matching results on public benchmark datasets.

Relation-Aggregated Cross-Graph Correlation Learning for Fine-Grained Image-Text Retrieval

Fine-grained image-text retrieval has been a hot research topic to bridge the vision and languages, and its main challenge is how to learn the semantic correspondence across different modalities. The existing methods mainly focus on learning the global semantic correspondence or intramodal relation correspondence in separate data representations, but which rarely consider the intermodal relation that interactively provide complementary hints for fine-grained semantic correlation learning. To address this issue, we propose a relation-aggregated cross-graph (RACG) model to explicitly learn the fine-grained semantic correspondence by aggregating both intramodal and intermodal relations, which can be well utilized to guide the feature correspondence learning process. More specifically, we first build semantic-embedded graph to explore both fine-grained objects and their relations of different media types, which aim not only to characterize the object appearance in each modality, but also to capture the intrinsic relation information to differentiate intramodal discrepancies. Then, a cross-graph relation encoder is newly designed to explore the intermodal relation across different modalities, which can mutually boost the cross-modal correlations to learn more precise intermodal dependencies. Besides, the feature reconstruction module and multihead similarity alignment are efficiently leveraged to optimize the node-level semantic correspondence, whereby the relation-aggregated cross-modal embeddings between image and text are discriminatively obtained to benefit various image-text retrieval tasks with high retrieval performance. Extensive experiments evaluated on benchmark datasets quantitatively and qualitatively verify the advantages of the proposed framework for fine-grained image-text retrieval and show its competitive performance with the state of the arts.

Fenton-like activity and pathway modulation via single-atom sites and pollutants comediates the electron transfer process

The studies on the origin of versatile oxidation pathways toward targeted pollutants in the single-atom catalysts (SACs)/peroxymonosulfate (PMS) systems were always associated with the coordination structures rather than the perspective of pollutant characteristics, and the analysis of mechanism commonality is lacking. In this work, a variety of single-atom catalysts (M-SACs, M: Fe, Co, and Cu) were fabricated via a pyrolysis process using lignin as the complexation agent and substrate precursor. Sixteen kinds of commonly detected pollutants in various references were selected, and their lnkobs values in M-SACs/PMS systems correlated well (R2 = 0.832 to 0.883) with their electrophilic indexes (reflecting the electron accepting/donating ability of the pollutants) as well as the energy gap (R2 = 0.801 to 0.840) between the pollutants and M-SACs/PMS complexes. Both the electron transfer process (ETP) and radical pathways can be significantly enhanced in the M-SACs/PMS systems, while radical oxidation was overwhelmed by the ETP oxidation toward the pollutants with lower electrophilic indexes. In contrast, pollutants with higher electrophilic indexes represented the weaker electron-donating capacity to the M-SACs/PMS complexes, which resulted in the weaker ETP oxidation accompanied with noticeable radical oxidation. In addition, the ETP oxidation in different M-SACs/PMS systems can be regulated via the energy gaps between the M-SACs/PMS complexes and pollutants. As a result, the Fenton-like activities in the M-SACs/PMS systems could be well modulated by the reaction pathways, which were determined by both electrophilic indexes of pollutants and single-atom sites. This work provided a strategy to establish PMS-based AOP systems with tunable oxidation capacities and pathways for high-efficiency organic decontamination.

MSFlow: Multiscale Flow-Based Framework for Unsupervised Anomaly Detection

Unsupervised anomaly detection (UAD) attracts a lot of research interest and drives widespread applications, where only anomaly-free samples are available for training. Some UAD applications intend to locate the anomalous regions further even without any anomaly information. Although the absence of anomalous samples and annotations deteriorates the UAD performance, an inconspicuous, yet powerful statistics model, the normalizing flows, is appropriate for anomaly detection (AD) and localization in an unsupervised fashion. The flow-based probabilistic models, only trained on anomaly-free data, can efficiently distinguish unpredictable anomalies by assigning them much lower likelihoods than normal data. Nevertheless, the size variation of unpredictable anomalies introduces another inconvenience to the flow-based methods for high-precision AD and localization. To generalize the anomaly size variation, we propose a novel multiscale flow-based framework (MSFlow) composed of asymmetrical parallel flows followed by a fusion flow to exchange multiscale perceptions. Moreover, different multiscale aggregation strategies are adopted for image-wise AD and pixel-wise anomaly localization according to the discrepancy between them. The proposed MSFlow is evaluated on three AD datasets, significantly outperforming existing methods. Notably, on the challenging MVTec AD benchmark, our MSFlow achieves a new state-of-the-art (SOTA) with a detection AUORC score of up to 99.7%, localization AUCROC score of 98.8% and PRO score of 97.1%.

The Multi-domain Lifestyle Intervention for Cognitive Impairment in Community-Dwelling Older Adults in Hangzhou (The Heritage Study): Study Design and Protocol

BACKGROUND

The globe has been working to promote a multi-domain lifestyle intervention for dementia prevention in older adults, referring to the Worldwide-FINGERS (Finnish Geriatric Intervention Study to Prevent Cognitive Impairment and Disability) initiative. In China, the multi-domain lifestyle intervention has been implemented in rural communities (MIND-China), yet the adaptability of such intervention based on the urban communities in China has not been verified.

OBJECTIVE

To examine the effectiveness and feasibility of the multi-domain lifestyle intervention on dementia prevention in at-risk community-dwelling older adults in China.

DESIGN, SETTING, PARTICIPANTS

The multi-domain lifestyle intervention study is a community-based 2-year cluster randomized controlled trial (RCT). A total of 1200 participants aged 60-80 years old will be recruited from twelve communities in Hangzhou, Zhejiang. Inclusion criteria were the Montreal Cognitive Assessment 5 minutes protocol (5 min MoCA) score of 6-9 or the Ascertain Dementia 8 (AD 8) score of ≥2, and having modifiable lifestyle factors.

INTERVENTION, MEASUREMENTS, RESULTS

Participating communities will be randomized into either the structured multi-domain intervention (SMI) arm or the self-guided intervention (SGI, general health education) arm. The SMI consists of cognitive training, physical exercise, and nutritional and dietary instruction for the first 12 months; and vascular risks monitoring and control for 24 months. The primary outcome is the global cognitive performance, measured by the comprehensive Neuropsychological Test Battery (NTB). The secondary outcomes include domain-specific cognitive performances, physical function, mental health, physiological and biochemical indices, adherence to healthy lifestyles, and neuroimaging metrics. The feasibility of intervention will be evaluated around the five dimensions of the RE-AIM framework and in conjunction with quantitative data, operational data and results of focus group discussions.

CONCLUSIONS

Following the Worldwide-FINGERS, this cluster RCT will verify the adaptability of the multi-domain lifestyle intervention in the urban community settings in China. This study will add evidence for global dementia prevention and management among older adults.

Mass spectrometry-based methods for investigating the dynamics and organization of the surfaceome: exploring potential clinical implications

INTRODUCTION

Cell-surface proteins are extremely important for many cellular events, such as regulating cell-cell communication and cell-matrix interactions. Aberrant alterations in surface protein expression, modification (especially glycosylation), and interactions are directly related to human diseases. Systematic investigation of surface proteins advances our understanding of protein functions, cellular activities, and disease mechanisms, which will lead to identifying surface proteins as disease biomarkers and drug targets.

AREAS COVERED

In this review, we summarize mass spectrometry (MS)-based proteomics methods for global analysis of cell-surface proteins. Then, investigations of the dynamics of surface proteins are discussed. Furthermore, we summarize the studies for the surfaceome interaction networks. Additionally, biological applications of MS-based surfaceome analysis are included, particularly highlighting the significance in biomarker identification, drug development, and immunotherapies.

EXPERT OPINION

Modern MS-based proteomics provides an opportunity to systematically characterize proteins. However, due to the complexity of cell-surface proteins, the labor-intensive workflow, and the limit of clinical samples, comprehensive characterization of the surfaceome remains extraordinarily challenging, especially in clinical studies. Developing and optimizing surfaceome enrichment methods and utilizing automated sample preparation workflow can expand the applications of surfaceome analysis and deepen our understanding of the functions of cell-surface proteins.

Fruit development and ripening orchestrating the biosynthesis and regulation of Lycium barbarum polysaccharides in goji berry

Lycium barbarum polysaccharides (LBPs) are the primary bioactive components in fruits of L. barbarum, commonly known as goji berry. Despite significant progress in understanding the chemical structures and health benefits of LBPs, the biosynthesis and regulation of LBPs in goji berry remains largely unknown. In this study, physiological indicators, including LBPs, were monitored in goji berry during fruit development and ripening (FDR), suggesting that pectin might be the major component of LBPs with increased content reaching 235.8 mg/g DW. Proteomic and transcriptomic analysis show that 6410 differentially expressed genes (DEGs) and 2052 differentially expressed proteins (DEPs) were identified with overrepresentation of flavonoids and polysaccharides-related gene ontology (GO) terms and KEGG pathways. Weighted gene co-expression network analysis (WGCNA) showed that LBPs coexpress with genes involved in pectin biosynthesis (LbGALS3, LbGATL5, LbQUA1, LbGAUT1/4/7, LbRGGAT1, LbRRT1/7, and LbRHM2), modification (LbSBT1.7), and regulation (LbAP2, LbGL2 LbTLP2, LbERF4, and LbTTG2), as well as with novel transcription factors (LbSPL9 and LbRIN homologs) and glycosyltransferases. Transgenic hairy roots overexpressing LbRIN validated that LbRIN modulate the expression of WGCNA-predicted regulators, including LbERF4, LbTTG2, and LbSPL9. These findings suggest that the biosynthesis and regulation of LBPs is conserved partially to those in Arabidopsis pectin. Taken together, this study provides valuable insights into the biosynthesis and regulation of LBPs, which can facilitate future studies on synthetic biology applications and genetic improvement of LBPs.

Exogenous l-fucose attenuates neuroinflammation induced by lipopolysaccharide

α1,6-Fucosyltransferase (Fut8) catalyzes the transfer of fucose to the innermost GlcNAc residue of N-glycan to form core fucosylation. Our previous studies showed that lipopolysaccharide (LPS) treatment highly induced neuroinflammation in Fut8 homozygous KO (Fut8-/-) or heterozygous KO (Fut8+/-) mice, compared with the WT (Fut8+/+) mice. To understand the underlying mechanism, we utilized a sensitive inflammation-monitoring mouse system that contains the human interleukin-6 (hIL6) bacterial artificial chromosome transgene modified with luciferase (Luc) reporter cassette. We successfully detected LPS-induced neuroinflammation in the central nervous system by exploiting this bacterial artificial chromosome transgenic monitoring system. Then we examined the effects of l-fucose on neuroinflammation in the Fut8+/- mice. The lectin blot and mass spectrometry analysis showed that l-fucose preadministration increased the core fucosylation levels in the Fut8+/- mice. Notably, exogenous l-fucose attenuated the LPS-induced IL-6 mRNA and Luc mRNA expression in the cerebral tissues, confirmed using the hIL6-Luc bioluminescence imaging system. The activation of microglial cells, which provoke neuroinflammatory responses upon LPS stimulation, was inhibited by l-fucose preadministration. l-Fucose also suppressed the downstream intracellular signaling of IL-6, such as the phosphorylation levels of JAK2 (Janus kinase 2), Akt (protein kinase B), and STAT3 (signal transducer and activator of transcription 3). l-Fucose administration increased gp130 core fucosylation levels and decreased the association of gp130 with the IL-6 receptor in Fut8+/- mice, which was further confirmed in BV-2 cells. These results indicate that l-fucose administration ameliorates the LPS-induced neuroinflammation in the Fut8+/- mice, suggesting that core fucosylation plays a vital role in anti-inflammation and that l-fucose is a potential prophylactic compound against neuroinflammation.

[Comparison of the efficiency of different etiological assays for detection of Schistosoma japonicum infections in wild mice]

OBJECTIVE

To compare the efficiency of multiple etiological techniques for detection of Schistosoma japonicum infections in wild mice, so as to provide technical supports to assessment of schistosomiasis transmission risk.

METHODS

Wild mice were captured with baited traps at night in Oncomelania hupensis snail-infested settings in schistosomiasis-endemic foci of Anhui Province from October to November, 2022. S. japonicum infections were detected in wild mice using microscopy of mouse liver tissues, microscopy of mouse mesenteric tissues, microscopy of mouse liver tissue homogenates, miracidial hatching test of mouse liver tissue homogenates, Kato-Katz technique and miracidial hatching test of mouse stool samples alone and in combinations. Identification of S. japonicum eggs or miracidia by any of these six assays was defined as an infection. The sensitivity of six assays alone or in combinations was compared for detection of S. japonicum infections in wild mice.

RESULTS

A total of 1 703 wild mice were captured, with 366 wild mice detected positive for S. japonicum (21.49%). There were significant differences in the prevalence of S. japonicum infections in wild mice by six assays (Q = 529.33, P < 0.001) and in the sensitivity of six assays for detection of S. japonicum infections in wild mice (χ2 = 527.78, P < 0.001). In addition, the combination of microscopy of mouse liver tissues and mesenteric tissues, combination of microscopy of mouse liver tissues and liver tissue homogenates and combination of microscopy of mouse liver tissues, microscopy of mesenteric tissues, microscopy of liver tissue homogenates and Kato-Katz technique showed 86.61%, 87.16% and 97.27% sensitivities for detection of S. japonicum infections in wild mice, respectively.

CONCLUSIONS

Diverse etiological assays show various efficiencies for detection of S. japonicum infections in wild mice. Combination of microscopy of mouse liver tissues and microscopy of mesenteric tissues, and combination of microscopy of mouse liver tissues and microscopy of liver tissue homogenates are potential approaches for field detection of S. japonicum infections in wild mice.

Aerobic glycolysis is the predominant means of glucose metabolism in neuronal somata, which protects against oxidative damage

It is generally thought that under basal conditions, neurons produce ATP mainly through mitochondrial oxidative phosphorylation (OXPHOS), and glycolytic activity only predominates when neurons are activated and need to meet higher energy demands. However, it remains unknown whether there are differences in glucose metabolism between neuronal somata and axon terminals. Here, we demonstrated that neuronal somata perform higher levels of aerobic glycolysis and lower levels of OXPHOS than terminals, both during basal and activated states. We found that the glycolytic enzyme pyruvate kinase 2 (PKM2) is localized predominantly in the somata rather than in the terminals. Deletion of Pkm2 in mice results in a switch from aerobic glycolysis to OXPHOS in neuronal somata, leading to oxidative damage and progressive loss of dopaminergic neurons. Our findings update the conventional view that neurons uniformly use OXPHOS under basal conditions and highlight the important role of somatic aerobic glycolysis in maintaining antioxidant capacity.

The effect of cold exposure on the levels of glucocorticoids, 11-hydroxysteroid dehydrogenase 2, and placental vascularization in a rat model

OBJECTIVE

Cold exposure (CE) before birth is one of the initial stressors that may impact mammalian pregnancy, changing placental and fetal development and affecting the health of the offspring. While glucocorticoids (GCs) participate in the body's response to the stress of CE, the specific mechanisms of their action are unclear. This study aims to determine the effect of CE stress on the placenta and to test whether stress, caused by cold exposure in pregnancy impairs fetal development by changing placental angiogenesis via excessive GC expression.

MATERIALS AND METHODS

CE rat model was created by exposing 30 SD rats to cold preconception, or during the first, second, and third weeks of pregnancy. Serum cortisol and soluble fms-like tyrosine kinase-1 (sFlt-1) expression levels, physiological index changes (food intake, body weight change and blood pressure), and pregnancy outcomes (fetal rat weight, number of live fetal rats, and placental weight) were collected at baseline and at different time points after the conception. Protein expression levels of 11 β-hydroxysteroid dehydrogenase 2 (11β-HSD2), glucocorticoid receptor, vascular endothelial growth factor A (VEGF-A), placental growth factor (PIGF), and sFlt-1 in placental tissues were measured by western blotting. Cytokeratin (CK) and laminin (LN) in trophoblasts, and α-actin in vascular smooth muscle of the spiral arteries of pregnant rats after the systemic cold treatment were assessed by immunofluorescence and visualized by fluorescent microscopy. To test the effect of 11β-HSD2 levels on the placental recasting, human first-trimester extravillous trophoblast cells (HTR8/SVneo) underwent knockdown using specific 11β-HSD2 siRNA constructs.  Expression levels of 11β-HSD2 were analyzed by quantitative real-time PCR (qPCR) and into HTR8 cells, and the expression levels of the 11β-HSD2 gene in each group were measured using qPCR. Cell migration and invasion was assessed by Transwell migration assay, and sFlt-1 levels in HTR8 cells were measured by ELISA.

RESULTS

CE pre-conception led to consistently increasing serum corticosterone and sFlt-1 levels throughout pregnancy, and persistently increased diastolic blood pressure (DBP) in rat CE model compared to control animals. CE during the second week of gestation (Gp.3) was associated with significantly lower placental weight (p=0.0003). Cold exposure in the third week (Gp.4) was associated with significantly (p=0.001) lower fetal weight. CE pre-conception was associated with significantly decreased placental levels of 11β-HSD2, glucocorticoid receptor, VEGF-A, PIGF, and sFlt-1 proteins and α-actin compared to the control group. Silencing 11β-HSD2 by siRNA led to reduced cell migrations and invasion, and markedly increased expression levels of sFlt-1 in HTR8/SVneo cells (p<0.05).

CONCLUSIONS

Pre-conception cold exposure and during early pregnancy leads to increased GCs levels and impaired placental 11β-HSD2 activity. We suggest that the subsequent 11β-HSD2-induced increase in the sFlt-1expression during early pregnancy may affect placental vascular remodeling and change placental morphological structure and function.

Prognostic value of plaque volume combined with CT fractional flow reserve in patients with suspected coronary artery disease

AIM

To investigate the prognostic value of quantitative plaque volume on coronary computed tomography (CT) angiography (CTA) combined with CT fractional flow reserve (CT-FFR) for major adverse cardiac events (MACE) in suspected coronary artery disease (CAD) patients.

MATERIALS AND METHODS

Patients who underwent coronary CTA with clinically suspected CAD were enrolled retrospectively in this study. Patients' baseline, Framingham Risk Score (FRS), coronary CTA plaque assessment, and CT-FFR were analysed retrospectively. Study outcomes included rehospitalisation and MACE (ST-segment elevation myocardial infarction, unstable angina, or non-ST-segment elevation myocardial infarction, revascularisation, and cardiac death).

RESULTS

There were 251 patients in the study, with a follow-up period of 1-6.58 years. Mean age was 61.16 ± 10.45 years and 146 (58%) patients were male. Higher CT-adapted Leaman score and quantitative plaque volume were found in patients with FRS >0.2 regardless of categorical or continuous variables. Coronary scores, quantitative plaque parameters, and CT-FFR were associated with MACE and rehospitalisation in univariate analysis. In model 1, CT-FFR was associated with MACE in multivariate Cox analysis when adjusted for FRS and CT-adapted Leaman score. Quantitative plaque parameters including calcified plaque volume, fibro-fatty plaque volume, low-attenuation plaque volume, non-calcified plaque volume, and total plaque volume were significantly associated with MACE and improved overall prognostic performance in a model adjusted for CT-FFR.

CONCLUSION

Additional quantitative plaque volume and CT-FFR further improve the predictive incremental value based on risk factor scores for prognostic prediction in patients. Adding quantitative plaque volume combined with CT-FFR analysis to anatomical and clinical assessment will be further beneficial to predict patients' prognosis of MACE.

Visual analysis of autoimmune dry eye studies based on bibliometrics

BACKGROUND

Dry eye is a chronic ocular surface disease caused by the instability of tear film or the imbalance of the ocular surface microenvironment which can lead to a diverse range of ocular discomfort symptoms. At present, the relevant mechanism of autoimmunity and treatment of dry eye is still unclear. Due to the proliferation of research papers in this field, visual analysis of existing papers can provide reference for future research.

METHODS

The academic papers of Web of Science were searched with the topics of "autoimmunity" and "dry eye," and the countries, institutions and keywords of the literatures selected in this domain were visualized by Citespace and Vosviewer software.

RESULTS

A total of 787 valid international papers were detected, and the publication count exhibited a consistent upward trend year by year. Within this field, the US has produced the highest number of papers (363), with Baylor College of Medicine being the most prolific institution (28 publications). High-producing authors in this field include Artemis P. Simopoulos and Stephen C. Pflugfelder.

CONCLUSION

International research in this field has focused on the pathogenesis, symptoms, and treatment of dry eye. It is predicted that the future international research hotspots will be the pathophysiology of autoimmune dry eye disease, data analysis of artificial intelligence-related diseases, and research on improving patients' quality of life.

Multiscale Information Diffusion Prediction With Minimal Substitution Neural Network

Information diffusion prediction is a complex task due to the dynamic of information substitution present in large social platforms, such as Weibo and Twitter. This task can be divided into two levels: the macroscopic popularity prediction and the microscopic information diffusion prediction (who is next), which share the essence of modeling the dynamic spread of information. While many researchers have focused on the internal influence of individual cascades, they often overlook other influential factors that affect information diffusion, such as competition and cooperation among information, the attractiveness of information to users, and the potential impact of content anticipation on further diffusion. To address this issue, we propose a multiscale information diffusion prediction with minimal substitution (MIDPMS) neural network. This model simultaneously enables macroscale popularity prediction and microscale diffusion prediction. Specifically, information diffusion is modeled as a substitution system among different information. First, the life cycle of content, user preferences, and potential content anticipation are considered in this system. Second, a minimal-substitution-theory-based neural network is first proposed to model this substitution system to facilitate joint training of macroscopic and microscopic diffusion prediction. Finally, extensive experiments are conducted on Weibo and Twitter datasets to validate the performance of our proposed model on multiscale tasks. The results confirmed that the proposed model performed well on both multiscale tasks on Weibo and Twitter.

Efficacy and nephrotoxicity of polymyxin B in elderly patients with carbapenem resistant bacterial infection

BACKGROUND

To study the efficacy and nephrotoxicity of polymyxin B in the treatment of elderly patients with carbapenem-resistant organism (CRO) infection.

METHODS

The clinical and microbiological data of patients with CRO-infected sepsis treated with polymyxin B were retrospectively analyzed. The effective rate, bacterial clearance, incidence and recovery rate of acute renal injury (AKI) and prognosis-related indicators in AKI at different stages were compared.

RESULTS

The effective rate of 215 elderly patients with CRO infection treated with polymyxin was 50.7%. The total bacterial clearance rate was 44.2%, the total incidence of AKI was 37.2%, the recovery rate of AKI was 35%, and the incidence range of polymyxin B-related AKI was 10.2-37.2%. Logistic multivariate regression analysis showed that the predictors of AKI in elderly patients were high APACHE II score, long duration of polymyxin, chronic renal insufficiency and ineffective outcome; the ROC curve showed that the cutoff value for predicting AKI was a serum creatinine concentration of 73 mmol/L before polymyxin B use, and the AUC was 0.931.

CONCLUSIONS

Rational use of polymyxin B is safe and effective in elderly patients with CRO infection, and its effective outcome can improve the recovery rate of AKI.

Preliminary exploration of the biomechanical properties of three novel cervical porous fusion cages using a finite element study

BACKGROUND

Porous cages are considered a promising alternative to high-density cages because their interconnectivity favours bony ingrowth and appropriate stiffness tuning reduces stress shielding and the risk of cage subsidence.

METHODS

This study proposes three approaches that combine macroscopic topology optimization and micropore design to establish three new types of porous cages by integrating lattices (gyroid, Schwarz, body-centred cubic) with the optimized cage frame. Using these three porous cages along with traditional high-density cages, four ACDF surgical models were developed to compare the mechanical properties of facet articular cartilage, discs, cortical bone, and cages under specific loads.

RESULTS

The facet joints in the porous cage groups had lower contact forces than those in the high-density cage group. The intervertebral discs in all models experienced maximum stress at the C5/6 segment. The stress distribution on the cortical bone surface was more uniform in the porous cage groups, leading to increased average stress values. The gyroid, Schwarz, and BCC cage groups showed higher average stress on the C5 cortical bone. The average stress on the surface of porous cages was higher than that on the surface of high-density cages, with the greatest difference observed under the lateral bending condition. The BCC cage demonstrated favourable mechanical stability.

CONCLUSION

The new porous cervical cages satifies requirements of low rigidity and serve as a favourable biological scaffold for bone ingrowth. This study provides valuable insights for the development of next-generation orthopaedic medical devices.

[Research progress on biomarkers and detection methods for Alzheimer's disease diagnosis in vitro]

Alzheimer's disease (AD) is a neurodegenerative disease with insidious onset, posing a serious threat to human physical and mental health. The cognitive impairments caused by AD are generally diffuse and overlap symptomatically with other neurodegenerative diseases. Moreover, the symptoms of AD are often covert, leading to missed opportunities for optimal treatment after diagnosis. Therefore, early diagnosis of AD is crucial. In vitro diagnostic biomarkers not only contribute to the early clinical diagnosis of AD but also aid in further understanding the disease's pathogenesis, predicting disease progression, and observing the effects of novel candidate therapeutic drugs in clinical trials. Currently, although there are numerous biomarkers associated with AD diagnosis, the complex nature of AD pathogenesis, limitations of individual biomarkers, and constraints of clinical detection methods have hindered the development of efficient, cost-effective, and convenient diagnostic methods and standards. This article provides an overview of the research progress on in vitro diagnostic biomarkers and detection methods related to AD in recent years.

Composition-Aware Image Steganography Through Adversarial Self-Generated Supervision

Steganography is an important and prevailing information hiding tool to perform secret message transmission in an open environment. Existing steganography methods can mainly fall into two categories: predefined rule-based and data-driven methods. The former is susceptible to the statistical attack, while the latter adopts the deep convolution neural networks to promote security. However, deep learning-based methods suffer from perceptible artificial artifacts or deep steganalysis. In this article, we introduce a novel composition-aware image steganography (CAIS) to guarantee both visual security and resistance to deep steganalysis through the self-generated supervision. The key innovation is an adversarial composition estimation module, which has integrated the rule-based composition method and generative adversarial network to help synthesize steganographic images with more naturalness. We first perform a rule-based image blending method to obtain infinite synthetically data-label pairs. Then, we utilize an adversarial composition estimation branch to recognize the message feature pattern from the composite image based on these self-generated data-label pairs. Through the adversarial training, we force the steganography function to synthesize steganographic images, which can fool the composition estimation network. Thus, the proposed CAIS can achieve better information hiding and higher security to resist deep steganalysis. Furthermore, an effective global-and-part checking is designed to alleviate visual artifacts caused by hiding secret information. We conduct a comprehensive analysis of CAIS from various aspects (e.g., security and robustness) to verify the superior performance of the proposed method. Comprehensive experimental results on three large-scale widely used datasets have demonstrated the superior performance of our CAIS compared with several state-of-the-art approaches.

Expression of ICAM-1 and E-selectin in different metabolic obesity phenotypes: discrepancy for endothelial dysfunction

OBJECTIVES

Endothelial dysfunction, the earliest vascular alteration, is a consequence of metabolic disorders associated with obesity. However, it is still unclear whether a proportion of obese individuals without metabolic alterations associated with obesity, defined as "metabolically healthy obesity (MHO)", exhibit better endothelial function. We therefore aimed to investigate the association of different metabolic obesity phenotypes with endothelial dysfunction.

METHODS

The obese participants without clinical cardiovascular disease from the MESA (Multi-Ethnic Study of Atherosclerosis) were allocated to the different metabolic obesity phenotypes based on their metabolic status, including MHO and metabolically unhealthy obesity (MUO). Associations of metabolic obesity phenotypes with the biomarkers of endothelial dysfunction, including soluble intercellular adhesion molecule-1 (sICAM-1) and soluble E-selectin (sE-selectin), were evaluated using multiple linear regression models.

RESULTS

Plasma levels of sICAM-1 and sE-selectin were respectively measured in 2371 and 968 participants. Compared to the non-obese participants, those with MUO were associated with higher concentrations of sICAM-1 (β 22.04, 95% CI 14.33-29.75, P < 0.001) and sE-selectin (β 9.87, 95% CI 6.00-13.75, P < 0.001) after adjusting for confounders. However, no differences were found for the concentrations of sICAM-1 (β 0.70, 95% CI - 8.91 to 10.32, P = 0.886) and sE-selectin (β 3.69, 95% CI - 1.13 to 8.51, P = 0.133) in the participants with MHO compared to the non-obese participants.

CONCLUSIONS

Individuals with MUO were associated with elevated biomarkers of endothelial dysfunction, but the association with endothelial dysfunction was not found in those with MHO, indicating that the individuals with MHO might exhibit better endothelial function.

HGF-Based CAR-T Cells Target Hepatocellular Carcinoma Cells That Express High Levels of c-Met

BACKGROUND

CAR-T is emerging as an effective treatment strategy for hematologic malignancies, however its effectiveness for treating solid tumors, such as Hepatocellular Carcinoma (HCC) is limited. Here, we screened a variety of CAR-T cells that target c-Met to investigate their potential to induce HCC cell death in vitro.

METHODS

Human T cells were transduced to express CARs by lentiviral vector transfection. c-Met expression in human HCC cell lines and CARs expression were monitored by flow cytometry. Tumor cell killing was evaluated by Luciferase Assay System Kit. The concentrations of cytokine were tested by Enzyme-linked immunosorbent assays. Knock down and overexpression studies targeting c-Met were conducted to assess the targeting specificity of CARs.

RESULTS

We found that CAR T cells expressing a minimal amino-terminal polypeptide sequence comprising the first kringle (kringle 1) domain (denoted as NK1 CAR-T cells), efficiently killed HCC cell lines that expressed high levels of the HGF receptor c-Met. Furthermore, we report that while NK1 CAR-T cells were efficient at targeting SMMC7221 cells for destruction, and its potency was significantly attenuated in parallel experiments with cells stably expressing short hairpin RNAs (shRNAs) that suppressed c-Met expression. Correspondingly, overexpression of c-Met in the embryonic kidney cell line HEK293T led to their enhanced killing by NK1 CAR-T cells.

CONCLUSION

Our studies demonstrate that a minimal amino-terminal polypeptide sequence comprising the kirngle1 domain of HGF is highly relevant to the design of effective CAR-T cell therapies that kill HCC cells expressing high levels of c-Met.