Propagation Structure Fusion for Rumor Detection Based on Node-Level Contrastive Learning.

With the rise of social media, the rapid spread of rumors online has resulted in numerous negative effects on society and the economy. The methods for rumor detection have attracted great interest from both academia and industry. Given the widespread effectiveness of contrastive learning, many graph contrastive learning models for rumor detection have been proposed by using the event propagation structure as graph data. However, the existing contrastive models usually treat the propagation structure of other events similar to the anchor events as negative samples. While this design choice allows for discriminative learning, on the other hand, it also inevitably pushes apart semantically similar samples and, thus, degrades model performance. In this article, we propose a novel propagation fusion model called propagation structure fusion model based on node-level contrastive learning (PFNC) for rumor detection based on node-level contrastive learning. PFNC first obtains three augmented propagation structures by masking the text of each node in the propagation structure randomly and perturbing some edges in the propagation structure based on the importance of edges. Then, PFNC applies the node-level contrastive learning method between every two augmented propagation structures to prevent the samples with similar propagation structure from far away. Finally, a convolutional neural network (CNN)-based model is proposed to capture the relevant information that is consistent and supplementary among three augmented propagation structures by regarding the propagation structure of the event as a color picture, three augmented propagation structures as color channels, and each node as a pixel. The experimental results on real datasets show that the PFNC significantly outperforms the state-of-the-art models for rumor detection.

Left atrial strain superior to structural remodeling in identifying occult atrial fibrillation.

OBJECTIVE: Occult atrial fibrillation (AF) is a significant risk factor for occult stroke but is difficult to detect. Myocardial strain, based on speckle tracking echocardiography (STE), can detect subclinical changes in left atrial (LA) function. This study aimed to investigate the diagnostic value of LA strain in identifying asymptomatic AF patients with normal electrocardiograms and healthy volunteers. METHODS: A total of 354 subjects were retrospectively enrolled. Patients were divided into two groups based on whether they had AF during echocardiography. Patients with sinus rhythm during echocardiography were further divided into healthy control group and occult AF group. Patients with AF during echocardiography were further divided into paroxysmal AF group and persistent AF group. LA mechanical function measured by STE was compared between patients with asymptomatic AF and volunteers. Conventional echocardiographic indicators were measured. The diagnostic value of LA strain for identifying asymptomatic AF was assessed by univariate and multivariate regression analysis and receiver operating characteristic (ROC) curves. RESULTS: Occult AF patients had higher NT-proBNP levels, larger RA area, larger LAVmax and decreased LAEF than control group. However, occult AF patients had lower NT-proBNP levels, RA area and LAVmax and higher LAEF than paroxysmal and persistent AF. The measured LA reservoir strain (LASr), LA conduit strain (LAScd), LA contraction strain (LASct) of occult AF group was significantly lower than that of control group. However, occult AF patients had preserved LAScd and LASct than paroxysmal and persistent AF. RA area, LAEF and LASr were remarkable correlation with occult AF after adjustment for NT-proBNP, LAVmax, and LAScd. The area under curve of ROC for LASr was the greatest among RA area, LAEF and LASr, with a cut-off value of 34.1% (sensitivity: 75.4%, specificity: 87.6%). CONCLUSION: LASr can identify occult AF in the asymptomatic population. Patients with LASr values ≤34.1% have a higher incidence of occult cardiac dysfunction. These findings help identify patients with occult AF and further risk stratification for the AF population.

Secondary leaflet tethering in patients with severe degenerative mitral regurgitation and its association with the severity of mitral regurgitation.

BACKGROUND: The purpose of the study was to determine the association between vena contracta area (VCA) and secondary leaflet tethering among mitral valve prolapse (MVP) patients, and thus to further identify and characterize an MVP with pathological leaflet tethering (MVPt+) phenotype. METHODS: We prospectively evaluated 94 consecutive MVP patients with significant mitral regurgitation (MR) and 21 healthy controls. MVPt+ group was defined as tenting volume index (TVi) > .7 mL/m2 . The three-dimensional (3D) geometry of mitral valve apparatus and VCA was measured with dedicated quantification software. RESULTS: Of the 94 patients with MVP and significant MR, 31 patients showed a TVi > .7 mL/m2 and entered the MVP with leaflet tethering (MVPt+) group. In stepwise multivariate analysis, only prolapse volume index and TVi were independently associated with 3D VCA. 3D VCA, annular area index, and plasma levels of NT-proBNP were independently correlated with the severity of leaflet tethering. ROC curve revealed that a 3D VCA ≥ .55 cm2 is the optimal cutoff point to predict MVPt+ phenotype. CONCLUSIONS: Secondary leaflet tethering is a significant mechanism behind severe degenerative MR, resulting in an MVPt+ phenotype featuring more advanced morphological and hemodynamical characteristics.

Mechanical Synchrony and Myocardial Work in Heart Failure Patients With Left Bundle Branch Area Pacing and Comparison With Biventricular Pacing.

Background: Little is known about the efficacy of permanent left bundle branch area pacing (LBBAP) in delivering cardiac resynchronization therapy (CRT). This study aimed to evaluate the effect of LBBAP on mechanical synchronization and myocardial work (MW) in heart failure (HF) patients and to compare LBBAP with biventricular pacing (BVP). Methods: This is a multicenter, prospective cohort study. From February 2018 to January 2021, 62 consecutive HF patients with reduced ejection fraction (LVEF ≤ 35%) and complete left bundle branch block (CLBBB) who underwent LBBAP or BVP were enrolled in this study. Echocardiograms and electrocardiograms and were conducted before and 3-6 months after implantation. Intra- and interventricular synchronization were assessed using two-dimensional speckle tracking imaging (2D-STI). The left ventricular pressure-strain loop was obtained by combining left ventricular strain with non-invasive blood pressure to evaluate mechanical efficiency. Results: The echocardiographic response rates were 68.6 and 88.9% in the BVP and LBBAP groups, respectively. Left bundle branch area pacing resulted in significant QRS narrowing (from 177.1 ± 16.7 to 113.0 ± 18.4 ms, P < 0.001) and improvement in LVEF (from 29.9 ± 4.8 to 47.1 ± 8.3%, P < 0.001). The global wasted work (GWW) (410.3 ± 166.6 vs. 283.0 ± 129.6 mmHg%, P = 0.001) and global work efficiency (GWE) (64.6 ± 7.8 vs. 80.5 ± 5.7%, P < 0.001) were significantly improved along with shorter peak strain dispersion (PSD) (143.4 ± 45.2 vs. 92.6 ± 35.1 ms, P < 0.001) and interventricular mechanical delay (IVMD) (56.4 ± 28.5 vs. 28.9 ± 19.0 ms, P < 0.001), indicating its efficiency in improving mechanical synchronization. In comparison with BVP, LBBAP delivered greater improvement of QRS narrowing (-64.1 ± 18.9 vs. -32.5 ± 22.3 ms, P < 0.001) and better mechanical synchronization and efficiency. Conclusions: Left bundle branch area pacing was effective in improving cardiac function, mechanical synchronization, and mechanical efficiency and may be a promising alternative cardiac resynchronization therapy.

Differentially Private Consensus With Quantized Communication.

This paper focuses on studying the differentially private consensus problem in multiagent networks under a quantized communication environment, where the exact real-value state is not available for transmission due to the range limitation of digital channels. We first extend the differentially private consensus model to the case of a quantized communication environment integrated with a dynamic encoding/decoding scheme and propose a differentially private communication algorithm utilizing the quantized state with a bounded quantizer instead of the exact real-value state to reach an agreement while protecting the initial or current states of the participants from information disclosure. Then, the convergence analysis of mean square consensus in the case of an unbounded quantizer is given to explain the sufficiency of the extended model and convergence conditions. To overcome the uncertainty of saturation in the case of a bounded quantizer, we also give a statistical analysis on the boundedness of quantization that the bounded quantizer with a finite number of bits can remain unsaturated with a desired high probability under certain conditions. Furthermore, we provide the statistical analysis on the convergent accuracy, which shows that the agreement value just converges to a random variable that falls in the neighboring range of the initial state average and the expectation of the agreement value is equal to the initial state average exactly. In addition, we provide the differential privacy analysis for individual agents and the whole network, and then establish the potential relationship between the dynamic encoding/decoding scheme and the differential privacy mechanism. Finally, the simulation results visually show that the proposed algorithm and the main theoretical results are effective and correct.