Search

Although more and more cardiovascular risk factors have been verified in peritoneal dialysis (PD) populations in different countries and regions, it is still difficult for clinicians to accurately and individually predict death in the near future. We aimed to develop and validate machine learning-based models to predict near-term all-cause and cardiovascular death. Machine learning models were developed among 7539 PD patients, which were randomly divided into a training set and an internal test set by five random shuffles of 5-fold cross-validation, to predict the cardiovascular death and all-cause death in 3 months. We chose objectively collected markers such as patient demographics, clinical characteristics, laboratory data, and dialysis-related variables to inform the models and assessed the predictive performance using a range of common performance metrics, such as sensitivity, positive predictive values, the area under the receiver operating curve (AUROC), and the area under the precision recall curve. In the test set, the CVDformer models had a AUROC of 0.8767 (0.8129, 0.9045) and 0.9026 (0.8404, 0.9352) and area under the precision recall curve of 0.9338 (0.8134,0.9453) and 0.9073 (0.8412, 0.9164) in predicting near-term all-cause death and cardiovascular death, respectively. The CVDformer models had high sensitivity and positive predictive values for predicting all-cause and cardiovascular deaths in 3 months in our PD population. Further calibration is warranted in the future. [ABSTRACT FROM AUTHOR]

In order to improve the effect of chinese named entity recognition, a method based on the XLNET- Transformer_P-CRF model was proposed, which used the Transformer_P encoder, improved the shortcomings of the traditional Transformer encoder that couldn’t obtain relative position information. Experiments show that the XLNET-Transformer_P-CRF model achieves 95.11%, 80.54%, 96.70%, and 71.46% F1 values on the four types of data sets: MSRA, OntoNotes4.0, Resume, and Weibo, which are all higher than other mainstream chinese NER model.

Background: Adaptive immunity is an important disease mediator of pulmonary vascular remodeling during pulmonary hypertension (PH) development, especially T‐cells lymphocytes. However, data for bibliometric analysis of T cell immunity in PH is currently vacant. This aimed to provide a comprehensive and visualized view of T‐cells research in PH pathogenesis and to lay a solid foundation for further studies. Methods: The data was acquired from the Web of Science Core Collection database. Web of Science analytic tool was used to analysis the publication years, authors, journals, countries, and organizations. CiteSpace 6.2.R3, VOSviewer 1.6.16, and Scimago Graphica 1.0.35.0 were applied to conduct a visualization bibliometric analysis about authors, countries, institutions, journals, references, and keywords. Results: Nine hundred and eight publications from 1992 to 2022 were included in the analysis. The results showed that Humbert Marc was the most prolific author. American Journal of Physiology Lung Cellular and Molecular Physiology had the most related articles. The institution with the most articles was Udice French Research University. The United States was far ahead in the article output. Keywords analysis showed that "Pulmonary hypertension" was the most usually appeared keyword in the relevant literature, and included "T‐cells", "Regulatory T cells", and "Activated T cell." "miRNA" of reference co‐citation clustering analysis demonstrated the possible T‐cell immunity activation mechanisms in PH. The most cited literature was published in the European Heart Journal by Galie N in 2016. The strongest citation burst of keyword is "gene expression" and terms such as "vascular remodeling," "growth," "proliferation," and "fibrosis" are among the list, indicating that T‐cells interact with stromal vascular cells to induce pulmonary vascular remodeling. The strongest burst of cited reference is "Galie N, 2016." Conclusions: T‐cell immunity is an important pathogenesis mechanism for PH development, which may have interaction with miRNAs and stromal vascular cells, but the possible T‐cell immunity activation mechanisms in PH need to be investigated further. [ABSTRACT FROM AUTHOR]

The existing multi-source contour matching studies have focused on the matching methods with consideration of topological relations and similarity measurement based on spatial Euclidean distance, while it is lack of taking the contour geometric features into account, which may lead to mismatching in map boundaries and areas with intensive contours or extreme terrain changes. In light of this, it is put forward that a matching strategy from coarse to precious based on the contour geometric features. The proposed matching strategy can be described as follows. Firstly, the point sequence is converted to feature sequence according to a feature descriptive function based on curvature and angle of normal vector. Then the level of similarity among multi-source contours is calculated by using the longest common subsequence solution. Accordingly, the identical contours could be matched based on the above calculated results. In the experiment for the proposed method, the reliability and efficiency of the matching method are verified using simulative datasets and real datasets respectively. It has been proved that the proposed contour matching strategy has a high matching precision and good applicability.

Aiming at the requirements for multi-scale Logo detection in natural scene images, a multi-scale Logo detection algorithm based on convolutional neural network was proposed. The algorithm was based on the realization of two-stage object detection. By constructing feature pyramids and adopting layer-by-layer prediction, multi-scale region proposals were generated. The multi-layer feature maps in convolutional neural networks were fused to enhance the feature representation. The experimental results on the FlickrLogos-32 dataset show that compared with the baseline, the proposed algorithm can improve the recall rate of region proposals, and can improve the performance of small Logo detection while ensuring the accuracy of large and middle Logo, proving the superiority of the proposed algorithm.

Previous studies of group percolation models in interdependent networks with reinforced nodes have rarely addressed the effects of the degree of reinforced nodes and the heterogeneity of group size distribution. In this paper, a cascading failure model in interdependent networks with reinforced crucial nodes and dependency groups is investigated numerically and analytically. For each group, we assume that if all the nodes in a group fail on one network, a node on another network that depends on that group will fail. We find that rich percolation transitions can be classified into three types: discontinuous, continuous, and hybrid phase transitions, which depend on the density of reinforced crucial nodes, the group size, and the heterogeneity of group size distribution. Importantly, our proposed crucial reinforced method has higher reinforcement efficiency than the random reinforced method. More significantly, we develop a general theoretical framework to calculate the percolation transition points and the shift point of percolation types. Simulation results show that the robustness of interdependent networks can be improved by increasing the density of reinforced crucial nodes, the group size, and the heterogeneity of group size distribution. Our theoretical results can well agree with numerical simulations. These findings might develop a new perspective for designing more resilient interdependent infrastructure networks. [ABSTRACT FROM AUTHOR]

Background To explore the cut-off values of haemoglobin (Hb) on adverse clinical outcomes in incident peritoneal dialysis (PD) patients based on a national-level database. Methods The observational cohort study was from the Peritoneal Dialysis Telemedicine-assisted Platform (PDTAP) dataset. The primary outcomes were all-cause mortality, major adverse cardiovascular events (MACE) and modified MACE (MACE+). The secondary outcomes were the occurrences of hospitalization, first-episode peritonitis and permanent transfer to haemodialysis (HD). Results A total of 2591 PD patients were enrolled between June 2016 and April 2019 and followed up until December 2020. Baseline and time-averaged Hb <100 g/l were associated with all-cause mortality, MACE, MACE+ and hospitalizations. After multivariable adjustments, only time-averaged Hb <100 g/l significantly predicted a higher risk for all-cause mortality {hazard ratio [HR] 1.83 [95% confidence interval (CI) 1.19–281], P  = .006}, MACE [HR 1.99 (95% CI 1.16–3.40), P  = .012] and MACE+ [HR 1.77 (95% CI 1.15–2.73), P  = .010] in the total cohort. No associations between Hb and hospitalizations, transfer to HD and first-episode peritonitis were observed. Among patients with Hb ≥100 g/l at baseline, younger age, female, use of iron supplementation, lower values of serum albumin and renal Kt/V independently predicted the incidence of Hb <100 g/l during the follow-up. Conclusion This study provided real-world evidence on the cut-off value of Hb for predicting poorer outcomes through a nation-level prospective PD cohort. [ABSTRACT FROM AUTHOR]

The development of adversarial technology, represented by adversarial text, has brought new challenges to rumor detection based on deep learning. In order to improve the robustness of rumor detection models under adversarial conditions, we propose a robust detection method based on the ensemble of multi-defense model on the basis of several mainstream defense methods such as data enhancement, random smoothing, and adversarial training. First, multiple robust detection models are trained based on different defense principles; then, two different ensemble strategies are used to integrate the above models, and the detection effect under different ensemble strategies is studied. The test results on the open-source dataset Twitter15 show that the proposed method is able to compensate for the shortcomings of a single model by ensembling different decision boundaries to effectively defend against mainstream adversarial text attacks and improve the robustness of rumor detection models compared to existing defense methods. [ABSTRACT FROM AUTHOR]