Search

Background: Sodium‐glucose cotransporter 2 inhibitors (SGLT2i) is reported to reduce incident atrial fibrillation (AF) in patients with or without diabetes; however, its cardiovascular (CV) benefit for AF patients remains unclear.SS Aims: To investigate the effect of SGLT2i on the incidence of CV events in patients with AF. Methods: Six randomized controlled trials (RCTs) assessing the effects of SGLT2i on CV outcomes in patients with or without AF were included (PROSPERO: CRD 42023431535). The primary endpoint was the composite outcome of heart failure (HF) hospitalization and CV death. Additionally, we assessed the effects of treatment in prespecified subgroups on HF hospitalization, CV death, and all‐cause mortality. Results: Among 38,529 participants from all trials, 5018 patients with AF were treated with SGLT2i. The follow‐up period of these trials ranged from 2.3 to 3.3 years. SGLT2i treatment was significantly associated with the risk reduction of primary endpoint in patients with AF (risk ratio [RR] 0.81, 95% confidence interval [CI] 0.74–0.88; p < 0.001), consistent with the finding in the general population (p for interaction = 0.76). SGLT2i was also associated with a consistent reduction in the risk of HF hospitalization in patients with AF (RR 0.76, 95% CI 0.69–0.84; p < 0.001) or not (RR 0.72, 95% CI 0.64–0.80; p < 0.0001), with no statistical difference between them (p for interaction = 0.41). Meta‐regression further revealed no significant association between the prevalence of HF with reduced ejection fraction or diabetes and the effect size of SGLT2i. Conclusions: The treatment effects of SGLT2i were associated with a lower incidence of CV events, especially HF hospitalization, in patients with AF. [ABSTRACT FROM AUTHOR]

In recent years, cross‐modal retrieval has been a popular research topic in both fields of computer vision and natural language processing. There is a huge semantic gap between different modalities on account of heterogeneous properties. How to establish the correlation among different modality data faces enormous challenges. In this work, we propose a novel end‐to‐end framework named Dual Multi‐Angle Self‐Attention (DMASA) for cross‐modal retrieval. Multiple self‐attention mechanisms are applied to extract fine‐grained features for both images and texts from different angles. We then integrate coarse‐grained and fine‐grained features into a multimodal embedding space, in which the similarity degrees between images and texts can be directly compared. Moreover, we propose a special multistage training strategy, in which the preceding stage can provide a good initial value for the succeeding stage and make our framework work better. Very promising experimental results over the state‐of‐the‐art methods can be achieved on three benchmark datasets of Flickr8k, Flickr30k, and MSCOCO. [ABSTRACT FROM AUTHOR]