Search

BackgroundEarly diagnosis of esophageal squamous cell carcinoma (ESCC) is critical for effective treatment and optimal prognosis; however, less study on serum biomarkers for the early ESCC detection has been reported. The aim of this study was to identify and evaluate several serum autoantibody biomarkers in early ESCC.MethodsWe initially screened candidate tumor-associated autoantibodies (TAAbs) associated with ESCC by serological proteome analysis (SERPA) combined with nanoliter-liquid chromatography combined with quadrupole time of flight tandem mass spectrometry (nano-LC-Q-TOF-MS/MS), and the TAAbs were further subjected to analysis by Enzyme-linked immunosorbent assay (ELISA) in a clinical cohort (386 participants, including 161 patients with ESCC, 49 patients with high-grade intraepithelial neoplasia [HGIN] and 176 healthy controls [HC]). Receiver operating characteristic (ROC) curve was plotted to evaluate the diagnostic performance.ResultsThe serum levels of CETN2 and POFUT1 autoantibodies which were identified by SERPA were statistically different between ESCC or HGIN patients and HC in ELISA analysis with the area under the curve (AUC) values of 0.709 (95%CI: 0.654-0.764) and 0.741 (95%CI: 0.689-0.793), 0.717 (95%CI: 0.634-0.800) and 0.703 (95%CI: 0.627-0.779) for detection of ESCC and HGIN, respectively. Combining these two markers, the AUCs were 0.781 (95%CI: 0.733-0.829), 0.754 (95%CI: 0.694-0.814) and 0.756 (95%CI: 0.686-0.827) when distinguishing ESCC, early ESCC and HGIN from HC, respectively. Meanwhile, the expression of CETN2 and POFUT1 was found to be correlated with ESCC progression.ConclusionsOur data suggest that CETN2 and POFUT1 autoantibodies have potential diagnostic value for ESCC and HGIN, which may provide novel insights for early ESCC and precancerous lesions detection.

N-ary facts composed of a primary triple (head entity, relation, tail entity) and an arbitrary number of auxiliary attribute-value pairs, are prevalent in real-world knowledge graphs (KGs). Link prediction on n-ary facts is to predict a missing element in an n-ary fact. This helps populate and enrich KGs and further promotes numerous downstream applications. Previous studies usually require a substantial amount of high-quality data to understand the elements in n-ary facts. However, these studies overlook few-shot relations, which have limited labeled instances, yet are common in real-world scenarios. Thus, this paper introduces a new task, few-shot link prediction on n-ary facts. It aims to predict a missing entity in an n-ary fact with limited labeled instances. We further propose a model for Few-shot Link prEdict on N-ary facts, thus called FLEN, which consists of three modules: the relation learning, support-specific adjusting, and query inference modules. FLEN captures relation meta information from limited instances to predict a missing entity in a query instance. To validate the effectiveness of FLEN, we construct three datasets based on existing benchmark data. Our experimental results show that FLEN significantly outperforms existing related models in both few-shot link prediction on n-ary facts and binary facts., Comment: 12 pages, submitted to IEEE for possible publication

Details of participating hospitals. (PDF 120 kb)

Event Causality Identification (ECI) aims to identify causal relations between events in unstructured texts. This is a very challenging task, because causal relations are usually expressed by implicit associations between events. Existing methods usually capture such associations by directly modeling the texts with pre-trained language models, which underestimate two kinds of semantic structures vital to the ECI task, namely, event-centric structure and event-associated structure. The former includes important semantic elements related to the events to describe them more precisely, while the latter contains semantic paths between two events to provide possible supports for ECI. In this paper, we study the implicit associations between events by modeling the above explicit semantic structures, and propose a Semantic Structure Integration model (SemSIn). It utilizes a GNN-based event aggregator to integrate the event-centric structure information, and employs an LSTM-based path aggregator to capture the event-associated structure information between two events. Experimental results on three widely used datasets show that SemSIn achieves significant improvements over baseline methods., Comment: Accepted at ACL 2023

The coronavirus disease 2019 (COVID-19) pandemic caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has occasioned worldwide alarm. Globally, the number of reported confirmed cases has exceeded 84.3 million as of this writing (January 2, 2021). Since there are no targeted therapies for COVID-19, the current focus is the repurposing of drugs approved for other uses. In some clinical trials, antiviral drugs such as remdesivir (Grein et al., 2020), lopinavir/ritonavir (LPV/r) (Cao et al., 2020), chloroquine (Gao et al., 2020), hydroxychloroquine (Gautret et al., 2020), arbidol (Wang et al., 2020), and favipiravir (Cai et al., 2020b) have shown efficacy in COVID-19 patients. LPV/r combined with arbidol, which is the basic regimen in some regional hospitals in China including Zhejiiang Province, has shown antiviral effects in COVID-19 patients (Guo et al., 2020; Xu et al., 2020). A retrospective cohort study also reported that this combination therapy showed better efficacy than LPV/r alone for the treatment of COVID-19 patients (Deng et al., 2020).

Linezolid is an oxazolidinone antibacterial drug, and its therapeutic drug monitoring and individualized treatment have been challenged since its approval. With the in-depth clinical research of linezolid, we have changed our attitude toward its therapeutic drug monitoring and our view of individualized treatment. On the basis of summarizing the existing clinical studies, and based on the practical experience of each expert in their respective professional fields, we have formed this expert consensus. Our team of specialists is a multidisciplinary team that includes pharmacotherapists, clinical pharmacology specialists, critical care medicine specialists, respiratory specialists, infectious disease specialists, emergency medicine specialists and more. We are committed to the safe and effective use of linezolid in patients in need, and the promotion of its therapeutic drug monitoring.

BackgroundTight junction (TJ) imbalance is associated with hepatocellular carcinoma (HCC). MarvelD3, which contains a conserved MARVEL (MAL and related proteins for vesicle trafficking and membrane link) domain similarly to occludin and tricellulin, is a recently identified integral membrane protein that forms TJs. However, little is known about the possible roles of marvelD3 in epithelial–mesenchymal transition (EMT) and metastasis of HCC. We aimed to demonstrate the role of marvelD3 in inhibiting HCC EMT and migration and further explore the underlying molecular mechanisms.MethodsMarvlD3 expression was assessed in HCC and normal liver tissues. Changes in marvelD3 expression were analyzed during transforming growth factor β1 (TGF-β1) and snail/slug-induced EMT. MarvelD3 knockdown HCC cell lines were established using marvelD3-siRNA to analyze correlations between marvelD3 and EMT-related proteins. Tumor cell behaviors were analyzed in marvelD3 knockdown HCC cells. Associations between marvelD3 and genes in the nuclear factor (NF)-κB pathway were also analyzed.ResultsLoss of marvelD3 expression was significantly correlated with the occurrence and TNM stage of HCC. MarvelD3 was downregulated in HCC cells with TGF-β and snail/slug-induced EMT. Expression of marvelD3 protein was significantly associated with E-cadherin and vimentin in HCC cell lines. Knockdown of marvelD3 promoted tumor cell migration concomitant with activation of the NF-κB signaling pathway and increased matrix metallopeptidase 9 expression.ConclusionsOur study demonstrated that MarvelD3 inhibited EMT and migration of HCC cells via NF-κB signaling pathway. This suggests that marvelD3 is a novel potential biomarker for the treatment and prognosis of HCC.

This study sought to discuss the correlation between the third-generation cephalosporins (3GC)-resistant Escherichia coli and Klebsiella pneumoniae and antibiotic consumption intensity from 143 Chinese tertiary hospitals in 2014. With a retrospective design, the correlation between antibiotic consumption and 3GC-resistant E. coli and K. pneumoniae were performed. 3GC-resistant E. coli was significantly correlated with the consumption of all antibiotics (r = 0.252, p

Acute kidney injury (AKI) caused by sepsis is a serious disease which mitochondrial oxidative stress and inflammatory play a key role in its pathophysiology. Ceria nanoparticles hold strong and recyclable reactive oxygen species (ROS)-scavenging activity, have been applied to treat ROS-related diseases. However, ceria nanoparticles can't selectively target mitochondria and the ultra-small ceria nanoparticles are easily agglomerated. To overcome these shortcomings and improve therapeutic efficiency, we designed an ROS-responsive nano-drug delivery system combining mitochondria-targeting ceria nanoparticles with atorvastatin for acute kidney injury. Methods: Ceria nanoparticles were modified with triphenylphosphine (TCeria NPs), followed by coating with ROS-responsive organic polymer (mPEG-TK-PLGA) and loaded atorvastatin (Atv/PTP-TCeria NPs). The physicochemical properties, in vitro drug release profiles, mitochondria-targeting ability, in vitro antioxidant, anti-apoptotic activity and in vivo treatment efficacy of Atv/PTP-TCeria NPs were examined. Results: Atv/PTP-TCeria NPs could accumulate in kidneys and hold a great ability to ROS-responsively release drug and TCeria NPs could target mitochondria to eliminate excessive ROS. In vitro study suggested Atv/PTP-TCeria NPs exhibited superior antioxidant and anti-apoptotic activity. In vivo study showed that Atv/PTP-TCeria NPs effectively decreased oxidative stress and inflammatory, could protect the mitochondrial structure, reduced apoptosis of tubular cell and tubular necrosis in the sepsis-induced AKI mice model. Conclusions: This ROS-responsive nano-drug delivery system combining mitochondria-targeting ceria nanoparticles with atorvastatin has favorable potentials in the sepsis-induced AKI therapy.

Greenland is the biggest island in the world, and the area of snow and ice is the second largest in the world, which is only smaller than Antarctic ice sheet. Studying the temperature change of the island is important for ice sheet melting, especially the summer temperature. In this study, we reconstructed monthly temperature data over past 10,000 years using ice core data, meteorological observation data and European reanalysis data, and linear regression equation– residual correction method was used. And the results are as follows: The temperature of September to May showed a significant increasing trend at the 0.05 level during period 9700 B. C. E. ~ 2019 C.E., and the increasing rate is 0.0017 ~ 0.0058 ℃ every two decades. However, the temperature of June to August showed a significant decreasing trend, and the decreasing rate is -0.0035 ~ -0.0040 ℃ every two decades. The climate change can be divided into four increasing periods, namely, period 9700 B. C. E. ~ 9600 B.C.E, period 9400 B. C. E. ~ 7720 B.C.E, period 6240 B. C. E. ~ 6000 B.C.E, and 1800 C.E. ~ 2019 C.E., and three decreasing periods, namely, period 9600 B. C. E. ~ 9400 B.C.E, period 6340 B. C. E. ~ 6240 B.C.E. and 6000 B.C.E. ~ 1800 C.E., and one stable periods, namely, 7720 B.C.E. ~ 6340 B.C.E. And the differences of temperature change rate of different altitudes of the same period and month is not obvious. And the temperature change rate in Winter is larger than that in Summer. The dataset can provide data basis for the study of Greenland ice sheet mass balance.

In this paper, we propose a relative pose estimation algorithm for micro-lens array (MLA)-based conventional light field (LF) cameras. First, by employing the matched LF-point pairs, we establish the LF-point-LF-point correspondence model to represent the correlation between LF features of the same 3D scene point in a pair of LFs. Then, we employ the proposed correspondence model to estimate the relative camera pose, which includes a linear solution and a non-linear optimization on manifold. Unlike prior related algorithms, which estimated relative poses based on the recovered depths of scene points, we adopt the estimated disparities to avoid the inaccuracy in recovering depths due to the ultra-small baseline between sub-aperture images of LF cameras. Experimental results on both simulated and real scene data have demonstrated the effectiveness of the proposed algorithm compared with classical as well as state-of-art relative pose estimation algorithms.

A Temporal Knowledge Graph (TKG) is a sequence of KGs corresponding to different timestamps. TKG reasoning aims to predict potential facts in the future given the historical KG sequences. One key of this task is to mine and understand evolutional patterns of facts from these sequences. The evolutional patterns are complex in two aspects, length-diversity and time-variability. Existing models for TKG reasoning focus on modeling fact sequences of a fixed length, which cannot discover complex evolutional patterns that vary in length. Furthermore, these models are all trained offline, which cannot well adapt to the changes of evolutional patterns from then on. Thus, we propose a new model, called Complex Evolutional Network (CEN), which uses a length-aware Convolutional Neural Network (CNN) to handle evolutional patterns of different lengths via an easy-to-difficult curriculum learning strategy. Besides, we propose to learn the model under the online setting so that it can adapt to the changes of evolutional patterns over time. Extensive experiments demonstrate that CEN obtains substantial performance improvement under both the traditional offline and the proposed online settings., Comment: ACL 2022 main conference

Hand-held light field (LF) cameras have unique advantages in computer vision such as 3D scene reconstruction and depth estimation. However, the related applications are limited by the ultra-small baseline, e.g., leading to the extremely low depth resolution in reconstruction. To solve this problem, we propose to rectify LF to obtain a large baseline. Specifically, the proposed method aligns two LFs captured by two hand-held LF cameras with a random relative pose, and extracts the corresponding row-aligned sub-aperture images (SAIs) to obtain an LF with a large baseline. For an accurate rectification, a method for pose estimation is also proposed, where the relative rotation and translation between the two LF cameras are estimated. The proposed pose estimation minimizes the degree of freedom (DoF) in the LF-point-LF-point correspondence model and explicitly solves this model in a linear way. The proposed pose estimation outperforms the state-of-the-art algorithms by providing more accurate results to support rectification. The significantly improved depth resolution in 3D reconstruction demonstrates the effectiveness of the proposed LF rectification.

Script is a kind of structured knowledge extracted from texts, which contains a sequence of events. Based on such knowledge, script event prediction aims to predict the subsequent event. To do so, two aspects should be considered for events, namely, event description (i.e., what the events should contain) and event encoding (i.e., how they should be encoded). Most existing methods describe an event by a verb together with only a few core arguments (i.e., subject, object, and indirect object), which are not precise. In addition, existing event encoders are limited to a fixed number of arguments, which are not flexible to deal with extra information. Thus, in this paper, we propose the Rich Event Prediction (REP) framework for script event prediction. Fundamentally, it is based on the proposed rich event description, which enriches the existing ones with three kinds of important information, namely, the senses of verbs, extra semantic roles, and types of participants. REP contains an event extractor to extract such information from texts. Based on the extracted rich information, a predictor then selects the most probable subsequent event. The core component of the predictor is a transformer-based event encoder to flexibly deal with an arbitrary number of arguments. Experimental results on the widely used Gigaword Corpus show the effectiveness of the proposed framework., Comment: AAAI 2023 (main conference)

In this paper we propose a generative adversarial network (GAN) framework to enhance the perceptual quality of compressed videos. Our framework includes attention and adaptation to different quantization parameters (QPs) in a single model. The attention module exploits global receptive fields that can capture and align long-range correlations between consecutive frames, which can be beneficial for enhancing perceptual quality of videos. The frame to be enhanced is fed into the deep network together with its neighboring frames, and in the first stage features at different depths are extracted. Then extracted features are fed into attention blocks to explore global temporal correlations, followed by a series of upsampling and convolution layers. Finally, the resulting features are processed by the QP-conditional adaptation module which leverages the corresponding QP information. In this way, a single model can be used to enhance adaptively to various QPs without requiring multiple models specific for every QP value, while having similar performance. Experimental results demonstrate the superior performance of the proposed PeQuENet compared with the state-of-the-art compressed video quality enhancement algorithms.

Accurately calibrating an unfocused plenoptic camera is essential to its applications. Rapid progress has been made in this area in the past decades. In this paper, detailed analysis is first performed toward the state-of-the-art projection model. Based on the analysis, parameters in the projection model are divided into two groups. Then, based on the parameter analysis, a new, to the best of our knowledge, form of the projection model, together with a new image feature light field structure point (LF-structure-point), is proposed. The LF-structure-point provides a completely non-redundant representation of the signal structure of the recorded light field raw data and induces a virtual space, “light field structure space,” which is related to the real physical space by a 3D-to-3D projective transformation. The extracting algorithm of the LF-structure-point is also presented. Finally, based on the 3D-to-3D projective transformation and parameter grouping, a simple two-step calibration method using the LF-structure-point as the input data is then proposed and achieves satisfactory experimental results.

The mutations in modifier genes may contribute to some inherited diseases including Wilson disease (WD). This study was designed to identify potential modifier genes that contribute to WD. A total of 10 WD patients with single or no heterozygous ATP7B mutations were recruited for whole-exome sequencing (WES). Five hundred and thirteen candidate genes, of which the genetic variants present in at least two patients, were identified. In order to clarify which proteins might be involved in copper transfer or metabolism processes, the isobaric tags for relative and absolute quantitation (iTRAQ) was performed to identify the differentially expressed proteins between normal and CuSO4-treated cell lines. Thirteen genes/proteins were identified by both WES and iTRAQ, indicating that disease-causing variants of these genes may actually contribute to the aberrant copper ion accumulation. Additionally, the c.86C > T (p.S29L) mutation in the SLC31A2 gene (coding CTR2) has a relative higher frequency in our cohort of WD patients (6/191) than reported (0.0024 in gnomAD database) in our healthy donors (0/109), and CTR2S29L leads to increased intracellular Cu concentration and Cu-induced apoptosis in cultured cell lines. In conclusion, the WES and iTRAQ approaches successfully identified several disease-causing variants in potential modifier genes that may be involved in the WD phenotype.

Objective The study conducted a multicenter study in China to explore the learning curve of contrast enhanced ultrasound (CEUS) for sentinel lymph nodes (SLNs), the feasibility of using this technique for the localization of SLNs and lymphatic channels (LCs) and its diagnostic performance for lymph node metastasis. Method Nine hundred two patients with early invasive breast cancer from six tertiary class hospitals in China were enrolled between December 2016 and December 2019. Each patient received general ultrasound scanning and SLN-CEUS before surgery. The locations and sizes of LCs and SLNs were marked on the body surface based on observations from SLN-CEUS. These body surface markers were then compared with intraoperative blue staining in terms of their locations. The first 40 patients from each center were included in determining the learning curve of SLN-CEUS across sites. The remaining patients were used to investigate the diagnostic efficacy of this technique in comparison with intraoperative blue staining and pathology respectively. Result The ultrasound doctor can master SLN-CEUS after 25 cases, and the mean operating time is 22.5 min. The sensitivity, specificity, negative predictive value, and positive predictive value of SLN-CEUS in diagnosing lymph node metastases were 86.47, 89.81, 74.90, and 94.97% respectively. Conclusion Ultrasound doctors can master SLN-CEUS with a suitable learning curve. SLN-CEUS is a feasible and useful approach to locate SLNs and LCs before surgery and it is helpful for diagnosing LN metastases.

Oxidative stress is an important pathological mechanism for acute kidney injury (AKI). SS-31, as a mitochondria-targeted peptide with strong antioxidant activity, is a good candidate for the treatment of AKI. However, an efficient treatment of AKI requires frequent administration of SS-31, which is due to its poor specific biodistribution and low delivery efficiency. To overcome these deficiencies, we designed pH-responsive and AKI-kidney targeted nanopolyplexes (NPs) for effectively delivering SS-31, which is new frontier for formulation of HA and CS. NPs are electrostatically complexed using anionic hyaluronic acid and cationic chitosan as materials, which could increase the accumulation in injured areas and uptake into CD44-overexpressed cells. Electrostatic balance of NPs is broken in low pH environment of lysosomes to allow SS-31 releasing and subsequently targeting to mitochondria to represent therapeutic effect. In vitro studies indicate that NPs exhibited higher antioxidative and antiapoptotic effects as compared with free SS-31. AKI mouse model suggests that NPs have significantly higher therapeutic efficiency than bare SS-31. It was found that NPs had excellent ability to decrease oxidative stress, protect mitochondrial structure, reduce inflammatory response, reduce apoptosis and necrosis of tubular cells after intravenious administration. Overall, the results suggest that the NPs have significant potential to enhance the specific biodistribution and delivery of SS-31, therefore have good effects on reducing oxidative stress and inflammation, preventing tubular apoptosis and necrosis. We believe NPs are effective delivery system for AKI treatment in clinical application.

Multiple drug resistance and the increase in the appearance of superbugs together with the exceedingly scant development of new potent antibiotic drugs pose an urgent global medical threat and imminent public security crisis. In the present study, we fabricated well-dispersed tocopherol polyethylene glycol succinate (TPGS)-capped silver nanoparticles (AgNPs) of about 10 nm in size. The hollow structure of the TPGS-capped AgNPs (TPGS/AgNPs) was confirmed and applied to load antibiotics. The TPGS/AgNPs proved to be able to cross the bacterial cell wall and penetrate into bacteria, thereby delivering more of the antibiotic to the interior of bacteria and thus enhancing the in vitro antibacterial effect of the antibiotic, even overcoming the drug-resistance in drug-resistant E. coli and Acinetobacter baumannii. It was found that the TPGS modification in the TPGS/AgNPs could decrease the activity of the efflux pumps AdeABC and AdeIJK in drug-resistant Acinetobacter baumannii via inhibiting the efflux pump genes adeB and adeJ, thus increasing the accumulation of the delivered antibiotic and overcoming the drug-resistance. Tigecycline delivered by TPGS/AgNPs could effectively antagonize drug-resistance in an acute peritonitis model mice, thereby increasing the survival rate and alleviating the inflammatory response. TPGS/AgNPs were developed as a novel and effective antibiotic delivery system and TPGS was demonstrated to have great potential as a pharmaceutical excipient for use in drug-resistant infection therapy.

Effective treatment for acute lung injury (ALI) is in high demand. Lung-targeted ternary nanoparticles containing anti-intercellular adhesion molecule-1 (ICAM-1) antibody-conjugated simvastatin-loaded nanostructured lipid carrier (ICAM/NLC), protamine (Pro), and angiopoietin-1 (Ang-1) gene (ICAM-NLC/Pro/Ang) were developed for ALI therapy. The ternary nanoparticles with different weight ratios of ICAM-NLC to Ang-1 gene were prepared via charge interaction. The anti-ICAM-1 antibody-conjugated ternary nanoparticles exhibited higher cellular uptake and transfection efficiency (from 26.7% to 30.9%) in human vascular endothelial cell line EAhy926 than the non-targeted control. The largest size of ICAM-NLC/Pro/Ang (357.1 nm) was employed for further study, which significantly up-regulated in vitro and in vivo Ang-1 protein expression. In vivo i.v. administration of ICAM-NLC/Pro/Ang (357.1 nm) significantly attenuated pulmonary TNF-α and IL-6 levels, inflammatory cell infiltration, and led to positive histological improvements in lipopolysaccharide-induced ALI mice. Collectively, the ICAM-NLC/Pro/Ang that co-delivered simvastatin and Ang-1 gene may represent a potential treatment modality for ALI.

The prevalence of antibiotic resistance and lack of alternative drugs have posed an increasing threat to public health. Here, we prepared β-Ga2O3:Cr3+ nanoparticles modified with ICAM1-antibody-conjugated TPGS (I-TPGS/Ga2O3) as a novel antibiotic carrier for the treatment of drug-resistant infections. Methods: I-TPGS/Ga2O3 were firstly characterized by measuring particle size, morphology, crystal structure, drug loading capacity, and in vitro drug release behaviors. The in vitro antibacterial activities of I-TPGS/Ga2O3/TIG were evaluated using standard and drug-resistant bacteria. The internalization of I-TPGS/Ga2O3 was observed by fluorescence confocal imaging, and the expression levels of the efflux pump genes of TRKP were analyzed by real-time RT-PCR. In vitro cellular uptake and in vivo biodistribution study were performed to investigate the targeting specificity of I-TPGS/Ga2O3 using HUEVC and acute pneumonia mice, respectively. The in vivo anti-infective efficacy and biosafety of I-TPGS/Ga2O3/TIG were finally evaluated using acute pneumonia mice. Results: It was found that TPGS could down-regulate the over-expression of the efflux pump genes, thus decreasing the efflux pump activity of bacteria. I-TPGS/Ga2O3 with small particle size and uniform distribution facilitated their internalization in bacteria, and the TPGS modification resulted in a significant reduction in the efflux of loaded antibiotics. These properties rendered the encapsulated tigecycline to exert a stronger antibacterial activity both in vitro and in vivo. Additionally, targeted delivery of I-TPGS/Ga2O3 mediated by ICAM1 antibodies contributed to a safe and effective therapy. Conclusion: It is of great value to apply I-TPGS/Ga2O3 as a novel and effective antibiotic delivery system for the treatment of drug-resistant infections.

Point clouds are widely used as representations of 3D objects and scenes in a number of applications, including virtual and mixed reality, autonomous driving, antiques reconstruction. To reduce the cost for transmitting and storing such data, this paper proposes an end-to-end learning-based point cloud attribute compression (PCAC) approach. The proposed network adopts a sparse convolution-based variational autoencoder (VAE) structure to compress the color attribute of point clouds. Considering the difficulty of stacked convolution operations in capturing long range dependencies, the attention mechanism is incorporated in which a non-local attention module is developed to capture the local and global correlations in both spatial and channel dimensions. Towards the practical application, an additional modulation network is offered to achieve the variable rate compression purpose in a single network, avoiding the memory cost of storing multiple networks for multiple bitrates. Our proposed method achieves state-of-the-art compression performance compared to other existing learning-based methods and further reduces the gap with the latest MPEG G-PCC reference software TMC13 version 14.

Colorectal cancer (CRC) is common worldwide. Aldehyde dehydrogenase 1B1 (ALDH1B1), a member of the ALDH1 family, serves as a biomarker for cancer stem cells. We hypothesized that ALDH1B1 expression is associated with colorectal tumors. Immunohistochemistry was used to detect ALDH1B1 expression across a commercial colorectal tissue microarray. The signal intensities of the positively stained tissues were expressed using the mean integrated optical density (mean IOD). We also analyzed ALDH1B1 mRNA expression in the Oncomine database. The associations between ALDH1B1 expression and CRC stage and prognosis were then evaluated using the web-based tools, GEPIA and UALCAN. Analysis of the tissue microarray revealed that the expression of ALDH1B1 was significantly higher in colorectal adenomas and colorectal adenocarcinoma (IOD/area values=0.117±0.070 and 0.168±0.0168, respectively) compared with normal and cancer-adjacent tissues (IOD/area values=0.051±0.028 and 0.068±0.053). For samples collected in the hospital, ALDH1B1 was highly expressed in the adenoma (IOD/area=0.103±0.054) and CRC (IOD/area=0.116±0.059) tissues compared with the cancer-adjacent tissues (IOD/area=0.066±0.024, p0.05). The expression of ALDH1B1 in tissues from two resources was not found to be significantly associated with CRC stage. In Oncomine, ALDH1B1 mRNA expression was increased in the colorectal tumor tissues compared with the normal colorectal tissues (p=0.024) and its expression was independent of CRC stage and prognosis (p0.05). Thus, while the protein and mRNA expression of ALDH1B1 suggests that it is a potential marker of colorectal tumors, its expression is independent of CRC stage and prognosis.

Recent years have witnessed the significant development of learning-based video compression methods, which aim at optimizing objective or perceptual quality and bit rates. In this paper, we introduce deep video compression with perceptual optimizations (DVC-P), which aims at increasing perceptual quality of decoded videos. Our proposed DVC-P is based on Deep Video Compression (DVC) network, but improves it with perceptual optimizations. Specifically, a discriminator network and a mixed loss are employed to help our network trade off among distortion, perception and rate. Furthermore, nearest-neighbor interpolation is used to eliminate checkerboard artifacts which can appear in sequences encoded with DVC frameworks. Thanks to these two improvements, the perceptual quality of decoded sequences is improved. Experimental results demonstrate that, compared with the baseline DVC, our proposed method can generate videos with higher perceptual quality achieving 12.27% reduction in a perceptual BD-rate equivalent, on average.

Scripts are structured sequences of events together with the participants, which are extracted from the texts.Script event prediction aims to predict the subsequent event given the historical events in the script. Two kinds of information facilitate this task, namely, the event-level information and the script-level information. At the event level, existing studies view an event as a verb with its participants, while neglecting other useful properties, such as the state of the participants. At the script level, most existing studies only consider a single event sequence corresponding to one common protagonist. In this paper, we propose a Transformer-based model, called MCPredictor, which integrates deep event-level and script-level information for script event prediction. At the event level, MCPredictor utilizes the rich information in the text to obtain more comprehensive event semantic representations. At the script-level, it considers multiple event sequences corresponding to different participants of the subsequent event. The experimental results on the widely-used New York Times corpus demonstrate the effectiveness and superiority of the proposed model., Comment: EMNLP 2021 long paper (main conference)

Knowledge Graph (KG) reasoning that predicts missing facts for incomplete KGs has been widely explored. However, reasoning over Temporal KG (TKG) that predicts facts in the future is still far from resolved. The key to predict future facts is to thoroughly understand the historical facts. A TKG is actually a sequence of KGs corresponding to different timestamps, where all concurrent facts in each KG exhibit structural dependencies and temporally adjacent facts carry informative sequential patterns. To capture these properties effectively and efficiently, we propose a novel Recurrent Evolution network based on Graph Convolution Network (GCN), called RE-GCN, which learns the evolutional representations of entities and relations at each timestamp by modeling the KG sequence recurrently. Specifically, for the evolution unit, a relation-aware GCN is leveraged to capture the structural dependencies within the KG at each timestamp. In order to capture the sequential patterns of all facts in parallel, the historical KG sequence is modeled auto-regressively by the gate recurrent components. Moreover, the static properties of entities such as entity types, are also incorporated via a static graph constraint component to obtain better entity representations. Fact prediction at future timestamps can then be realized based on the evolutional entity and relation representations. Extensive experiments demonstrate that the RE-GCN model obtains substantial performance and efficiency improvement for the temporal reasoning tasks on six benchmark datasets. Especially, it achieves up to 11.46\% improvement in MRR for entity prediction with up to 82 times speedup comparing to the state-of-the-art baseline., Comment: SIGIR 2021 Full Paper

Knowledge Graph Question Answering aims to automatically answer natural language questions via well-structured relation information between entities stored in knowledge graphs. When faced with a complex question with compositional semantics, query graph generation is a practical semantic parsing-based method. But existing works rely on heuristic rules with limited coverage, making them impractical on more complex questions. This paper proposes a Director-Actor-Critic framework to overcome these challenges. Through options over a Markov Decision Process, query graph generation is formulated as a hierarchical decision problem. The Director determines which types of triples the query graph needs, the Actor generates corresponding triples by choosing nodes and edges, and the Critic calculates the semantic similarity between the generated triples and the given questions. Moreover, to train from weak supervision, we base the framework on hierarchical Reinforcement Learning with intrinsic motivation. To accelerate the training process, we pre-train the Critic with high-reward trajectories generated by hand-crafted rules, and leverage curriculum learning to gradually increase the complexity of questions during query graph generation. Extensive experiments conducted over widely-used benchmark datasets demonstrate the effectiveness of the proposed framework.

Severe and critically ill patients with coronavirus disease 2019 (COVID-19) were usually with underlying diseases, which led to the problems of complicated drug use, potential drug-drug interactions and medication errors in special patients. Based on Diagnosis and treatment of novel coronavirus pneumonia ( trial version 6), and Management of COVID-19: the Zhejiang experience, we summarized the experience in the use of antiviral drugs, corticosteroids, vascular active drugs, antibacterial, probiotics, nutrition support schemes in severe and critically ill COVID-19 patients. It is also suggested to focus on medication management for evaluation of drug efficacy and duration of treatment, prevention and treatment of adverse drug reactions, identification of potential drug-drug interactions, individualized medication monitoring based on biosafety protection, and medication administration for special patients.

Coronavirus disease 2019 (COVID-19) caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is now a global outbreak of disease. The antiviral treatment acts as one of the most important means of SARS-CoV-2 infection. Alteration of physiological characteristics in special populations may lead to the change in drug pharmacokinetics, which may result in treatment failure or increased adverse drug reactions. Some potential drugs have shown antiviral effects on SARS-CoV-2 infections, such as chloroquine, hydroxychloroquine, favipiravir, lopinavir/ritonavir, arbidol, interferon alpha, and remedsivir. Here, we reviewed the literature on clinical effects in COVID-19 patients of these antiviral agents and provided the potential antiviral agent options for pregnant women, elderly patients, liver or renal dysfunction patients, and severe or critically ill patients receiving renal replacement therapy or ECMO after SARS-CoV-2 infection.

Background The purpose of the study is to discuss the correlation between the resistance rate of gram negative bacteria to fluoroquinolones (FQ) and antibiotic consumption intensity of 145 China tertiary hospitals in 2014. Methods This retrospective study adopted national monitoring data from 2014. Each participating hospital required to report annual consumption of each antibiotic, and the resistance rate of gram negative bacteria to FQ. Then the correlation between antibiotic usage and fluoroquinolones –resistant (FQR) rate was consequently investigated. Results One hundred forty-five hospitals were included in the study, and the median antibiotic consumption intensity was 46.30 (23.93–115.39) defined daily dosages (DDDs) per 100 patient-days. Cephalosporins ranks first in the antibiotics consumption, followed by fluoroquinolones, penicillins, and carbapenems. Fluoroquinolones resistance rate varied from hospital to hospital. The correlation analysis showed significant relationship between the percentage of FQR Escherichia coli and the consumption of FQs (r = 0.308, ppKlebsiella pneumoniae, not only FQs (r = 0.291, pppPseudomonas aeruginosa was observed to be correlated with the consumption of all antibiotics (r = 0.260, ppAcinetobacter baumannii was significantly correlated with the consumption of all antibiotics (r = 0.282, pppA. baumannii and the antibiotics consumption was not found. Conclusions A strong correlation was demonstrated between the antibiotic consumption and the rates of FQR gram-negative bacteria. As unreasonable antibiotics usage remains crucial in the proceeding of resistant bacteria selection, our study could greatly promote the avoidance of unnecessary antibiotic usage.

Background: Colorectal cancer (CRC) comprises a large proportion of malignant tumors, and early detection of CRC is critical for effective treatment and optimal prognosis. We aimed to discover and validate serum autoantibodies for early detection of CRC. Methods: Combined with CRC-associated autoantibodies discovered by serological proteome and multiplex analyses, 26 predefined autoantibodies were evaluated in 315 samples (130 CRCs, 75 advanced adenomas, and 110 healthy controls) by protein microarray analysis. Autoantibodies with potential detection value were verified by enzyme-linked immunosorbent assays (ELISAs). Receiver operating characteristic (ROC) curve analysis was conducted to evaluate the accuracy of the biomarkers. Results: Four serum autoantibodies (ALDH1B1, UQCRC1, CTAG1, and CENPF) showed statistically different levels between patients with advanced neoplasm (CRC or advanced adenoma) and controls in microarray analysis, which were validated by ELISAs. Among the four biomarkers, the ALDH1B1 autoantibody showed the highest detection value with area under the curve (AUC) values of 0.70 and 0.74 to detect CRC and advanced adenoma with sensitivities of 75.68 and 62.31% and specificities of 63.06 and 73.87%, respectively. By combining the four biomarkers, the performance was improved with an AUC of 0.79 to detect CRC and advanced adenomas. Conclusion: The ALDH1B1 autoantibody has a good potential for early detection of CRC and advanced adenoma, and measuring serum autoantibodies against tumor-associated antigens may improve detection of early CRC.

An SS31 prodrug improved AKI therapy via rapid distribution in the kidney and long-term retention in the renal tubule., The development of drugs with rapid distribution in the kidney and long-term retention in the renal tubule is a breakthrough for enhanced treatment of acute kidney injury (AKI). Here, l-serine–modified chitosan (SC) was synthesized as a potential AKI kidney–targeting agent due to the native cationic property of chitosan and specific interaction between kidney injury molecule–1 (Kim-1) and serine. Results indicated that SC was rapidly accumulated and long-term retained in ischemia-reperfusion–induced AKI kidneys, especially in renal tubules, which was possibly due to the specific interactions between SC and Kim-1. SC-TK-SS31 was then prepared by conjugating SS31, a mitochondria-targeted antioxidant, to SC via reactive oxygen species (ROS)–sensitive thioketal linker. Because of the effective renal distribution combined with ROS-responsive drug release behavior, the administration of SC-TK-SS31 led to an enhanced therapeutic effect of SS31 by protecting mitochondria from damage and reducing the oxidative stress, inflammation, and cell apoptosis.

This study sought to discuss the correlation between the third-generation cephalosporins (3GC)-resistant Escherichia coli and Klebsiella pneumoniae and antibiotic consumption intensity from 143 Chinese tertiary hospitals in 2014. With a retrospective design, the correlation between antibiotic consumption and 3GC-resistant E. coli and K. pneumoniae were performed. 3GC-resistant E. coli was significantly correlated with the consumption of all antibiotics (r = 0.252, p 0.01), β-Lactams antibiotics (r = 0.313, p 0.01), β-Lactams excluding combinations with β-lactamase inhibitors (r = 0.365, p 0.01), cephalosporin (r = 0.398, p 0.01), cephalosporins excluding combinations with β-lactamase inhibitors (r = 0.374, p 0.01), 3GC (r = 0.321, p 0.01), and 3GC excluding combinations with β-lactamase inhibitors (r = 0.343, p 0.01). 3GC-resistant K. pneumoniae was significantly correlated with the consumption of all antibiotics (r = 0.200, p 0.05), β-Lactams antibiotics (r = 0.232, p 0.01), cephalosporin (r = 0.215, p 0.05), 3GC (r = 0.383, p 0.01), 3GC excluding combinations with β-lactamase inhibitors (r = 0.245, p 0.01), and β-lactam-β-lactamase inhibitor combinations (r = 0.218, p 0.05). There was a significant relationship between the antibiotic consumption and the rates of 3GC-resistant E. coli and K. pneumoniae. Clinicians should grasp the indication of antibiotics use to reduce the production of drug-resistant bacteria.

Event coreference resolution aims to classify all event mentions that refer to the same real-world event into the same group, which is necessary to information aggregation and many downstream applications. To resolve event coreference, existing methods usually calculate the similarities between event mentions and between specific kinds of event arguments. However, they fail to accurately identify paraphrase relations between events and may suffer from error propagation while extracting event components (i.e., event mentions and their arguments). Therefore, we propose a new model based on Event-specific Paraphrases and Argument-aware Semantic Embeddings, thus called EPASE, for event coreference resolution. EPASE recognizes deep paraphrase relations in an event-specific context of sentences and can cover event paraphrases of more situations, bringing about a better generalization. Additionally, the embeddings of argument roles are encoded into event embedding without relying on a fixed number and type of arguments, which results in the better scalability of EPASE. Experiments on both within- and cross-document event coreference demonstrate its consistent and significant superiority compared to existing methods.

Knowledge inference on knowledge graph has attracted extensive attention, which aims to find out connotative valid facts in knowledge graph and is very helpful for improving the performance of many downstream applications. However, researchers have mainly poured attention to knowledge inference on binary facts. The studies on n-ary facts are relatively scarcer, although they are also ubiquitous in the real world. Therefore, this paper addresses knowledge inference on n-ary facts. We represent each n-ary fact as a primary triple coupled with a set of its auxiliary descriptive attribute-value pair(s). We further propose a neural network model, NeuInfer, for knowledge inference on n-ary facts. Besides handling the common task to infer an unknown element in a whole fact, NeuInfer can cope with a new type of task, flexible knowledge inference. It aims to infer an unknown element in a partial fact consisting of the primary triple coupled with any number of its auxiliary description(s). Experimental results demonstrate the remarkable superiority of NeuInfer.