Publications by authors named "Qinghua Hu"

208 Publications

Outbreak dynamics of foodborne pathogen Vibrio parahaemolyticus over a seventeen year period implies hidden reservoirs.

Nat Microbiol 2022 Aug 2;7(8):1221-1229. Epub 2022 Aug 2.

Shenzhen Center for Disease Control and Prevention, Shenzhen, Guangdong Province, China.

Controlling foodborne diseases requires robust outbreak detection and a comprehensive understanding of outbreak dynamics. Here, by integrating large-scale phylogenomic analysis of 3,642 isolates and epidemiological data, we performed 'data-driven' outbreak detection and described the long-term outbreak dynamics of the leading seafood-associated pathogen, Vibrio parahaemolyticus, in Shenzhen, China, over a 17-year period. Contradictory to the widely accepted notion that sporadic patients and independent point-source outbreaks dominated foodborne infections, we found that 71% of isolates from patients grouped into within-1-month clusters that differed by ≤6 single nucleotide polymorphisms, indicating putative outbreaks. Furthermore, we showed that despite the long time spans between clusters, 70% of them were genomically closely related and were inferred to arise from a small number of common sources, which provides evidence that hidden persistent reservoirs generated most of the outbreaks rather than independent point-sources. Phylogeographical analysis further revealed the geographical heterogeneity of outbreaks and identified a coastal district as the potential hotspot of outbreaks and as the hub and major source of cross-district spread events. Our findings provide a comprehensive picture of the long-term spatiotemporal dynamics of foodborne outbreaks and present a different perspective on the major source of foodborne infections, which will inform the design of future disease control strategies.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1038/s41564-022-01182-0DOI Listing
August 2022

Autoencoder in Autoencoder Networks.

IEEE Trans Neural Netw Learn Syst 2022 Jul 15;PP. Epub 2022 Jul 15.

Modeling complex correlations on multiview data is still challenging, especially for high-dimensional features with possible noise. To address this issue, we propose a novel unsupervised multiview representation learning (UMRL) algorithm, termed autoencoder in autoencoder networks (AE -Nets). The proposed framework effectively encodes information from high-dimensional heterogeneous data into a compact and informative representation with the proposed bidirectional encoding strategy. Specifically, the proposed AE -Nets conduct encoding in two directions: the inner-AE-networks extract view-specific intrinsic information (forward encoding), while the outer-AE-networks integrate this view-specific intrinsic information from different views into a latent representation (backward encoding). For the nested architecture, we further provide a probabilistic explanation and extension from hierarchical variational autoencoder. The forward-backward strategy flexibly addresses high-dimensional (noisy) features within each view and encodes complementarity across multiple views in a unified framework. Extensive results on benchmark datasets validate the advantages compared to the state-of-the-art algorithms.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TNNLS.2022.3189239DOI Listing
July 2022

Application of retrograde cerebral perfusion in aortic root replacement combined with right half aortic arch replacement.

Zhong Nan Da Xue Xue Bao Yi Xue Ban 2022 May;47(5):650-654

Department of Cardiovascular Surgery, Xiangya Hospital, Central South University, Changsha 410008.

Objectives: After cardiac surgery involving the aortic arch, the incidence of neurological complications remains high, therefore it is very important to take measures to protect brain. This study is to investigate the safety and effectiveness of deep hypothermic circulatory arrest and retrograde cerebral perfusion for aortic root combined with right half aortic arch replacement.

Methods: Clinical data of 31 patients, who underwent aortic root and right half aortic arch replacement with deep hypothermic circulatory arrest and retrograde cerebral perfusion in Xiangya Hospital, Central South University, were retrospectively analyzed. This cohort included 23 aortic aneurysms and 8 aortic dissections. Aortic root replacement was conducted in 26 patients by Bentall procedures, and 5 patients by David procedures. Time of deep hypothermic circulatory arrest and retrograde cerebral perfusion in surgery was (21.9±5.2) min. The in-hospital mortality, postoperative neurological dysfunction and other major adverse complications were observed and recorded.

Results: No in-hospital death and permanent neurological dysfunction occurred. Two patients had transient neurological dysfunction and 2 patients with aortic dissection requiring long-time ventilation due to hypoxemia, 1 patient underwent resternotomy. During 6-36 months of follow-up, all patients recovered satisfactorily.

Conclusions: Deep hypothermic circulatory arrest and retrograde cerebral perfusion can be safely and effectively applied in aortic root and right half aortic arch replacement, and which can simplify the surgical procedures and be worth of clinical promotion.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.11817/j.issn.1672-7347.2022.210288DOI Listing
May 2022

Efficacy of subthreshold micropulse laser for chronic central serous chorioretinopathy: A meta-analysis.

Photodiagnosis Photodyn Ther 2022 May 27;39:102931. Epub 2022 May 27.

Aier Eye Hospital of Wuhan University, Wuhan, 430050, China. Electronic address:

Purpose: To evaluate the efficacy of subthreshold micropulse laser (SML) treatment for chronic central serous chorioretinopathy (CSC).

Methods: PubMed, Cochrane Library, and Embase were comprehensively searched for studies published up to April 19, 2021. Randomized controlled trials (RCTs), retrospective and prospective cohort studies that compared SML with any other intervention for chronic CSC were selected. The primary outcomes were best corrected visual acuity (BCVA) and central macular thickness (CMT). Meta-analysis was performed using Review Manager 5.3. Random-effect model was used for pooled analysis.

Results: Eleven studies including 834 eyes were included, with 428 eyes undergoing SML treatment and 406 eyes receiving other interventions. Pooled results showed no significant differences between SML and control with respect to BCVA, CMT, or complete resolution of subretinal fluid, while SML treatment was inferior to control in terms of subfoveal choroidal thickness (SFCT) reduction [weighted mean difference (WMD)=44.75, 95% Cl, 14.31-75.20, P<0.05, I=72%]. However, SML was not superior to photodynamic therapy with respect to SFCT (WMD=61.29, 95% Cl, 29.50-93.08, P<0.05, I=56%), retinal sensitivity (WMD=-1.85, 95% Cl, -3.05--0.65, P<0.01, I=0) or NEI-VFQ25 score (WMD=-2.44, 95% Cl, -4.72--0.16, P<0.05, I=0). No serious side effects of SML treatment were observed.

Conclusion: Available evidence suggests that the clinical efficacy of SML therapy is similar to other treatment modalities for chronic CSC. However, because of the lack of serious side effects, SML may be the most promising alternative therapy for chronic CSC. The study protocol was registered on PROSPERO (CRD42021 258837).
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.pdpdt.2022.102931DOI Listing
May 2022

Rapid Multilateral and Integrated Public Health Response to a Cross-City Outbreak of Enteritidis Infections Combining Analytical, Molecular, and Genomic Epidemiological Analysis.

Front Microbiol 2022 4;13:772489. Epub 2022 May 4.

Shenzhen Major Infectious Disease Control Key Laboratory, Shenzhen Center for Disease Control and Prevention, Shenzhen, China.

On September 21, 2019, the Shenzhen and Dongguan Centers for Disease Control and Prevention received notification of a large cluster of suspected gastroenteritis involving primarily children who sought medical care at hospitals throughout two adjacent cities in China, Shenzhen, and Dongguan. A joint outbreak response was promptly initiated across jurisdictions in a concerted effort between clinical microbiologists, epidemiologists, and public health scientists. Concurrently, multiplex PCRs were used for rapid laboratory diagnosis of suspected cases; epidemiological investigations were conducted to identify the outbreak source, complemented by near real-time multicenter whole-genome analyses completed within 34 h. Epidemiological evidence indicated that all patients had consumed egg sandwiches served on September 20 as snacks to children and staff at a nursery in Dongguan, located near Shenzhen. Enteritidis was isolated from case-patients, food handlers, kitchenware, and sandwiches with kitchen-made mayonnaise. Whole-genome single-nucleotide polymorphism (SNP)-based phylogenetic analysis demonstrated a well-supported cluster with pairwise distances of ≤1 SNP between genomes for outbreak-associated isolates, providing the definitive link between all samples. In comparison with historical isolates from the same geographical region, the minimum pairwise distance was >14 SNPs, suggesting a non-local outbreak source. Genomic source tracing revealed the possible transmission dynamics of a . Enteritidis clone throughout a multi-provincial egg distribution network. The efficiency and scale with which multidisciplinary and integrated approaches were coordinated in this foodborne disease outbreak response was unprecedented in China, leading to the timely intervention of a large cross-jurisdiction outbreak.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.3389/fmicb.2022.772489DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC9117964PMC
May 2022

Meta-analysis of Cd input-output fluxes in agricultural soil.

Chemosphere 2022 Sep 18;303(Pt 2):134974. Epub 2022 May 18.

Fujian Provincial Environmental Monitoring Center, Fuzhou, 350003, China.

Heavy metal pollution of agricultural soil, especially Cd, has become a global threat to food safety and human health. Analysis of Cd fluxes through different input/output pathways is widely used to predict the change of Cd content in agricultural soil, identify the critical pathways, and assist in developing effective management strategies to protect the environmental quality of agricultural soils. In the present study, literature recording input/output fluxes of Cd through different pathways in agricultural soils were investigated, with study areas primarily located in China, Japan, and Europe. Fluxes of Cd at the study sites were calculated, and comparative analyses were carried out. Results indicated that the dominant input pathway of Cd was strongly associated with the intensity of local industrial activities. Atmospheric deposition was the predominant input pathway of Cd for 75% of the study cases. Irrigation and livestock manure were also major pathways of Cd input in China. The main output pathways were influenced by the planting structure, precipitation, topography, etc. Crop harvesting and leaching to groundwater played important roles among all Cd output pathways in China, and crop harvesting alone could remove a significant amount of Cd from the soil, with an estimated average flux of 6.27 g/ha/yr. Leaching was the dominant Cd output pathway in Europe, accounting for 77%-93% of total outflux. To mitigate the accumulation of Cd in agricultural soil, standards to regulate Cd in the atmospheric environment, irrigation water, and agricultural additives should be tightened, and regulated removal and disposal of crops harvested from the heavily contaminated field should be promoted.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.chemosphere.2022.134974DOI Listing
September 2022

Collaborative Decision-Reinforced Self-Supervision for Attributed Graph Clustering.

IEEE Trans Neural Netw Learn Syst 2022 May 18;PP. Epub 2022 May 18.

Attributed graph clustering aims to partition nodes of a graph structure into different groups. Recent works usually use variational graph autoencoder (VGAE) to make the node representations obey a specific distribution. Although they have shown promising results, how to introduce supervised information to guide the representation learning of graph nodes and improve clustering performance is still an open problem. In this article, we propose a Collaborative Decision-Reinforced Self-Supervision (CDRS) method to solve the problem, in which a pseudo node classification task collaborates with the clustering task to enhance the representation learning of graph nodes. First, a transformation module is used to enable end-to-end training of existing methods based on VGAE. Second, the pseudo node classification task is introduced into the network through multitask learning to make classification decisions for graph nodes. The graph nodes that have consistent decisions on clustering and pseudo node classification are added to a pseudo-label set, which can provide fruitful self-supervision for subsequent training. This pseudo-label set is gradually augmented during training, thus reinforcing the generalization capability of the network. Finally, we investigate different sorting strategies to further improve the quality of the pseudo-label set. Extensive experiments on multiple datasets show that the proposed method achieves outstanding performance compared with state-of-the-art methods. Our code is available at https://github.com/Jillian555/TNNLS_CDRS.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TNNLS.2022.3171583DOI Listing
May 2022

Calcium Sensing Receptor Variants Increase Pulmonary Hypertension Susceptibility.

Hypertension 2022 07 28;79(7):1348-1360. Epub 2022 Apr 28.

Department of Pathophysiology, School of Basic Medicine, Tongji Medical College, Huazhong University of Science and Technology (HUST), Wuhan, China (B.L., X.F., Z.C., X.L.,Y.X., L.Z., Q.H.).

Background: Pulmonary arterial hypertension is an incurable disease, in which the extracellular CaSR (calcium sensing receptor) is mechanistically important. This study was aimed to genetically link the gene and function to the disease severity.

Methods: Sanger sequencing, Sugen/hypoxia pulmonary arterial hypertension rat model, mutated rat, transcriptional reporter assay and measurement of CaSR activity were used.

Results: Sanger sequencing identified a significant association between the variant rs1042636(A>G), located in exon 7, and idiopathic pulmonary arterial hypertension (IPAH) formation in patients. The frequency of 2968G homozygotes was higher in patients with IPAH compared with healthy individuals (23.6% versus 17.5%; =0.001, OR=1.864), and the minor alleles of rs6776158, rs1048213, and rs9883099, located in promoter, raised the IPAH odds ratio to 2.173. Patients with IPAH carrying heterozygotes or homozygotes genotype of rs1042636 showed markedly higher pulmonary artery pressure and reduced survival compared with individuals carrying the wild-type allele. The minor alleles of rs6776158, rs1048213, and rs9883099 increased CaSR expression in reporter assay. In Sugen/hypoxia pulmonary arterial hypertension rats, the point mutation replicating rs1042636 found in IPAH exacerbated pulmonary arterial hypertension severity by promoting the overexpression and the enhanced activity of CaSR.

Conclusions: Our functional genomic analysis thus indicates that the minor alleles of rs1042636, rs6776158, rs1048213, and rs9883099 contribute to the development and severity of IPAH. These findings may benefit clinical prognosis and treatment for IPAH.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1161/HYPERTENSIONAHA.121.18399DOI Listing
July 2022

Query-efficient Black-box Adversarial Attack with Customized Iteration and Sampling.

IEEE Trans Pattern Anal Mach Intell 2022 Apr 25;PP. Epub 2022 Apr 25.

It is challenging to fool an image classifier based on deep neural networks under the black-box setting where the target model can only be queried. Among existing black-box attacks, transfer-based methods tend to overfit the substitute model on parameter settings. Decision-based methods have low query efficiency due to fixed sampling and greedy search strategy. To alleviate the above problems, we present a new framework for query-efficient black-box adversarial attack by bridging transfer-based and decision-based attacks. We reveal the relationship between current noise and variance of sampling, the monotonicity of noise compression, and the influence of transition function. Guided by the new framework, we propose a black-box adversarial attack named Customized Iteration and Sampling Attack (CISA). CISA estimates the distance from nearby decision boundary to set the stepsize, and uses a dual-direction iterative trajectory to find the intermediate adversarial example. Based on the intermediate adversarial example, CISA conducts customized sampling according to the noise sensitivity of each pixel to further compress noise, and relaxes the state transition function to achieve higher query efficiency. We embed and benchmark existing adversarial attack methods under the new framework. Extensive experiments demonstrate CISA's advantage in query efficiency of black-box adversarial attacks.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TPAMI.2022.3169802DOI Listing
April 2022

GPR105-Targeted Therapy Promotes Gout Resolution as a Switch Between NETosis and Apoptosis of Neutrophils.

Front Immunol 2022 30;13:870183. Epub 2022 Mar 30.

State Key Laboratory of Natural Medicines, Key Laboratory of Drug Metabolism and Pharmacokinetics, China Pharmaceutical University, Nanjing, China.

The fate of infiltrating neutrophils in inflamed joints determines the development of acute gouty arthritis (AGA). GPR105 highly expressed in human neutrophils is sensitive to monosodium urate crystals (MSU); nevertheless, the roles of GPR105 in AGA remain unclear. Here, we show that GPR105 is significantly upregulated in peripheral polymorphonuclear neutrophils of AGA patients. GPR105 knockout (GPR105) prevented NETosis and induced apoptosis of neutrophils under MSU exposure, as well as attenuating inflammatory cascades in AGA. Mechanistically, GPR105 deletion activated cAMP-PKA signals, thereby disrupting Raf-Mek1/2-Erk1/2 pathway-mediated NADPH oxidase activation, contributing to inhibition of NETosis. Whereas, cAMP-PKA activation resulting in GPR105 deficiency modulated PI3K-Akt pathway to regulate apoptosis. More importantly, suppression of cAMP-PKA pathway by SQ22536 and H-89 restored NETosis instead of apoptosis in GPR105 neutrophils, promoting MSU-induced gout flares. Interestingly, lobetyolin was screened out as a potent GPR105 antagonist using molecular docking-based virtual screening and activity test, which efficiently attenuated MSU-induced inflammatory response interacting with GPR105. Taken together, our study implicated that modulating cell death patterns between NETosis and apoptosis through targeting GPR105 could be a potential therapeutic strategy for the treatment of AGA.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.3389/fimmu.2022.870183DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC9005804PMC
March 2022

CrabNet: Fully Task-Specific Feature Learning for One-Stage Object Detection.

IEEE Trans Image Process 2022 11;31:2962-2974. Epub 2022 Apr 11.

Object detection is usually solved by learning a deep architecture involving classification and localization tasks, where feature learning for these two tasks is shared using the same backbone model. Recent works have shown that suitable disentanglement of classification and localization tasks has the great potential to improve performance of object detection. Despite the promising performance, existing feature disentanglement methods usually suffer from two limitations. First, most of them only focus on the disentangled proposals or predication heads for classification and localization tasks after RPN. While little consideration has been given to that the features for these two different tasks actually are obtained by a shared backbone model before RPN. Second, they are suggested for two-stage objectors and are not applicable to one-stage methods. To overcome these limitations, this paper presents a novel fully task-specific feature learning method for one-stage object detection. Specifically, our method first learns disentangled features for classification and localization tasks using two separated backbone models, where auxiliary classification and localization heads are inserted at the end of the two backbone models for providing a fully task-specific features for classification and localization. Then, a feature interaction module is developed for aligning and fusing task-specific features, which are further used to produce the final detection result. Experiments on MS COCO show that our proposed method (dubbed CrabNet) can achieve clear improvement over counterparts with increasing limited inference time, while performing favorably against state-of-the-arts.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TIP.2022.3162099DOI Listing
April 2022

577 nm subthreshold micropulse laser treatment for acute central serous chorioretinopathy: a comparative study.

BMC Ophthalmol 2022 Mar 5;22(1):105. Epub 2022 Mar 5.

Wuhan Aier Eye Hanyang Hospital, Wu Han, China.

Background: To assess the efficacy of 577 nm subthreshold micropulse laser (SML) treatment for acute central serous chorioretinopathy (CSC).

Methods: This retrospective comparative case-series included 34 eyes of 34 patients with acute CSC who received either 577 nm SML treatment (SML group, n = 16 eyes) or were only monitored (observation group, n = 18 eyes). Acute CSC was defined as disease course < 3 months. Eyes with any history of treatment in the past were excluded. Data were collected over a period of 6 months. The best corrected visual acuity (BCVA), central macular thickness (CMT), and subfoveal choroidal thickness (SFCT) were observed.

Results: SML group showed significantly greater improvement in the BCVA (logMAR) compared to observation group at 1 month (0.20 ± 0.10 vs 0.30 ± 0.12, P < 0.01), 3 months (0.13 ± 0.06 vs 0.21 ± 0.06, P < 0.01) and 6 months (0.01 ± 0.06 vs 0.09 ± 0.66, P < 0.01). The CMT reduction was significantly greater in the SML group at 1 month (337.19 ± 62.96 µm vs 395.11 ± 91.30 µm, P < 0.05), 3 months (312.94 ± 49.50 µm vs 364.50 ± 70.30 µm, P < 0.05) and 6 months (291.38 ± 26.46 µm vs 348.56 ± 54.65 µm, P < 0.05). In the SML group, the SFCT did not show a significant decrease at 1 month (468.88 ± 42.19 µm, P > 0.05) but showed a significant reduction at 3 months (451.75 ± 39.36 µm, P < 0.05) and 6 months (450.50 ± 34.24 µm, P < 0.05) from baseline (489.94 ± 45.86 µm). In the observation group, there was no significant change in SFCF during follow-up. No adverse events occurred in the SML group.

Conclusions: Although some patients with acute CSC show spontaneous healing, timely intervention with 577 nm SML can shorten the disease course, improve visual acuity, and reduce the risk of chronic transformation without adverse events.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1186/s12886-022-02330-0DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC8898420PMC
March 2022

Platelet-Derived TGF (Transforming Growth Factor)-β1 Enhances the Aerobic Glycolysis of Pulmonary Arterial Smooth Muscle Cells by PKM2 (Pyruvate Kinase Muscle Isoform 2) Upregulation.

Hypertension 2022 05 2;79(5):932-945. Epub 2022 Mar 2.

Department of Pharmacology, School of Basic Medicine, Tongji Medical College of Huazhong University of Science and Technology, Wuhan, China (Y.Z., D.S., X.G., M.L., X.-B.Z., J.G., Y.W.G., L.L., R.M., Z.-Y.M.).

Background: Metabolic reprogramming is a hallmark of pulmonary arterial hypertension. Platelet activation has been implicated in pulmonary arterial hypertension (PAH), whereas the role of platelet in the pathogenesis of PAH remains unclear.

Methods: First, we explored the platelet function of semaxanib' a inhibitor of VEGF receptor (SU5416)/hypoxia mice and monocrotaline-injected rats PAH model. Then we investigated pulmonary arterial smooth muscle cell aerobic glycolysis after being treated with platelet supernatant. TGF (transforming growth factor)-βRI, pyruvate kinase muscle 2, and other antagonists were applied to identify the underlying mechanism. In addition, platelet-specific deletion TGF-β1 mice were exposed to chronic hypoxia and SU5416. Cardiopulmonary hemodynamics, vascular remodeling, and aerobic glycolysis of pulmonary arterial smooth muscle cell were determined.

Results: Here, we demonstrate that platelet-released TGF-β1 enhances the aerobic glycolysis of pulmonary arterial smooth muscle cells after platelet activation via increasing pyruvate kinase muscle 2 expression. Mechanistically, platelet-derived TGF-β1 regulate spyruvate kinase muscle 2 expression through mTOR (mammalian target of rapamycin)/c-Myc/PTBP-1(polypyrimidine tract binding protein 1)/hnRNPA-1(heterogeneous nuclear ribonucleoprotein A1) pathway. Platelet TGF-β1 deficiency mice are significantly protected from SU5416 plus chronic hypoxia-induced PAH, including attenuated increases in right ventricular systolic pressure and less pulmonary vascular remodeling. Also, in mice, pulmonary arterial smooth muscle cells showed lower glycolysis capacity and their pyruvate kinase muscle 2 expression decreased.

Conclusions: Our data demonstrate that TGF-β1 released by platelet contributes to the pathogenesis of PAH and further highlights the role of platelet in PAH.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1161/HYPERTENSIONAHA.121.18684DOI Listing
May 2022

Investigation and Identification of Food Poisoning Caused by Type B1 in Shenzhen, China.

Foodborne Pathog Dis 2022 03 28;19(3):226-231. Epub 2022 Feb 28.

Shenzhen Major Infectious Disease Control Key Laboratory, Shenzhen Center for Disease Control and Prevention, Shenzhen, China.

produces botulinum neurotoxins (BoNTs), which cause people who ingest them to become seriously ill and sometimes die. In recent years, sporadic food poisoning cases associated with have occurred across the world. In 2016, two men were admitted to our hospital in Shenzhen, China, with foodborne botulism. In this study, we report on these two typical -related food poisoning incidents and the steps taken to identify and characterize the causative pathogen. We characterized the bacterial pathogen isolated from the first patient using cooked meat medium and egg yolk agar bacterial cultures under anaerobic conditions, and morphologically identified the isolate using Gram staining. The bioassay results in mice showed that the minimum lethal dose of the BoNTs produced by our isolate was 0.001-0.0001 mg/mL (LD50 of the culture was estimated to be 1.5812 mg/kg). Whole genome sequencing (WGS) results showed that the isolate was identified as B1 Okra. The causative strain was successfully isolated from the intestinal lavage fluid collected from the initial patient.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1089/fpd.2021.0060DOI Listing
March 2022

Construction of a rabbit model with vinorelbine administration via peripherally inserted central catheter and dynamic monitoring of changes in phlebitis and thrombosis.

Exp Ther Med 2022 Mar 11;23(3):212. Epub 2022 Jan 11.

Nursing Faculty, School of Medicine, Jinhua Polytechnic, Jinhua, Zhejiang 321007, P.R. China.

Peripherally inserted central catheters (PICCs) are used for the administration of chemotherapy drugs, including vinorelbine. The present study aimed to construct a rabbit model with vinorelbine administration via PICC, and to dynamically monitor the formation of phlebitis and thrombosis. PICC was inserted into 48 rabbits following specific clinical procedures. The rabbits were randomly divided (n=6 per group) into the following eight groups: i) Control (PICC in place for 1 day); ii) 2nd day of PICC placement (received the first cycle of vinorelbine administration); iii) 3rd day of PICC placement; iv) 7th day of PICC placement; v) 14th day of PICC placement; vi) 21st day of PICC placement; vii) 23rd day of PICC placement (received the second cycle of vinorelbine administration); and viii) 24th day of PICC placement. Hematoxylin and eosin staining was performed on catheter, ear vein and anterior vena specimens. Prothrombin time was measured using an automatic coagulation analyzer, followed by routine blood tests. Serum levels of inflammation- and thrombosis-related factors, including C-reactive protein, D-dimer, interleukin-2, interleukin-6, P-selectin and E-selectin, were measured using ELISAs. X-ray examination confirmed that the rabbit model with vinorelbine administration via PICC was successfully constructed. On the 1st and 23rd day of PICC placement, thrombosis was observed in the catheter. Furthermore, on the 1st day of PICC placement, thrombosis was clearly observed in the ear vein and anterior vena samples. After vinorelbine administration, phlebitis occurred in the ear vein and anterior vena cava samples. With increasing time after vinorelbine administration via PICC, thrombosis and phlebitis were notably ameliorated. Moreover, on the day of vinorelbine administration, prothrombin time was significantly decreased and the serum levels of inflammation- and thrombosis-related factors were significantly increased compared with previous days. Collectively, the present study observed the formation and specific evolution of phlebitis and venous thrombosis after vinorelbine administration, providing a reference for the early prediction, timely prevention and treatment of PICC-related chemotherapy complications.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.3892/etm.2022.11135DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC8796649PMC
March 2022

Development and multi-center clinical trials of an up-converting phosphor technology-based point-of-care (UPT-POCT) assay for rapid COVID-19 diagnosis and prediction of protective effects.

BMC Microbiol 2022 02 3;22(1):42. Epub 2022 Feb 3.

State Key Laboratory of Pathogen and Biosecurity, Beijing Institute of Microbiology and Epidemiology, Beijing, 100071, People's Republic of China.

Background: Quantitative point-of-care testing assay for detecting antibodies is critical to COVID-19 control. In this study, we established an up-conversion phosphor technology-based point-of-care testing (UPT-POCT), a lateral flow assay, for rapid COVID-19 diagnosis, as well as prediction of seral neutralizing antibody (NAb) activity and protective effects.

Methods: UPT-POCT was developed targeting total antibodies against the receptor-binding domain (RBD) of SARS-CoV-2 spike protein. Using ELISA as a contrast method, we evaluated the quantitation accuracy with NAb and serum samples. Cutoff for serum samples was determined through 70 healthy and 140 COVID-19 patients. We evaluated the cross-reactions with antibodies against other viruses. Then, we performed multi-center clinical trials of UPT-POCT, including 782 patients with 387 clinically confirmed COVID-19 cases. Furthermore, RBD-specific antibody levels were detected using UPT-POCT and microneutralization assay for samples from both patients and vaccinees. Specifically, the antibodies of recovered patients with recurrent positive (RP) reverse transcriptase-polymerase chain reaction test results were discussed.

Results: The ratios of signal intensities between the test and control bands on the lateral flow strip, namely, T/C ratios, was defined as the results of UPT-POCT. T/C ratios had excellent correlations with concentrations of NAb, as well as OD values of ELISA for serum samples. The sensitivity and specificity of UPT-POCT were 89.15% and 99.75% for 782 cases in seven hospitals in China, respectively. We evaluated RBD-specific antibodies for 528 seral samples from 213 recovered and 99 RP COVID-19 patients, along with 35 seral samples from inactivated SARS-CoV-2 vaccinees, and we discovered that the total RBD-specific antibody level indicated by T/C ratios of UPT-POCT was significantly related to the NAb titers in both COVID-19 patients (r = 0.9404, n = 527; ρ = 0.6836, n = 528) and the vaccinees (r = 0.9063, ρ = 0.7642, n = 35), and it was highly relevant to the protection rate against RP (r = 0.9886, n = 312).

Conclusion: This study reveals that the UPT-POCT for quantitative detection of total RBD-specific antibody could be employed as a surrogate method for rapid COVID-19 diagnosis and prediction of protective effects.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1186/s12866-022-02450-zDOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC8811342PMC
February 2022

Isolation of From COVID-19 Swab Kits.

Front Microbiol 2021 4;12:799150. Epub 2022 Jan 4.

Department of Public Health Laboratory Sciences, School of Public Health, University of South China, Hengyang, China.

To investigate and characterize the putative contaminant isolated from throat and anal swab samples of patients from three fever epidemic clusters, which were not COVID-19 related, in Shenzhen, China, during COVID-19 pandemic. Bacteria were cultured from throat ( = 28) and anal ( = 3) swab samples from 28 fever adolescent patients. The isolated bacterial strains were identified using matrix-assisted laser desorption/ionization time of flight mass spectrometry (MALDI-TOF/MS) and the VITEK2 automated identification system. Nucleic acids were extracted from the patient samples ( = 31), unopened virus collection kits from the same manufacturer as the patient samples ( = 35, blank samples) and from unopened throat swab collection kits of two other manufacturers ( = 22, control samples). Metagenomic sequencing and quantitative real-time PCR (qPCR) detection were performed. Blood serum collected from patients ( = 13) was assessed for the presence of antibodies to . The genomic characteristics, antibiotic susceptibility, and heat resistance of isolates ( = 31) were analyzed. The isolates were identified by MALDI-TOF/MS and VITEK2 as . DNA sequence analysis confirmed isolates to be . The patients' samples and blank samples were positive for . Control samples were negative for . The sera from a sub-sample of 13 patients were antibody-negative for isolated . Most of the isolates were highly homologous and carried multiple β-lactamase genes ( , , and ). The isolates displayed resistance to nitrofurans, penicillins, and most β-lactam drugs. The bacteria survived heating at 56°C for 30 min. The unopened commercial virus collection kits from the same manufacturer as those used to swab patients were contaminated with . Patients were not infected with and the causative agent for the fevers remains unidentified. The relevant authorities were swiftly notified of this discovery and subsequent collection kits were not contaminated. DNA sequence-based techniques are the definitive method for species identification. The isolates were multidrug-resistant, with partial heat resistance, making them difficult to eradicate from contaminated surfaces. Such resistance indicates that more attention should be paid to disinfection protocols, especially in hospitals, to avoid outbreaks of infection.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.3389/fmicb.2021.799150DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC8763855PMC
January 2022

Interaction-Aware Graph Neural Networks for Fault Diagnosis of Complex Industrial Processes.

IEEE Trans Neural Netw Learn Syst 2021 Dec 17;PP. Epub 2021 Dec 17.

Fault diagnosis of complex industrial processes becomes a challenging task due to various fault patterns in sensor signals and complex interactions between different units. However, how to explore the interactions and integrate with sensor signals remains an open question. Considering that the sensor signals and their interactions in an industrial process with the form of nodes and edges can be represented as a graph, this article proposes a novel interaction-aware and data fusion method for fault diagnosis of complex industrial processes, named interaction-aware graph neural networks (IAGNNs). First, to describe the complex interactions in an industrial process, the sensor signals are transformed into a heterogeneous graph with multiple edge types, and the edge weights are learned by the attention mechanism, adaptively. Then, multiple independent graph neural network (GNN) blocks are employed to extract the fault feature for each subgraph with one edge type. Finally, each subgraph feature is concatenated or fused by a weighted summation function to generate the final graph embedding. Therefore, the proposed method can learn multiple interactions between sensor signals and extract the fault feature from each subgraph by message passing operation of GNNs. The final fault feature contains the information from raw data and implicit interactions between sensor signals. The experimental results on the three-phase flow facility and power system (PS) demonstrate the reliable and superior performance of the proposed method for fault diagnosis of complex industrial processes.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TNNLS.2021.3132376DOI Listing
December 2021

Cardiovascular Protective Effects of Plant Polysaccharides: A Review.

Front Pharmacol 2021 18;12:783641. Epub 2021 Nov 18.

State Key Laboratory of Natural Medicines, China Pharmaceutical University, Nanjing, China.

Cardiovascular disease is a kind of heart, brain, and blood vessel injury disease by the interaction of various pathological factors. The pathogenesis of cardiovascular disease is complex with various risk factors, including abnormally elevated blood pressure, glucose, and lipid metabolism disorders, atherosclerosis, thrombosis, etc. Plant polysaccharides are a special class of natural products derived from plant resources, which have the characteristics of wide sources, diverse biological activities, and low toxicity or side effects. Many studies have shown that plant polysaccharides improve cardiovascular diseases through various mechanisms such as anti-oxidative stress, restoring the metabolism of biological macromolecules, regulating the apoptosis cascade to reduce cell apoptosis, and inhibiting inflammatory signal pathways to alleviate inflammation. This article reviews the pharmacological effects and protective mechanisms of some plant polysaccharides in modulating the cardiovascular system, which is beneficial for developing more effective drugs with low side effects for management of cardiovascular diseases.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.3389/fphar.2021.783641DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC8639026PMC
November 2021

A biocompatible dual-AIEgen system without spectral overlap for quantitation of microbial viability and monitoring of biofilm formation.

Mater Horiz 2021 06 23;8(6):1816-1824. Epub 2021 Apr 23.

Department of Chemistry, Department of Chemical and Biological Engineering, Hong Kong Branch of Chinese Nation-al Engineering Research Center for Tissue Restoration and Reconstruction, Institute of Advanced Study and Division of Life Science The Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong, China.

The lack of rapid and reliable microbial detection and sensing platforms and insufficient understanding of microbial behavior may delay precautions that could be made, which is a great threat to human life and increases the heavy financial burden on society. In this contribution, a dual-aggregation-induced emission luminogen (AIEgen) system is successfully developed for microbial imaging and metabolic status sensing. This system consists of two AIEgens (DCQA and TPE-2BA) that bear positively charged groups or boronic acid groups, providing universal microbial staining ability and specific affinity for dead microbes, respectively. Based on the distinctive fluorescence response produced by the diverse interaction of AIEgens with live or dead microbes, this dual-AIEgen system can detect all the microbes and identify their viabilities. Furthermore, the morphology and metabolic status of a sessile biofilm can also be imaged and monitored. The system exhibits rapid labelling properties that suitable for various microbes, and good biocompatibilities.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1039/d1mh00149cDOI Listing
June 2021

Genomic Epidemiology and Antimicrobial Susceptibility Profile of Enterotoxigenic From Outpatients With Diarrhea in Shenzhen, China, 2015-2020.

Front Microbiol 2021 28;12:732068. Epub 2021 Oct 28.

Shenzhen Center for Disease Control and Prevention, Shenzhen, China.

Enterotoxigenic (ETEC) is the leading cause of severe diarrhea in children and the most common cause of diarrhea in travelers. However, most ETEC infections in Shenzhen, China were from indigenous adults. In this study, we characterized 106 ETEC isolates from indigenous outpatients with diarrhea (77% were adults aged >20 years) in Shenzhen between 2015 and 2020 by whole-genome sequencing and antimicrobial susceptibility testing. Shenzhen ETEC isolates showed a remarkable high diversity, which belonged to four phylogroups (A: 71%, B1: 13%, E: 10%, and D: 6%) and 15 ETEC lineages, with L11 (25%, O159:H34/O159:H43, ST218/ST3153), novel L2/4 (21%, O6:H16, ST48), and L4 (15%, O25:H16, ST1491) being major lineages. Heat-stable toxin (ST) was most prevalent (76%, STh: 60% STp: 16%), followed by heat-labile toxin (LT, 17%) and ST + LT (7%). One or multiple colonization factors (CFs) were identified in 68 (64%) isolates, with the common CFs being CS21 (48%) and CS6 (34%). Antimicrobial resistance mutation/gene profiles of genomes were concordant with the phenotype testing results of 52 representative isolates, which revealed high resistance rate to nalidixic acid (71%), ampicillin (69%), and ampicillin/sulbactam (46%), and demonstrated that the novel L2/4 was a multidrug-resistant lineage. This study provides novel insight into the genomic epidemiology and antimicrobial susceptibility profile of ETEC infections in indigenous adults for the first time, which further improves our understanding on ETEC epidemiology and has implications for the development of vaccine and future surveillance and prevention of ETEC infections.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.3389/fmicb.2021.732068DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC8581654PMC
October 2021

Discovery and computational studies of 2-phenyl-benzoxazole acetamide derivatives as promising P2YR antagonists with anti-gout potential.

Eur J Med Chem 2022 Jan 19;227:113933. Epub 2021 Oct 19.

Department of Medicinal Chemistry, College of Pharmaceutical Sciences, Soochow University, Suzhou, 215123, China. Electronic address:

The P2Y nucleotide receptor, a subtype of P2Y receptors, is implicated in many human inflammatory diseases. Based on the identification of favorable residues of two screening hits in the almost symmetrical P2Y binding domain, we describe the structural optimization of previously identified virtual screening hits 6 and 7 that result in the development of P2YR antagonists with a novel 2-phenyl-benzoxazole acetamide chemical scaffold. Notably, compound 52 showed potent P2YR antagonistic activity (IC = 2 nM), and a stronger inhibitory effect on MSU-induced inflammatory in vitro, better than a previously described P2YR antagonist PPTN. In vivo evaluation demonstrated that compound 52 also had satisfactory inhibitory activity on the inflammatory response of gout flares in mice. Moreover, P2YR antagonist 52 decreased paw swelling and inflammatory cell infiltration through cAMP/NLRP3/GSDMD signaling pathways in MSU-induced acute gouty arthritis mice. The discussions on the binding mechanism that employ MM/GBSA free energy calculations/decompositions also provide some useful clues for further structural designing of compound 52. Taken together, 2-phenyl-benzoxazole acetamide derivative 52 with potent P2YR antagonistic activity and in vivo potency could be a promising strategy for gout therapy and deserves further optimization.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.ejmech.2021.113933DOI Listing
January 2022

Detection and Tracking Meet Drones Challenge.

IEEE Trans Pattern Anal Mach Intell 2021 Oct 14;PP. Epub 2021 Oct 14.

Drones, or general UAVs, equipped with cameras have been fast deployed with a wide range of applications, including agriculture, aerial photography, and surveillance. Consequently, automatic understanding of visual data collected from drones becomes highly demanding, bringing computer vision and drones more and more closely. To promote and track the developments of object detection and tracking algorithms, we have organized three challenge workshops in conjunction with ECCV 2018, ICCV 2019 and ECCV 2020, attracting more than 100 teams around the world. We provide a large-scale drone captured dataset, VisDrone, which includes four tracks, i.e., (1) image object detection, (2) video object detection, (3) single object tracking, and (4) multi-object tracking. We first present a thorough review of object detection and tracking datasets and benchmarks, and discuss the challenges of collecting large-scale drone-based object detection and tracking datasets with fully manual annotations. Being the largest such dataset ever published, VisDrone enables extensive evaluation and investigation of visual analysis algorithms for the drone platform. We provide a detailed analysis of the current state of the field of large-scale object detection and tracking on drones, and conclude the challenge as well as propose future directions.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TPAMI.2021.3119563DOI Listing
October 2021

Peptide Blocking Self-Polymerization of Extracellular Calcium-Sensing Receptor Attenuates Hypoxia-Induced Pulmonary Hypertension.

Hypertension 2021 11 27;78(5):1605-1616. Epub 2021 Sep 27.

From the Department of Pathophysiology, School of Basic Medicine (R.X., S.L., T.Z., L.Z., Q.H.).

[Figure: see text].
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1161/HYPERTENSIONAHA.120.16712DOI Listing
November 2021

VPsero: Rapid Serotyping of Using Serogroup-Specific Genes Based on Whole-Genome Sequencing Data.

Front Microbiol 2021 2;12:620224. Epub 2021 Sep 2.

BGI-Shenzhen, Shenzhen, China.

has emerged as a significant enteropathogen in human and marine habitats worldwide, notably in regions where aquaculture products constitute a major nutritional source. It is a growing cause of diseases including gastroenteritis, wound infections, and septicemia. Serotyping assays use commercially available antisera to identify strains, but this approach is limited by high costs, complicated procedures, cross-immunoreactivity, and often subjective interpretation. By leveraging high-throughput sequencing technologies, we developed an method based on comparison of gene clusters for lipopolysaccharide (LPSgc) and capsular polysaccharide (CPSgc) by firstly using the unique-gene strategy. The algorithm, VPsero, which exploits serogroup-specific genes as markers, covers 43 K and all 12 O serogroups in serotyping assays. VPsero is capable of predicting serotypes from assembled draft genomes, outputting LPSgc/CPSgc sequences, and recognizing possible novel serogroups or populations. Our tool displays high specificity and sensitivity in prediction toward strains, with an average sensitivity in serogroup prediction of 0.910 for O and 0.961 for K serogroups and a corresponding average specificity of 0.990 for O and 0.998 for K serogroups.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.3389/fmicb.2021.620224DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC8443796PMC
September 2021

Enhancing Geometric Factors in Model Learning and Inference for Object Detection and Instance Segmentation.

IEEE Trans Cybern 2022 Aug 19;52(8):8574-8586. Epub 2022 Jul 19.

Deep learning-based object detection and instance segmentation have achieved unprecedented progress. In this article, we propose complete-IoU (CIoU) loss and Cluster-NMS for enhancing geometric factors in both bounding-box regression and nonmaximum suppression (NMS), leading to notable gains of average precision (AP) and average recall (AR), without the sacrifice of inference efficiency. In particular, we consider three geometric factors, that is: 1) overlap area; 2) normalized central-point distance; and 3) aspect ratio, which are crucial for measuring bounding-box regression in object detection and instance segmentation. The three geometric factors are then incorporated into CIoU loss for better distinguishing difficult regression cases. The training of deep models using CIoU loss results in consistent AP and AR improvements in comparison to widely adopted l -norm loss and IoU-based loss. Furthermore, we propose Cluster-NMS, where NMS during inference is done by implicitly clustering detected boxes and usually requires fewer iterations. Cluster-NMS is very efficient due to its pure GPU implementation, and geometric factors can be incorporated to improve both AP and AR. In the experiments, CIoU loss and Cluster-NMS have been applied to state-of-the-art instance segmentation (e.g., YOLACT and BlendMask-RT), and object detection (e.g., YOLO v3, SSD, and Faster R-CNN) models. Taking YOLACT on MS COCO as an example, our method achieves performance gains as +1.7 AP and +6.2 AR for object detection, and +1.1 AP and +3.5 AR for instance segmentation, with 27.1 FPS on one NVIDIA GTX 1080Ti GPU. All the source code and trained models are available at https://github.com/Zzh-tju/CIoU.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TCYB.2021.3095305DOI Listing
August 2022

Molecular epidemiology of Bordetella pertussis and analysis of vaccine antigen genes from clinical isolates from Shenzhen, China.

Ann Clin Microbiol Antimicrob 2021 Aug 18;20(1):53. Epub 2021 Aug 18.

Shenzhen Center for Disease Control and Prevention, 8 Longyuan Road, Nanshan District, Shenzhen, China.

Background: Although pertussis cases globally have been controlled through the Expanded Programme on Immunization (EPI), the incidence of pertussis has increased significantly in recent years, with a "resurgence" of pertussis occurring in developed countries with high immunization coverage. Attracted by its fast-developing economy, the population of Shenzhen has reached 14 million and has become one of the top five largest cities by population size in China. The incidence of pertussis here was about 2.02/100,000, far exceeding that of the whole province and the whole country (both < 1/100,000). There are increasing numbers of reports demonstrating variation in Bordetella pertussis antigens and genes, which may be associated with the increased incidence. Fifty strains of Bordetella pertussis isolated from 387 suspected cases were collected in Shenzhen in 2018 for genotypic and molecular epidemiological analysis.

Methods: There were 387 suspected cases of pertussis enrolled at surveillance sites in Shenzhen from June to August 2018. Nasopharyngeal swabs from suspected pertussis cases were collected for bacterial culture and the identity of putative Bordetella pertussis isolates was confirmed by real-time PCR. The immunization history of each patient was taken. The acellular pertussis vaccine (APV) antigen genes for pertussis toxin (ptxA, ptxC), pertactin (prn) and fimbriae (fim2 and fim3) together with the pertussis toxin promoter region (ptxP) were analyzed by second-generation sequencing. Genetic and phylogenetic analysis was performed using sequences publicly available from GenBank, National Institutes of Health, Bethesda, MD, USA ( https://www.ncbi.nlm.nih.gov/genbank/ ). The antimicrobial susceptibility was test by Kirby-Bauer disk diffusion.

Results: Fifty strains of Bordetella pertussis were successfully isolated from nasopharyngeal swabs of 387 suspected cases, with a positivity rate of 16.79%, including 28 males and 22 females, accounting for 56.0% and 44.0% respectively. Thirty-eight of the 50 (76%) patients were found to be positive for B. pertussis by culture. Among the positive cases with a history of vaccination, 30 of 42 (71.4%) cases had an incomplete pertussis vaccination history according to the national recommendation. Three phylogenetic groups (PG1-PG3) were identified each containing a predominant genotype. The two vaccines strains, CS and Tohama I, were distantly related to these three groups. Thirty-one out of fifty (62%) isolates belonged to genotype PG1, with the allelic profile prn2/ptxC2/ptxP3/ptxA1/fim3-1/fim2-1. Eighteen out of fifty (36%) isolates contained the A2047G mutation and were highly resistant to erythromycin, and all belonged to genotype PG3 (prn1/ptxA1/ptxP1/ptxC1/fim3-1/fim2-1), which is closely related to the recent epidemic strains found in northern China.

Conclusions: The positive rate of cases under one-year-old was significantly higher than that of other age groups and should be monitored. The dominant antigen genotypes of 50 Shenzhen isolates are closely related to the epidemic strains in the United States, Australia and many countries in Europe. Despite high rates of immunization with APV, epidemics of pertussis have recently occurred in these countries. Therefore, genomic analysis of circulating isolates of B. pertussis should be continued, for it will benefit the control of whooping cough and development of improved vaccines and therapeutic strategies.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1186/s12941-021-00458-3DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC8371876PMC
August 2021

Coarse-to-Fine: Progressive Knowledge Transfer-Based Multitask Convolutional Neural Network for Intelligent Large-Scale Fault Diagnosis.

IEEE Trans Neural Netw Learn Syst 2021 Aug 9;PP. Epub 2021 Aug 9.

In modern industry, large-scale fault diagnosis of complex systems is emerging and becoming increasingly important. Most deep learning-based methods perform well on small number of fault diagnosis, but cannot converge to satisfactory results when handling large-scale fault diagnosis because the huge number of fault types will lead to the problems of intra/inter-class distance unbalance and poor local minima in neural networks. To address the above problems, a progressive knowledge transfer-based multitask convolutional neural network (PKT-MCNN) is proposed. First, to construct the coarse-to-fine knowledge structure intelligently, a structure learning algorithm is proposed via clustering fault types in different coarse-grained nodes. Thus, the intra/inter-class distance unbalance problem can be mitigated by spreading similar tasks into different nodes. Then, an MCNN architecture is designed to learn the coarse and fine-grained task simultaneously and extract more general fault information, thereby pushing the algorithm away from poor local minima. Last but not least, a PKT algorithm is proposed, which can not only transfer the coarse-grained knowledge to the fine-grained task and further alleviate the intra/inter-class distance unbalance in feature space, but also regulate different learning stages by adjusting the attention weight to each task progressively. To verify the effectiveness of the proposed method, a dataset of a nuclear power system with 66 fault types was collected and analyzed. The results demonstrate that the proposed method can be a promising tool for large-scale fault diagnosis.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TNNLS.2021.3100928DOI Listing
August 2021

Enhanced mPGES-1 Contributes to PD-Related Peritoneal Fibrosis via Activation of the NLRP3 Inflammasome.

Front Med (Lausanne) 2021 18;8:675363. Epub 2021 May 18.

Department of Nephrology, Shunde Hospital, Southern Medical University (The First People's Hospital of Shunde), Foshan, China.

Microsomal prostaglandin E synthase-1 (mPGES-1)-derived prostaglandin E (PGE2) is a chief mediator of inflammation. However, the role and mechanism of mPGES-1 in peritoneal dialysis (PD)-associated peritoneal fibrosis have not been investigated. In PD patients, mPGES-1 expression in peritoneum tissues and the levels of PGE2, IL-1β, and IL-18 in the dialysate were examined. In rat peritoneal mesothelial cells (RPMCs), the regulation and function of mPGES-1 and NLRP3 inflammasome were investigated. The expression of extracellular matrix proteins and the components of NLRP3 inflammasome were detected by Western blotting or real-time quantitative PCR. In PD patients with ultrafiltration failure (UFF), mPGES-1 was enhanced in the peritoneum, which was associated with the degree of peritoneal fibrosis. Accordingly, the intraperitoneal PGE2 levels were also positively related to the PD duration, serum C-reactive protein levels, and serum creatinine levels in incident PD patients. In RPMCs, high-glucose treatment significantly induced mPGES-1 expression and PGE2 secretion without affecting the expressions of mPGES-2 and cPGES. Inhibition of mPGES-1 via short hairpin RNA significantly ameliorated the expression of extracellular matrix proteins of RPMCs induced by high glucose. Additionally, high glucose markedly activated NLRP3 inflammasome in RPMCs that was blunted by mPGES-1 inhibition. Furthermore, silencing NLRP3 with siRNA significantly abrogated the expression of extracellular matrix proteins in RPMCs treated with high glucose. Finally, we observed increased IL-1β and IL-18 levels in the dialysate of incident PD patients, showing a positive correlation with PGE2. These data demonstrate that mPGES-1-derived PGE2 plays a critical role in PD-associated peritoneal fibrosis through activation of the NLRP3 inflammasome. Targeting mPGES-1 may offer a novel strategy to treat peritoneal fibrosis during PD.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.3389/fmed.2021.675363DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC8167893PMC
May 2021
-->