Search

Abstract Background The metabolic patterns of human placental-derived mesenchymal stem cell (hP-MSC) treatment for primary sclerosing cholangitis (PSC) remain unclear, and therapeutic effects significantly vary due to individual differences. Therefore, it is crucial to investigate the serological response to hP-MSC transplantation through small molecular metabolites and identify easily detectable markers for efficacy evaluation. Methods Using Mdr2−/− mice as a PSC model and Mdr2+/+ mice as controls, the efficacy of hP-MSC treatment was assessed based on liver pathology, liver enzymes, and inflammatory factors. Serum samples were collected for 12C-/13C-dansylation and DmPA labeling LC–MS analysis to investigate changes in metabolic pathways after hP-MSC treatment. Key metabolites and regulatory enzymes were validated by qRT-PCR and Western blotting. Potential biomarkers of hP-MSC efficacy were identified through correlation analysis and machine learning. Results Collectively, the results of the liver histology, serum liver enzyme levels, and inflammatory factors supported the therapeutic efficacy of hP-MSC treatment. Based on significant differences, 41 differentially expressed metabolites were initially identified; these were enriched in bile acid, lipid, and hydroxyproline metabolism. After treatment, bile acid transport was accelerated, whereas bile acid production was reduced; unsaturated fatty acid synthesis was upregulated overall, with increased FADS2 and elongase expression and enhanced fatty acid β-oxidation; hepatic proline 4-hydroxylase expression was decreased, leading to reduced hydroxyproline production. Correlation analysis of liver enzymes and metabolites, combined with time trends, identified eight potential biomarkers: 2-aminomuconate semialdehyde, l-1-pyrroline-3-hydroxy-5-carboxylic acid, l-isoglutamine, and maleamic acid were more abundant in model mice but decreased after hP-MSC treatment. Conversely, 15-methylpalmitic, eicosenoic, nonadecanoic, and octadecanoic acids were less abundant in model mice but increased after hP-MSC treatment. Conclusions This study revealed metabolic regulatory changes in PSC model mice after hP-MSC treatment and identified eight promising biomarkers, providing preclinical evidence to support therapeutic applications of hP-MSC.

Abstract Organoids are miniature, highly accurate representations of organs that capture the structure and unique functions of specific organs. Although the field of organoids has experienced exponential growth, driven by advances in artificial intelligence, gene editing, and bioinstrumentation, a comprehensive and accurate overview of organoid applications remains necessary. This review offers a detailed exploration of the historical origins and characteristics of various organoid types, their applications—including disease modeling, drug toxicity and efficacy assessments, precision medicine, and regenerative medicine—as well as the current challenges and future directions of organoid research. Organoids have proven instrumental in elucidating genetic cell fate in hereditary diseases, infectious diseases, metabolic disorders, and malignancies, as well as in the study of processes such as embryonic development, molecular mechanisms, and host–microbe interactions. Furthermore, the integration of organoid technology with artificial intelligence and microfluidics has significantly advanced large‐scale, rapid, and cost‐effective drug toxicity and efficacy assessments, thereby propelling progress in precision medicine. Finally, with the advent of high‐performance materials, three‐dimensional printing technology, and gene editing, organoids are also gaining prominence in the field of regenerative medicine. Our insights and predictions aim to provide valuable guidance to current researchers and to support the continued advancement of this rapidly developing field.

Query execution time prediction is essential for database query optimization tasks, such as query scheduling, progress monitoring, and resource allocation. In the query execution time prediction tasks, the query plan is often used as the modeling object of a prediction model. Although the learning-based prediction models have been proposed to capture plan features, there are two limitations need to be considered more. First, the parent–child dependencies between plan operators can be captured, but the operator’s branch independence cannot be distinguished. Second, each operator’s output row is its following operator input, but the data iterate transfer operations between operators are ignored. In this study, we propose a graph query execution time prediction model containing a plan module, a query module, a plan-query module, and a prediction module to improve prediction effectiveness. Specifically, the plan module is used to capture the data iterate transfer operations and distinguish independent of branch operators; the query module is used to learn features of query terms that have an influence on the composition of operators; the plan-query interaction module is used to learn the logical correlations of plan and query. The experiment on datasets proves the effectiveness of the operator iterate-aware and query-plan interaction method in our proposed graph query execution prediction model.

The early and accurate identification of predictive biomarkers for antiviral treatment efficacy remains a significant clinical challenge, particularly in the management of chronic hepatitis B (CHB). This study aimed to assess whether the plasma metabolome could reliably predict the success of antiviral therapy in CHB patients. We conducted a retrospective analysis on 56 treatment-naive CHB patients at the First Affiliated Hospital of Zhejiang University from December 2013 to March 2016. Patients who underwent a 48-week treatment regimen of entecavir (ETV) and interferon-alpha (IFN-α) were randomly assigned to either a discovery cohort (n=29) or a validation cohort (n=27). Based on the outcome of the treatment, patients were classified as HBeAg seroconversion group (High responders, Hrp) or the non-remission group (Low responder, Lrp). Our methodology involved an untargeted analysis of the amine/phenol and carboxylic acid submetabolomes in the CHB patients under treatment, utilizing chemical isotope labeling (CIL) techniques with liquid chromatography-mass spectrometry (LC-MS). Several metabolites were identified as having significant diagnostic potential for distinguishing Hrp from Lrp, with areas under the receiver operating characteristic curve (AUC) exceeding those typical clinical indicators. Notably, four metabolites, namely 2-methyl-3-ketovaleric acid, 2-ketohexanoic acid, 6-oxo-1,4,5,6-tetrahydronicotinic acid, and α-ketoisovaleric acid, demonstrated exceptionally high sensitivity and specificity in both cohorts, nearing 100%. In contrast, the clinical indicators, including HBcAb, log(HBsAg), and HBeAb, demonstrated lower and inconsistent sensitivity and specificity between the discovery and validation cohorts. Using HBcAb as a marker, the sensitivity was 87.5% with 76.9% specificity in the discovery cohort; however, the sensitivity dropped to 46.7% with 91.7% specificity in the validation cohort. Using log(HBsAg), the sensitivity was 84.6% with 69.2% specificity in the discovery cohort, compared to 85.7% sensitivity and 83.3% specificity in the validation cohort. For HBeAb, the separation of Hrp and Lrp had a sensitivity of 87.5% with 69.2% specificity in the discovery cohort, while the validation cohort showed 86.7% sensitivity and 91.7% specificity.

Abstract The research and application of bearing fault diagnosis techniques are crucial for enhancing equipment reliability, extending bearing lifespan, and reducing maintenance expenses. Nevertheless, most existing methods encounter challenges in discriminating between signals from machines operating under normal and faulty conditions, leading to unstable detection results. To tackle this issue, the present study proposes a novel approach for bearing fault detection based on graph neural networks and ensemble learning. Our key contribution is a novel stochasticity-based compositional method that transforms Euclidean-structured data into a graph format for processing by graph neural networks, with feature fusion and a newly proposed ensemble learning strategy for outlier detection specifically designed for bearing fault diagnosis. This approach marks a significant advancement in accurately identifying bearing faults, highlighting our study's pivotal role in enhancing diagnostic methodologies.

Focused on the problem that the stream computing platform was suffering from the high energy consumption and low efficiency due to the lack of consideration for energy efficiency in designing process, an energy-efficient optimization strategy based on elastic data migration in big data streaming platform (EEDM-BDSP) was proposed.Firstly, models of the load prediction and the resource judgment were set up, and the load prediction algorithm was designed, which predicted the load tendency and determine node resource occupancy, so as to find nodes of resource overload and redundancy.Secondly, models of the resource constraint and the optimal data migration were set up, and the optimal data migration algorithm was proposed, which data migration for the purpose of improving node resource utilization.Finally, model of the energy consumption was set up to calculate the energy consumption saved by the cluster after data migration.The experimental results show that the EEDM-BDSP changes node resources in the cluster can responded on time, the resource utilization and the energy-efficient are improved.

Query execution time estimation is an essential task for databases, accurate estimation results can help administrators to manage and monitor systems. This study proposes an interaction-aware and dependency-aware query execution time estimation approach that utilizes graph neural networks to capture dependence and interaction relationships. We divide graph query execution time estimation tasks into three stages: workload generation and running, graph-based feature modeling and representation, training and estimation. Specifically, we generate query workloads and run them to collect the database and plan information when queries are executed. Then, the collected plan and database components are modeled into vertexes, the interaction and dependency between them are modeled into edges of graph-based feature representation. We develop an estimation model based on graph neural networks, in which the vertex embedding network is proposed to deal with the vertex heterogeneity, and the message passing network is proposed to aggregate the local representation into the global representation to obtain an embedding that can represent the higher-order feature information of the whole graph, and the estimation network is proposed to estimate execution times. The experiment results on datasets show that our estimation approach can improve estimation quality and outperform other estimation approaches in terms of estimation accuracy.

Abstract This study explores the analysis and modeling of energy consumption in the context of database workloads, aiming to develop an eco-friendly database management system (DBMS). It leverages vibration energy harvesting systems with self-sustaining wireless vibration sensors (WVSs) in combination with the least square support vector machine algorithm to establish an energy consumption model (ECM) for relational database workloads. Through experiments, the performance of self-sustaining WVS in providing power is validated, and the accuracy of the proposed ECM during the execution of Structured Query Language (SQL) statements is evaluated. The findings demonstrate that this approach can reliably predict the energy consumption of database workloads, with a maximum prediction error rate of 10% during SQL statement execution. Furthermore, the ECM developed for relational databases closely approximates actual energy consumption for query operations, with errors ranging from 1 to 4%. In most cases, the predictions are conservative, falling below the actual values. This finding underscores the high predictive accuracy of the ECM in anticipating relational database workloads and their associated energy consumption. Additionally, this paper delves into prediction accuracy under different types of operations and reveals that ECM excels in single-block read operations, outperforming multi-block read operations. ECM exhibits substantial accuracy in predicting energy consumption for SQL statements in sequential and random read modes, especially in specialized database management system environments, where the error rate for the sequential read model is lower. In comparison to alternative models, the proposed ECM offers superior precision. Furthermore, a noticeable correlation between model error and the volume of data processed by SQL statements is observed. In summary, the relational database ECM introduced in this paper provides accurate predictions of workload and database energy consumption, offering a theoretical foundation and practical guidance for the development of eco-friendly DBMS.

Many drug and therapeutic modalities have emerged over the past few years. However, successful commercialization is dependent on their safety and efficacy evaluations. Several preclinical models are available for drug-screening and safety evaluations, including cellular- and molecular-level models, tissue and organoid models, and animal models. Organoids are three-dimensional cell cultures derived from primary tissues or stem cells that are structurally and functionally similar to the original organs and can self-renew, and they are used to establish various disease models. Human hepatobiliary organoids have been used to study the pathogenesis of diseases, such as hepatitis, liver fibrosis, hepatocellular carcinoma, primary sclerosing cholangitis and biliary tract cancer, as they retain the physiological and histological characteristics of the liver and bile ducts. Here, we review recent research progress in validating drug toxicity, drug screening and personalized therapy for hepatobiliary-related diseases using human hepatobiliary organoid models, discuss the challenges encountered in current research and evaluate the possible solutions.

Focused on the problem that the planned routes cannot meet the personalized demand of different users in route planning of personalized self-driving tour, a keyword-aware optimal route planning method based on different user interests was proposed.Firstly, the road network information preprocessing model was set up and the road network information query graph was built by the road network information preprocessing algorithm.Secondly, the inverted index algorithm was proposed to prune the road network information query graph according to the personalized requirements from users, which improved the execution efficiency of keyword-aware optimal route planning method and reduced the memory cost of large-scale data processing effectively.Finally, the keyword-aware optimal route planning algorithm was proposed to realize personalized recommendation according to user interest by bidirectional parallel extension.The experimental results show that the method not only realizes the route planning to meet the individual needs of users but also improves the execution efficiency of the method through pruning and bidirectional parallel extension.

Abstract Background Mesenchymal stem cell (MSC) treatment plays a major role in the management of acute lung injury (ALI), and neutrophils are the initial line of defense against ALI. However, the effect of MSCs on neutrophils in ALI remains mostly unknown. Methods We investigated the characteristics of neutrophils in lung tissue of ALI mice induced by lipopolysaccharide after treatment with MSCs using single-cell RNA sequencing. Neutrophils separated from lung tissue in ALI were co-cultured with MSCs, and then samples were collected for reverse transcription-polymerase chain reaction and flow cytometry. Results During inflammation, six clusters of neutrophils were identified, annotated as activated, aged, and circulatory neutrophils. Activated neutrophils had higher chemotaxis, reactive oxygen species (ROS) production, and nicotinamide adenine dinucleotide phosphate (NADPH) oxidase scores than aged neutrophils. Circulatory neutrophils occurred mainly in healthy tissue and were characterized by higher expression of Cxcr2 and Sell. Activated neutrophils tended to exhibit higher expression of Cxcl10 and Cd47, and lower expression of Cd24a, while aged neutrophils expressed a lower level of Cd47 and higher level of Cd24a. MSC treatment shifted activated neutrophils toward an aged neutrophil phenotype by upregulating the expression of CD24, thereby inhibiting inflammation by reducing chemotaxis, ROS production, and NADPH oxidase. Conclusion We identified the immunosuppressive effects of MSCs on the subtype distribution of neutrophils and provided new insight into the therapeutic mechanism of MSC treatment in ALI.

Primary sclerosing cholangitis (PSC) is an autoimmune cholangiopathy characterized by chronic inflammation of the biliary epithelium and periductal fibrosis, with no curative treatment available, and liver transplantation is inevitable for end-stage patients. Human placental mesenchymal stem cell (hpMSC)-derived exosomes have demonstrated the ability to prevent fibrosis, inhibit collagen production and possess immunomodulatory properties in autoimmune liver disease. Here, we prepared hpMSC-derived exosomes (ExoMSC) and further investigated the anti-fibrotic effects and detailed mechanism on PSC based on Mdr2−/− mice and multicellular organoids established from PSC patients. The results showed that ExoMSC ameliorated liver fibrosis in Mdr2−/− mice with significant collagen reduction in the preductal area where Th17 differentiation was inhibited as demonstrated by RNAseq analysis, and the percentage of CD4+IL-17A+T cells was reduced both in ExoMSC-treated Mdr2−/− mice (Mdr2−/−-Exo) in vivo and ExoMSC-treated Th17 differentiation progressed in vitro. Furthermore, ExoMSC improved the hypersecretory phenotype and intercellular interactions in the hepatic Th17 microenvironment by regulating PERK/CHOP signaling as supported by multicellular organoids. Thus, our data demonstrate the anti-fibrosis effect of ExoMSC in PSC disease by inhibiting Th17 differentiation, and ameliorating the Th17-induced microenvironment, indicating the promising potential therapeutic role of ExoMSC in liver fibrosis of PSC or Th17-related diseases.

Abstract Background Pain intensity may be varied during the needle advancing through different skin layers, injection into the intradermal layer may exclude mixed pain from deeper planes. This study aimed to investigate whether compressing a three-dimensional (3D)-printed disk against the skin may relieve pain associated with intradermal injection of local anesthetic which mimics the skin test procedure. Methods After institutional review board approval, 3D-printed disks with projections were designed for this study. Enrolled patients were randomized to receive either a disk compressing against the axillary skin during the intradermal injection of local anesthesia (compressing disk group) or an intradermal injection of local anesthesia without any compression (no compressing disk group). The primary outcomes were pain intensity (100-mm visual analog scale) and satisfaction (5-point Likert scale) as assessed by patients. Results Ninety patients with American Society of Anesthesiologists I–II physical status receiving intradermal local anesthesia prior to an ultrasound-guided axillary approach were included. Eighty-seven patients completed the study, with 44 and 43 patients in disk and no disk groups, respectively. Pain scores were significantly different (P

Abstract Outlier detection is an important topic in machine learning and has been used in a wide range of applications. Outliers are objects that are few in number and deviate from the majority of objects. As a result of these two properties, we show that outliers are susceptible to a mechanism called fluctuation. This article proposes a method called fluctuation-based outlier detection (FBOD) that achieves a low linear time complexity and detects outliers purely based on the concept of fluctuation without employing any distance, density or isolation measure. Fundamentally different from all existing methods. FBOD first converts the Euclidean structure datasets into graphs by using random links, then propagates the feature value according to the connection of the graph. Finally, by comparing the difference between the fluctuation of an object and its neighbors, FBOD determines the object with a larger difference as an outlier. The results of experiments comparing FBOD with eight state-of-the-art algorithms on eight real-worlds tabular datasets and three video datasets show that FBOD outperforms its competitors in the majority of cases and that FBOD has only 5% of the execution time of the fastest algorithm. The experiment codes are available at: https://github.com/FluctuationOD/Fluctuation-based-Outlier-Detection .

Cardinality estimation is critical for optimizing database queries, and accurate results are essential for a good query plan. Traditional models use statistical principles but struggle with complex data associations. Learning-based methods solve these problems but need to improve accuracy and reduce parameter size, and adapt to multi-table training. Therefore, we propose the Convolutional Masked Autoencoder for Distribution Estimation(ConvMADE) model, which uses the Re-parameterization Convolution(RepConv) structure, which enhances the ability of the model to obtain data features, thereby improving the accuracy of cardinality estimation. At the same time, the DepthWise Multilayer Perceptron (DWMP) structure is added to reduce the number of model parameters, and each table is explicitly trained to improve the ability to capture multi-table data features. We compare the ConvMADE model with traditional and learning-based methods on the DMV and IMDB datasets. The results show that the performance of the ConvMADE model in both single-table and multi-table models is superior to other models, and the parameter amount of the ConvMADE model is much lower than that of the baseline model. The single table can be as low as 18% of the baseline model, the multi-table can be as low as 81%, and the multi-table average q-error is 27.2% lower than the baseline model.

Hypoxia-inducible factor-2α (HIF-2α) is a transcription factor responsible for regulating genes related to angiogenesis and metabolism. This study aims to explore the effect of a previously unreported mutation c.C2473T (p.R825S) in the C-terminal transactivation domain (CTAD) of HIF-2α that we detected in tissue of patients with liver disease. We sequenced available liver and matched blood samples obtained during partial liver resection or liver transplantation performed for clinical indications including hepatocellular carcinoma and liver failure. In tandem, we constructed cell lines and a transgenic mouse model bearing the corresponding identified mutation in HIF-2α from which we extracted primary hepatocytes. Lipid accumulation was evaluated in these cells and liver tissue from the mouse model using Oil Red O staining and biochemical measurements. We identified a mutation in the CTAD of HIF-2α (c.C2473T; p.R825S) in 5 of 356 liver samples obtained from patients with hepatopathy and dyslipidemia. We found that introduction of this mutation into the mouse model led to an elevated triglyceride level, lipid droplet accumulation in liver of the mutant mice and in their extracted primary hepatocytes, and increased transcription of genes related to hepatic fatty acid transport and synthesis in the mutant compared to the control groups. In mutant mice and cells, the protein levels of nuclear HIF-2α and its target perilipin-2 (PLIN2), a lipid droplet-related gene, were also elevated. Decreased lipophagy was observed in mutant groups. Our study defines a subpopulation of dyslipidemia that is caused by this HIF-2α mutation. This may have implications for personalized treatment.

Schematic diagram of the protective effect of MSC-derived exosomes on maintaining SLC7A11 function during ferroptosis involved in CCl4-induced ALI.

Abstract Background The radial arteriovenous fistula (AVF) is a rare complication occurring after transradial cardiac catheterization. Patients with AVF typically present with signs of venous dilation, such as swelling or palpable thrills. However, neurological complications secondary to radial AVFs are rare. This paper reported a rare case of ischemic steal syndrome that occurred 11 months after the transradial cardiac catheterization, most likely as a consequence of radial arteriovenous fistula. Case presentation This paper described a case of a 73-year-old female, who complained of right forearm swelling and radial 1–3 fingers numbness for several months after the catheterized stent surgery through radial approach. Upon Clinical examination, this patient presented with a slight bump and palpable thrill at the distal third of right forearm, and the sensory of radial 1–3 fingers and pinch force was compromised. The Ultrasonography and computed tomography angiography (CTA) of the upper extremity revealed AVF between the right radial artery and the adjacent vein. Microsurgery was performed successfully to ligate the fistula and reconstruct the radial artery. The numbness has gradually improved about 1 week after the surgery, with no recurred swelling. The two-point pinch force and digital sensitivity recovered at the 20-month follow-up. What’s more, due to the scarcity of cases, the optimal therapy for iatrogenic radial AVF is controversial. Accordingly, we provided a literature review of previous reports of catheter-related radial AVFs and proposed an algorithm to manage them. Conclusions We believe that once an AVF is diagnosed, early treatment options such as compression or surgery are necessary to relieve symptoms and prevent further complications. Otherwise, serious complications can occur, including the ischemic steal syndrome.

Outlier detection is a significant research direction in machine learning and has many applications in finance, network security, and other areas. Outlier detection of Euclidean datasets is a mainstream problem in outlier detection. Most detection methods often ignore the connection of its nodes. To collect the representation information of feature sets and node connections to improve the detection of outliers in Euclidean datasets Accuracy rate, we propose a novel Graph Convolutional and Attention-Based Outlier Detection (GCA).The GCA first converts the Euclidean structure data into directed graphs using locally sensitive hashing; then, by applying a Graph Convolutional Network, the data features and their connectivity graph are fed into the neural network; secondly, it fuses the extracted features and the features reconstructed by the attention mechanism; finally, calculating the outlier factors of the objects. Comparing eight state-of-art algorithms on ten real-world datasets shows that GCA achieves the highest Area Under ROC Curve (AUC) on datasets and also achieves equally good results in Accuracy (ACC) and False Alarm Rate (FAR). This study fills the gap of upgraded GCNs in detecting outliers to the best of our knowledge and provides a new way to convert Euclidean data to graphs.

Primary sclerosing cholangitis (PSC) is a biliary disease accompanied by chronic inflammation of the liver and biliary stricture. Mesenchymal stem cells (MSCs) are used to treat liver diseases because of their immune regulation and regeneration-promoting functions. This study was performed to explore the therapeutic potential of human placental MSCs (hP-MSCs) in PSC through the Takeda G protein-coupled receptor 5 (TGR5) receptor pathway. Liver tissues were collected from patients with PSC and healthy donors (n = 4) for RNA sequencing and intrahepatic cholangiocyte organoid construction. hP-MSCs were injected via the tail vein into Mdr2−/−, bile duct ligation (BDL), and 3,5-diethoxycarbonyl-1,4-dihydrocollidine (DDC) mouse models or co-cultured with organoids to confirm their therapeutic effect on biliary cholangitis. Changes in bile acid metabolic profile were analyzed by liquid chromatography/tandem mass spectrometry (LC-MS/MS). Compared with healthy controls, liver tissues and intrahepatic cholangiocyte organoids from PSC patients were characterized by inflammation and cholestasis, and marked downregulation of bile acid receptor TGR5 expression. hP-MSC treatment apparently reduced the inflammation, cholestasis, and fibrosis in Mdr2−/−, BDL, and DDC model mice. By activating the phosphatidylinositol 3 kinase/extracellular signal-regulated protein kinase pathway, hP-MSC treatment promoted the proliferation of cholangiocytes, and affected the transcription of downstream nuclear factor κB through regulation of the binding of TGR5 and Pellino3, thereby affecting the cholangiocyte inflammatory phenotype.

Cardinality estimation is critical for database management systems (DBMSs) to execute query optimization tasks, which can guide the query optimizer in choosing the best execution plan. However, traditional cardinality estimation methods cannot provide accurate estimates because they cannot accurately capture the correlation between multiple tables. Several recent studies have revealed that learning-based cardinality estimation methods can address the shortcomings of traditional methods and provide more accurate estimates. However, the learning-based cardinality estimation methods still have large errors when an SQL query involves multiple tables or is very complex. To address this problem, we propose a sampling-based tree long short-term memory (TreeLSTM) neural network to model queries. The proposed model addresses the weakness of traditional methods when no sampled tuples match the predicates and considers the join relationship between multiple tables and the conjunction and disjunction operations between predicates. We construct subexpressions as trees using operator types between predicates and improve the performance and accuracy of cardinality estimation by capturing the join-crossing correlations between tables and the order dependencies between predicates. In addition, we construct a new loss function to overcome the drawback that Q-error cannot distinguish between large and small cardinalities. Extensive experimental results from real-world datasets show that our proposed model improves the estimation quality and outperforms traditional cardinality estimation methods and the other compared deep learning methods in three evaluation metrics: Q-error, MAE, and SMAPE.

Remote sensing image object detection holds significant research value in resources and the environment. Nevertheless, complex background information and considerable size differences between objects in remote sensing images make it challenging. This paper proposes an efficient remote sensing image object detection model (MSA-YOLO) to improve detection performance. First, we propose a Multi-Scale Strip Convolution Attention Mechanism (MSCAM), which can reduce the introduction of background noise and fuse multi-scale features to enhance the focus of the model on foreground objects of various sizes. Second, we introduce the lightweight convolution module GSConv and propose an improved feature fusion layer, which makes the model more lightweight while improving detection accuracy. Finally, we propose the Wise-Focal CIoU loss function, which can reweight different samples to balance the contribution of different samples to the loss function, thereby improving the regression effect. Experimental results show that on the remote sensing image public datasets DIOR and HRRSD, the performance of our proposed MSA-YOLO model is significantly better than other existing methods.

The most significant technical challenges of current aerial image object-detection tasks are the extremely low accuracy for detecting small objects that are densely distributed within a scene and the lack of semantic information. Moreover, existing detectors with large parameter scales are unsuitable for aerial image object-detection scenarios oriented toward low-end GPUs. To address this technical challenge, we propose efficient-lightweight You Only Look Once (EL-YOLO), an innovative model that overcomes the limitations of existing detectors and low-end GPU orientation. EL-YOLO surpasses the baseline models in three key areas. Firstly, we design and scrutinize three model architectures to intensify the model’s focus on small objects and identify the most effective network structure. Secondly, we design efficient spatial pyramid pooling (ESPP) to augment the representation of small-object features in aerial images. Lastly, we introduce the alpha-complete intersection over union (α-CIoU) loss function to tackle the imbalance between positive and negative samples in aerial images. Our proposed EL-YOLO method demonstrates a strong generalization and robustness for the small-object detection problem in aerial images. The experimental results show that, with the model parameters maintained below 10 M while the input image size was unified at 640 × 640 pixels, the APS of the EL-YOLOv5 reached 10.8% and 10.7% and enhanced the APs by 1.9% and 2.2% compared to YOLOv5 on two challenging aerial image datasets, DIOR and VisDrone, respectively.

Abstract Background We explored whether stem cell therapy was effective for animal models and patients with Crohn’s disease (CD). Methods We searched five online databases. The relative outcomes were analyzed with the aid of GetData Graph Digitizer 2.26 and Stata 16.0 software. The SYRCLE risk of bias tool and the MINORS tool were used to assess study quality. Results We evaluated 46 studies including 28 animal works (n = 567) and 18 human trials (n = 360). In the animal studies, the disease activity index dramatically decreased in the mesenchymal stem cell (MSC) treatment groups compared to the control group. Rats and mice receiving MSCs exhibited longer colons [mice: standardized mean difference (SMD) 2.84, P = 0.000; rats: SMD 1.44, P = 0.029], lower histopathological scores (mice: SMD − 4.58, p = 0.000; rats: SMD − 1.41, P = 0.000) and lower myeloperoxidase levels (SMD − 6.22, P = 0.000). In clinical trials, stem cell transplantation reduced the CD activity index (SMD − 2.10, P = 0.000), the CD endoscopic index of severity (SMD − 3.40, P = 0.000) and simplified endoscopy score for CD (SMD − 1.71, P = 0.000) and improved the inflammatory bowel disease questionnaire score (SMD 1.33, P = 0.305) compared to control values. CD patients maintained high remission rates for 3–24 months after transplantation. Conclusions Stem cell transplantation is a valuable supplementary therapy for CD.

Sequential recommendation uses contrastive learning to randomly augment user sequences and alleviate the data sparsity problem. However, there is no guarantee that the augmented positive or negative views remain semantically similar. To address this issue, we propose graph neural network-guided contrastive learning for sequential recommendation (GC4SRec). The guided process employs graph neural networks to obtain user embeddings, an encoder to determine the importance score of each item, and various data augmentation methods to construct a contrast view based on the importance score. Experimental validation is conducted on three publicly available datasets, and the experimental results demonstrate that GC4SRec improves the hit rate and normalized discounted cumulative gain metrics by 1.4% and 1.7%, respectively. The model can enhance recommendation performance and mitigate the data sparsity problem.

Graph convolutional networks (GCNs) show great potential in recommendation applications, as they have excellent performance in propagation node information propagation and capturing high-order connectivity in user-item interaction graphs. However, in the current recommendation model based on GCN, incoming information from neighbors is aggregated during information propagation, and some of this information may be noisy due to negative information. Additionally, the over-smoothing problem occurs when the model layers are stacked too high. During the embedding learning of users in the graph convolution operation, an important factor is that high-order neighbor users with no interest are involved, leading to similar embedding for users with no interest. These issues can degrade the recommendation performance. To address these problems, this paper proposes a method called IMPLayerGCN. In this method, high-order graph convolution is performed within subgraphs, which are composed of users with similar interests and their interaction items. The higher-order graph convolution is carried out in the subgraph and the layer representation in the process of GCN re-information propagation and node update is refined. The convolution process uses a symmetrical matrix. This approach avoids the spread of negative information from higher-order neighbors to embedded learning.

The convolution module in Conformer is capable of providing translationally invariant convolution in time and space. This is often used in Mandarin recognition tasks to address the diversity of speech signals by treating the time-frequency maps of speech signals as images. However, convolutional networks are more effective in local feature modeling, while dialect recognition tasks require the extraction of a long sequence of contextual information features; therefore, the SE-Conformer-TCN is proposed in this paper. By embedding the squeeze-excitation block into the Conformer, the interdependence between the features of channels can be explicitly modeled to enhance the model’s ability to select interrelated channels, thus increasing the weight of effective speech spectrogram features and decreasing the weight of ineffective or less effective feature maps. The multi-head self-attention and temporal convolutional network is built in parallel, in which the dilated causal convolutions module can cover the input time series by increasing the expansion factor and convolutional kernel to capture the location information implied between the sequences and enhance the model’s access to location information. Experiments on four public datasets demonstrate that the proposed model has a higher performance for the recognition of Mandarin with an accent, and the sentence error rate is reduced by 2.1% compared to the Conformer, with only 4.9% character error rate.

Abstract Background Cholangiocyte senescence is an important pathological process in diseases such as primary sclerosing cholangitis (PSC) and primary biliary cirrhosis (PBC). Stem cell/induced pluripotent stem cell-derived exosomes have shown anti-senescence effects in various diseases. We applied novel organoid culture technology to establish and characterize cholangiocyte organoids (cholangioids) with oxidative stress-induced senescence and then investigated whether human placenta mesenchymal stem cell (hPMSC)-derived exosomes exerted a protective effect in senescent cholangioids. Methods We identified the growth characteristics of cholangioids by light microscopy and confocal microscopy. Exosomes were introduced concurrently with H2O2 into the cholangioids. Using immunohistochemistry and immunofluorescence staining analyses, we assessed the expression patterns of the senescence markers p16INK4a, p21WAF1/Cip1, and senescence-associated β-galactosidase (SA-β-gal) and then characterized the mRNA and protein expression levels of chemokines and senescence-associated secretory phenotype (SASP) components. Results Well-established cholangioids expressed cholangiocyte-specific markers. Oxidative stress-induced senescence enhanced the expression of the senescence-associated proteins p16INK4a, p21WAF1/Cip1, and SA-β-gal in senescent cholangioids compared with the control group. Treatment with hPMSC-derived exosomes delayed the cholangioid aging progress and reduced the levels of SASP components (i.e., interleukin-6 and chemokine CC ligand 2). Conclusions Senescent organoids are a potential novel model for better understanding senescence progression in cholangiocytes. hPMSC-derived exosomes exert protective effects against senescent cholangioids under oxidative stress-induced injury by delaying aging and reducing SASP components, which might have therapeutic potential for PSC or PBC.

As one of the most popular high-performance data processing technology, existing task and resource scheduling strategies for stream computing platforms are suffering from the problem of triggering hysteresis, which seriously affects the cluster performance. To address this problem, the idea of performance prediction based on timeline series data is proposed. Firstly, the performance variation rule of the stream processing platform is analyzed, which provides a basis for proposing the performance prediction model. Secondly, the basic topology paradigm, periodic load prediction model, and real-time throughput prediction model are proposed as the theoretical foundation for performance prediction. Thirdly, the performance prediction algorithm is proposed to predict the variation trend of the processing load and throughput of the cluster. The processing load is predicted periodically while the throughput of the cluster is predicted in a real-time manner. Finally, the performance evaluation algorithm is proposed to evaluate the cluster performance based on the prediction results and trigger the corresponding scheduling strategies in advance. The experimental results showed that the prediction accuracy of the proposed algorithm meets the requirements in practical applications. Meanwhile, the proposed method improves the performance of stream computing performance by triggering the scheduling strategies in advance.

Neural headline generation models have recently shown great results since neural network methods have been applied to text summarization. In this paper, we focus on news headline generation. We propose a news headline generation model based on a generative pre-training model. In our model, we propose a rich features input module. The headline generation model we propose only contains a decoder incorporating the pointer mechanism and the n-gram language features, while other generation models use the encoder-decoder architecture. Experiments on news datasets show that our model achieves comparable results in the field of news headline generation.

BackgroundOrganoids, which enable disease modeling and drug screening closer to an in vivo environment, can be isolated and grown from organs such as the brain, small intestine, kidney, lungs, and liver. To facilitate the establishment of liver and small intestinal organoids, we developed efficient protocols for cholangiocytes and intestine crypts collecting and organoid culturing.MethodsCholangiocytes were collected from intrahepatic bile ducts, the gallbladder, and small intestine crypts by gravity settling and multistep centrifugation methods. The cells isolated were embedded with Matrigel and grew in three-dimensional spheroids in a suitable culture medium. The stability of organoid cells was assessed by subculture, cryopreservation, and thawing. RNA and DNA extraction of organoids, as well as immunostaining procedure, were also optimized. Hand-picking procedures were developed and performed to ensure similar growth characteristics of organoids.ResultsA large number of cholangiocytes and small intestine crypts were collected under these protocols. Cholangiocytes developed into cyst-like structures after 3–4 days in Matrigel. After 1–2 weeks of cultivation, small intestinal organoids (in-orgs) developed buds and formed a mature structure. Compared to organoids derived from the gallbladder, cholangiocyte organoids (Cho-orgs) from intrahepatic the bile ducts grew more slowly but had a longer culture term, expressed the cholangiocytes markers Krt19 and Krt7, and recapitulated in vivo tissue organization.ConclusionsOur protocols simplified the cell collection procedure and avoided the possibility of exposing tissue-derived stem cells to mechanical damage or chemical injury by gravity settling and multistep centrifugation. In addition, our approach allowed similar growth characteristics of organoids from different mammalian tissue sources. The protocol requires 2–4 weeks to establish a stable organoid growth system. Organoids could be stably passaged, cryopreserved, and recovered under protocol guidance. Besides, the organoids of cholangiocytes and small intestines retained their original tissue characteristics, such as tissue-specific marker expression, which prepares them for further experiments such as preclinical in vitro trials and mechanism research studies.

In order to solve the problem that the load of big data stream computing platform fluctuates drastically while the cluster was suffering from the performance bottleneck due to the shortage of computing resources,the load prediction based elastic resource scheduling strategy in Flink (LPERS-Flink) was proposed.Firstly,the load prediction model was set up as the foundation to propose the load prediction algorithm and predict the variation tendency of the processing load.Secondly,the resource judgment model was set up to identify the performance bottleneck and resource redundancy of the cluster while the resource scheduling algorithm was proposed to draw up the resource rescheduling plan.Finally,the online load migration algorithm was proposed to execute the resource rescheduling plan and migrate processing load among nodes efficiently.The experimental results show that the strategy provides better performance promotion in the application with drastically fluctuating processing load.The scale and resource configuration of the cluster responded to the variation of processing load in time and the communication overhead of the load migration was reduced effectively.

Abstract Background Immune system disorders play important roles in acute lung injury (ALI), and mesenchymal stem cell (MSC) treatment can reduce inflammation during ALI. In this study, we compared the changes in lung B cells during MSC treatment. Methods We investigated the effects of MSCs on lung B cells in a mouse model of lipopolysaccharide (LPS)-induced ALI. MSCs were administered intratracheally 4 h after LPS. As vehicle-treated controls, mice were treated with phosphate-buffered saline (PBS) containing 2% C57BL/6 (PBS group). Histopathological changes, survival rate, inflammatory factor levels, and the number of neutrophils in bronchoalveolar lavage fluid (BALF) were determined. Single-cell RNA sequencing (scRNA-Seq) analysis was performed to evaluate the transcriptional changes in lung B cells between the PBS, LPS, and LPS/MSC groups on days 3 and 7. Results MSC treatment ameliorated LPS-induced ALI, as indicated by the reductions in mortality, the levels of chemokines and cytokines in BALF, and the severity of lung tissue histopathology in ALI mice. Lung B cells in the PBS group remained undifferentiated and had an inhibitory phenotype. Based on our scRNA-Seq results, the differentially expressed genes (DEGs) in lung B cells in both the PBS group and LPS group were involved in chemotaxis processes and some proinflammatory pathways. MSC treatment inhibited the expression of chemokine genes that were upregulated by LPS and were related to the recruitment of neutrophils into lung tissues. Immunoglobulin-related gene expression was decreased in lung B cells of mice treated with LPS/MSC for 7 days. The DEGs regulated by MSCs were enriched in biological processes, including humoral immune response and apoptotic signaling. Conclusions Lung B cells played an important role in the effects of treatment of ALI with MSCs. These observations provide new insights into the mechanisms underlying the effects of MSC treatment for ALI.

ABSTRACTThe seroprevalenc of autoimmune hepatitis (AIH)-related antibodies in patients, particularly Asians, with acute hepatitis E (AHE) is unclear. In this study, we investigated whether acute hepatitis E virus (HEV) infection is associated with the seroprevalence of AIH-related autoantibodies and assessed their impact on the disease characteristics. AIH-related autoantibodies were detected by indirect immunofluorescence in 198 AHE patients and 50 type 1 AIH patients. The positivity rates of against nuclear antigen (ANA) and smooth muscles antibody (SMA) in AHE patients were 37.4% and 22.7%, and the total positivity rate was 50%. Compared to those in AIH patients, the positivity rates of ANA-H and SMA-AA were significantly lower (35.1% vs. 82.1% and 4.4% vs. 88.4%). Female gender and the ALT level, but not immunosuppressive or antiviral drugs, were independently predictive of the presence of AIH-related autoantibodies in AHE patients. Fifty-two patients positive for AIH-related autoantibodies were followed up for 12 months. During this period, 33 of them became negative and 19 remained positive, albeit with significantly decreased titres. In conclusions, the seroprevalence of AIH-related autoantibodies in AHE patients was elevated, particularly in females, but their subspecificities and titres differed from those of type 1 AIH. Acute HEV infection may be related to AIH.Abbreviations: AIH: autoimmune hepatitis; AHE: acute hepatitis E; ANA: against nuclear antigen; SMA: smooth muscles antibody; ANA-H: ANA with homogeneous pattern; SMA-AA: SMA with anti-actin pattern; Anti-LKM1: anti- liver-kidney microsomes-1 antibody; ANCA: anti-neutrophil cytoplasmic antibody; AMA: anti-mitochondrial antibody; Anti-SLA: anti-soluble liver antigen; Anti-LC1: anti-liver cytoplasmic type 1 antibody; pANCA: perinuclear antineutrophil cytoplasmic antibody

The volume and type of streaming data is increasing rapidly, thus, real-time processing scenarios for streaming data have continued to increase. The inherent volatility of streaming data causes large changes in the throughput volatility of stream processing platforms, making evaluations of the maximum sustainable throughput (MST) for such platforms a challenge. To address these problems and improve the low efficiency of manual evaluations, a naïve method is typically used to evaluate the MST of platforms by periodically measuring throughput. The throughput volatility is detected using a skip sliding window that is defined as a bounded sequence in which a limited number of throughput values stored to measure volatility in a skipping fashion in each data growth cycle, which is a fixed time interval with respect to the continuous data rate increases. A skipping sliding window that is used to describe the procedure with which the MST is calculated in an approximate way is based on a sliding window, which is used usually in the context of stateful streaming operators. Then, according to the throughput volatility, the method judges whether the MST has been reached using a latency guarantee. However, because this method has low efficiency and a large error rate, we propose an adaptive MST evaluation method that adds a data-growth factor function to the naïve method cycle that dynamically and adaptively tunes the data rate for each data growth cycle. The experimental results from four open-source benchmarks running on three mainstream stream processing platforms show that the adaptive MST evaluation method has a lower error rate and executes faster compared with the naïve evaluation method. Moreover, the proposed method is insensitive to parameter variations and is suitable for mainstream stream processing platforms.

To solve the problems of a poor manual garbage sorting environment, including heavy tasks and low sorting efficiency, we propose the Lightweight Feature Fusion Single Shot Multibox Detector (L-SSD) algorithm to realize intelligent trash classification and recognition. Since waste has a small volume and the image resolution of garbage is always low, the algorithm that we propose is an enhanced single shot multibox detector (SSD) with a lightweight and novel feature fusion module. This SSD can significantly improve the performance of rubbish detection. In this feature fusion module, features from different layers with different scales are connected in series. A new feature pyramid was generated by using downsampling blocks, which will be fed to appointed multibox detectors to predict the final detection results. Due to the extremely unbalanced ratio of positive samples to negative samples, which leads to a low accuracy of SSD, Focal Loss using balanced cross-entropy is employed, which is provided by easy examples that corresponds to difficult samples with a decline in the loss weight. Thus, the training is biased towards meaningful samples. We have replaced the backbone network of VGG16 with ResNet-101 to achieve more accurate detection. We analyzed the performance of a nonmaximum suppression (NMS) algorithm and discovered that Soft-NMS was more suitable for learning better image representations. The strategy of Soft-NMS is to suppress the undesirable detection box rather than remove it completely. The experimental results show that the L-SSD exceeds a large number of state-of-the-art object detection algorithms in both accuracy and speed.

Cloud Data Computing (CDC) is conducive to precise energy-saving management of user data centers based on the real-time energy consumption monitoring of Information Technology equipment. This work aims to obtain the most suitable energy-saving strategies to achieve safe, intelligent, and visualized energy management. First, the theory of Convolutional Neural Network (CNN) is discussed. Besides, an intelligent energy-saving model based on CNN is designed to ameliorate the variable energy consumption, load, and power consumption of the CDC data center. Then, the core idea of the policy gradient (PG) algorithm is introduced. In addition, a CDC task scheduling model is designed based on the PG algorithm, aiming at the uncertainty and volatility of the CDC scheduling tasks. Finally, the performance of different neural network models in the training process is analyzed from the perspective of total energy consumption and load optimization of the CDC center. At the same time, simulation is performed on the CDC task scheduling model based on the PG algorithm to analyze the task scheduling demand. The results demonstrate that the energy consumption of the CNN algorithm in the CDC energy-saving model is better than that of the Elman algorithm and the ecoCloud algorithm. Besides, the CNN algorithm reduces the number of virtual machine migrations in the CDC energy-saving model by 9.30% compared with the Elman algorithm. The Deep Deterministic Policy Gradient (DDPG) algorithm performs the best in task scheduling of the cloud data center, and the average response time of the DDPG algorithm is 141. In contrast, the Deep Q Network algorithm performs poorly. This paper proves that Deep Reinforcement Learning (DRL) and neural networks can reduce the energy consumption of CDC and improve the completion time of CDC tasks, offering a research reference for CDC resource scheduling.

Camera surveillance is widely used in residential areas, highways, schools and other public places. The monitoring and scanning of sudden abnormal events depend on humans. Human anomaly monitoring not only consumes a lot of manpower and time but also has a large error in anomaly detection. Video anomaly detection based on AE (Auto-Encoder) is currently the dominant research approach. The model has a highly symmetrical network structure in the encoding and decoding stages. The model is trained by learning standard video sequences, and the anomalous events are later determined in terms of reconstruction error and prediction error. However, in the case of limited computing power, the complex model will greatly reduce the detection efficiency, and unnecessary background information will seriously affect the detection accuracy of the model. This paper uses the AE loaded with dynamic prototype units as the basic model. We introduce an attention mechanism to improve the feature representation ability of the model. Deep separable convolution operation can effectively reduce the number of model parameters and complexity. Finally, we conducted experiments on three publicly available datasets of real scenarios (UCSD Ped1, UCSD Ped2 and CUHK Avenue). The experimental results show that compared with the baseline model, the accuracy of our model improved by 1.9%, 1.4% and 6.6%, respectively, across the three datasets. Compared with many popular models, the validity of our model in anomaly detection is verified.

Background and Aims: Viral hepatitis are one of the main causes of liver cirrhosis. The treatment of portal hypertension caused by liver cirrhosis is difficult and diverse, and the therapeutic effect is unknown. Bayesian network meta-analysis was performed to compare the efficacy and safety of treatments for patients with portal hypertension and cirrhosis, including a transjugular intrahepatic portosystemic shunt (TIPS), endoscopic therapy, surgical therapy and medications.Methods: Eligible articles were searched for in PubMed, Embase, Cochrane Library and Web of Science databases from their inception until June 2020. Using the “gemtc-0.8.4” package in R v.3.6.3 software and the Just Another Gibbs Sampler v.4.2.0 program, network meta-analysis was performed using a random effects model within a Bayesian framework. The odds ratios for all-cause rebleeding, bleeding-related mortality, overall survival (OS), treatment failure and hepatic encephalopathy were determined within the Bayesian framework.Results: Forty randomized controlled trials were identified, including 4,006 adult patients and nine treatment strategies. Our results showed that distal splenorenal shunt and TIPS provided the best control of hemorrhage. Endoscopic variceal ligation with medication resulted in the highest OS rate. Medication alone resulted in poor OS and treatment failure.Conclusions: We performed a systematic comparison of diverse treatments for cirrhotic patients with portal hypertension. Our meta-analysis indicated that a TIPS and distal splenorenal shunt resulted in lower rates of rebleeding than did other therapies. Furthermore, drugs are more suitable for combination therapy than monotherapy.

Storm is suffering the problems of high energy consumption but low efficiency.Aiming at this problem,the resource constraint model,the optimal principle of data reorganization in executors and node voltage reduction principle were proposed based on the analysis of the architecture and topology of Storm,and further the energy-efficient strategy for data migration and merging was put forward in Storm(DMM-Storm),which was composed of resource constraint algorithm,data migration and merging algorithm as well as node voltage reduction algorithm.The resource constraint algorithm estimates whether work nodes are appropriate for data migration according to the resource constraint model.The data migration and merging algorithm designs an optimal method to migrate data according to the the optimal principle of data reorganization in executors.The node voltage reduction algorithm reduces voltage of work nodes according to node voltage reduction principle.The experimental results show that the DMM-Storm can reduce energy consumption efficiently without affecting the performance of cluster compared with the existing researches.

In order to solve the problem that the load of big data stream computing platform is increasing with fluctuation while the cluster was not able to rescale efficiently,the Flow-network based auto rescale strategy for Flink was proposed.Firstly,the flow-network model was set up and the capacity of each edge that was calculated by self-learning algorithm.Secondly,the bottleneck of the cluster was acquired by maximum-flow algorithm and the resource rescheduling plan was drawn up.Finally,the resource rescheduling plan was executed and the stateful data was migrated efficiently by the data migration algorithm based on the strategy of data partitioning by bulk and bucket.The experimental results show that the strategy can effectively provide performance promotion in the application with complex stateful data.It improved the throughput of the cluster and reduced the time overhead of the data migration on the premise of satisfying the latency constrain of the application,which means that the strategy promotes the scalability of the cluster efficiently.

Abstract Background The efficacy of mesenchymal stem cell (MSC)-based therapy for acute liver injury (ALI) involves coordination with the hepatic immune system, a complex and coordinated network of immune-cell interactions. However, studies of the immunomodulatory effects of MSCs have focused on a limited number of cell subsets rather than a systematic assessment. Methods Carbon tetrachloride (CCl4) was used to induce ALI in mice. To determine the efficacy of MSCs, ALI mice were injected with MSCs via the tail vein, and histopathological changes, survival rate, and the serum levels of liver enzymes were determined. To assess the immune response induced by MSCs, a mass cytometry panel of 43 metal isotope-tagged antibodies was used to characterize the hepatic immune compartment at days 1, 2, 3, and 7 after administration of MSCs or PBS. Results MSC treatment significantly alleviated CCl4-induced ALI and improved the survival rate. MSC treatment also modulated the hepatic immune system in terms of the distribution of immune-cell subsets and the phenotype of single cells. During the injured phase, MSCs inhibited a systemic response by reducing the numbers of Ly6ClowCD8+ TRM cells, conventional NK cells, and IgM+IgD+ B cells; suppressing the activation of Ly6ChiCD8+ TRM cells; downregulating MHC II and IgM expression in IgM+IgD+ B cells; and increasing the number of immunosuppressive monocyte-derived macrophages. During the recovery phase, MSCs promoted the retention of Ly6ClowCD8+ TRM cells and maintained the immunosuppressive activity of monocyte-derived macrophages. The response to MSC treatment differed between the injured and recovery phases, emphasizing the benefit of dynamic assessment of the immunomodulatory effects of MSCs. Conclusions We determined the immunomodulatory effects of MSC treatment on the subtype distribution and phenotypes of hepatic immune cells.

Introduction: Although epinephrine is universally acknowledged to increase return of spontaneous circulation (ROSC) after cardiac arrest, its balanced effects on later outcomes remain uncertain, causing potential harm during post-resuscitation phase. Recent studies have questioned the efficacy and potential deleterious effects of epinephrine on long-term survival and neurological outcomes, despite that the adverse relationship between epinephrine dose and outcome can be partially biased by longer CPR duration and underlying comorbidities. This study explored the long-term effect of epinephrine when used in a cohort of patients that underwent cardiac arrest during cardiopulmonary resuscitation.Methods: The data were originally collected from a retrospective institutional database from January 2007 to December 2015 and are now available on Dryad (via: https://doi.org/10.5061/dryad.qv6fp83). Use of epinephrine was coded by dose (

Objectives: Evidence has shown that angiotensin-converting enzyme 2 (ACE2), which can be upregulated after angiotensin-converting enzyme inhibitor (ACEi) and angiotensin receptor blocker (ARB) treatment, may play a dual role in the pathogenesis and progression of coronavirus disease 2019 (COVID-19). We aimed to assess the association between the use of ACEi/ARB and the outcome of COVID-19 patients with preexisting hypertension in non-endemic areas.Methods: From January 17, 2020, to February 19, 2020, 286 patients with hypertension were enrolled in this retrospective study out of 1,437 COVID-19 patients from 47 centers in Zhejiang and Jiangsu Province. The composite endpoints consisted of mechanical ventilation, intensive care unit (ICU) admission, or death. Cox proportional hazards analysis was performed to assess the association between ACEi/ARB and clinical outcomes of COVID-19 patients with hypertension.Results: In the main analysis, 103 patients receiving ACEi/ARB were compared with 173 patients receiving other regimens. Overall, 44 patients (15.94%) had an endpoint event. The risk probability of crude endpoints in the ACEi/ARB group (12.62%) was lower than that in the non-ACEi/ARB group (17.92%). After adjusting for confounding factors by inverse probability weighting, the results showed that the use of ACEi/ARB reduced the occurrence of end events by 47% [hazard ratio (HR) = 0.53; 95% CI, 0.34–0.83]. Similar results were obtained in multiple sensitivity analyses.Conclusions: In this retrospective study, among COVID-19 patients with hypertension, the use of ACEi/ARB is not associated with an increased risk of disease severity compared with patients without ACEi/ARB. The trends of beneficial effects of ACEi/ARB need to be further evaluated in randomized clinical trials.

You Only Look Once (YOLO) series detectors are suitable for aerial image object detection because of their excellent real-time ability and performance. Their high performance depends heavily on the anchor generated by clustering the training set. However, the effectiveness of the general Anchor Generation algorithm is limited by the unique data distribution of the aerial image dataset. The divergence in the distribution of the number of objects with different sizes can cause the anchors to overfit some objects or be assigned to suboptimal layers because anchors of each layer are generated uniformly and affected by the overall data distribution. In this paper, we are inspired by experiments under different anchors settings and proposed the Layered Anchor Generation (LAG) algorithm. In the LAG, objects are layered by their diagonals, and then anchors of each layer are generated by analyzing the diagonals and aspect ratio of objects of the corresponding layer. In this way, anchors of each layer can better match the detection range of each layer. Experiment results showed that our algorithm is of good generality that significantly uprises the performance of You Only Look Once version 3 (YOLOv3), You Only Look Once version 5 (YOLOv5), You Only Learn One Representation (YOLOR), and Cascade Regions with CNN features (Cascade R-CNN) on the Vision Meets Drone (VisDrone) dataset and the object DetectIon in Optical Remote sensing images (DIOR) dataset, and these improvements are cost-free.

Outlier detection is an important research direction in the field of data mining. Aiming at the problem of unstable detection results and low efficiency caused by randomly dividing features of the data set in the Isolation Forest algorithm in outlier detection, an algorithm CIIF (Cluster-based Improved Isolation Forest) that combines clustering and Isolation Forest is proposed. CIIF first uses the k-means method to cluster the data set, selects a specific cluster to construct a selection matrix based on the results of the clustering, and implements the selection mechanism of the algorithm through the selection matrix; then builds multiple isolation trees. Finally, the outliers are calculated according to the average search length of each sample in different isolation trees, and the Top-n objects with the highest outlier scores are regarded as outliers. Through comparative experiments with six algorithms in eleven real data sets, the results show that the CIIF algorithm has better performance. Compared to the Isolation Forest algorithm, the average AUC (Area under the Curve of ROC) value of our proposed CIIF algorithm is improved by 7%.

Purpose: Venous anastomosis is the key procedure for Ishikawa zone Ⅱ fingertip replantation. Both palmar and lateral veins provide efficient venous drainage. This study compared the clinical effects between these venous anastomoses for fingertip replantation. Methods: In 2016–2018, 61 patients underwent Ishikawa zone Ⅱ fingertip replantation with venous anastomosis (28 and 33 cases with palmar and lateral anastomoses, respectively). Retrospective comparative analyses evaluated surgical technique and function, including operative time; rates of finger survival, venous congestion, and infection; sensation; joint motion; cold intolerance symptom severity (CISS), Disabilities of the Arm, Shoulder, and Hand (DASH), and Vancouver scar scores; and chronic regional pain syndrome (CRPS) rates. Results: There were 33 patients with lateral vein anastomosis and 28 patients with palmar vein anastomosis. The average patient follow-up was 18.2 months. The survival rates did not differ significantly between groups (87.8% (29/33) vs. 85.7% (24/28), p > 0.05); however, the operative time was shorter in the lateral vein group than in the palmar vein group (78.57 ± 7.08 min vs. 67.88 ± 5.77 min, p < 0.05). Venous congestion and infection rates did not differ significantly between groups ( p > 0.05). The replanted finger function, including joint motion, sensation, DASH scores, Vancouver scar scores, and CRPS rates, did not differ significantly between groups ( p > 0.05). However, the CISS score was higher in the palmar vein group than in the lateral vein group (44.39 ± 5.16 vs. 38.09 ± 4.49, p < 0.05). Conclusions: Venous anastomosis with either palmar or lateral veins showed high survival rates and good function in fingertip replantation. The lateral vein had a shorter operative time and benefit to arterial revision and was especially suitable for patients with finger pulp damage.

Background and Aims. The involvement of cellular immunity in the development of hepatitis E virus (HEV) infection is rare. We aimed to study the roles of viral load and Th cell responses in acute hepatitis E (AHE) and HEV-related acute liver failure (HEV-ALF). Methods. We evaluated viral load and Th1/Th2 cytokine levels in 34 patients with HEV infection, including 17 each with AHE or HEV-ALF. Seventeen healthy controls (HCs) were also included who were negative for anti-HEV IgM and IgG. Results. There was no significant difference in viral load and HEV RNA in the AHE and HEV-ALF groups (both P>0.05). The Th lymphocyte levels (CD3+, CD4+) in the AHE and HEV-ALF groups were significantly higher than those in the HC group (both P0.05). Both IFN-γ and IL-10 showed gradual upward trend from the HC group to the AHE (both P0.05). Th2 bias was observed from the AHE (ratio=58.65) to HEV-ALF (ratio=1.20) groups. The level of IFN-γ was associated with the outcome of HEV-ALF patients. Conclusions. HEV viral load was not associated with aggravation of AHE, and the HEV-ALF patients showed significant Th2 bias, which may be involved in the aggravation of AHE.

In most of the existing multi-task learning (MTL) models, multiple tasks’ public information is learned by sharing parameters across hidden layers, such as hard sharing, soft sharing, and hierarchical sharing. One promising approach is to introduce model pruning into information learning, such as sparse sharing, which is regarded as being outstanding in knowledge transferring. However, the above method performs inefficiently in conflict tasks, with inadequate learning of tasks’ private information, or through suffering from negative transferring. In this paper, we propose a multi-task learning model (Pruning-Based Feature Sharing, PBFS) that merges a soft parameter sharing structure with model pruning and adds a prunable shared network among different task-specific subnets. In this way, each task can select parameters in a shared subnet, according to its requirements. Experiments are conducted on three benchmark public datasets and one synthetic dataset; the impact of the different subnets’ sparsity and tasks’ correlations to the model performance is analyzed. Results show that the proposed model’s information sharing strategy is helpful to transfer learning and superior to the several comparison models.