Zhihui Zhang

Background: The timely identification of left atrial appendage thrombus (LAAT) is paramount for stroke prevention in non-valvular atrial fibrillation (NVAF), yet current reliance on transesophageal echocardiography (TEE) limits accessibility. Objectives: To develop and validate a non-invasive, electrocardiogram (ECG)-based machine-learning model for predicting LAAT in patients with first-diagnosed NVAF. Methods: A retrospective cohort of 486 patients who underwent ECG within 2 days before TEE or cardiac CT was analyzed. Eight predictive ECG features were selected using Lasso regression. Five machine-learning models were trained and evaluated. Model performance was assessed by the area under the receiver operating characteristic curve (AUC), calibration, and decision curve analysis (DCA). Continuous net reclassification improvement (NRI) and integrated discrimination improvement (IDI) were used to compare against the CHA ₂ DS ₂ –VASc score. Results: LAAT was present in 87 patients (17.9%). The logistic regression (LR) model demonstrated optimal performance, achieving an AUC of 0.803 (95% CI 0.750–0.855), sensitivity of 0.77, specificity of 0.81, and excellent calibration (Hosmer–Lemeshow χ ² = 7.295, df = 8, P = 0.505). DCA showed net clinical benefit across thresholds of 4–75%. It significantly outperformed CHA ₂ DS ₂ –VASc (AUC = 0.570), yielding a continuous NRI of 0.9228 and an IDI of 0.1882. Adding CHA ₂ DS ₂ –VASc did not improve prediction. The final model was deployed as an interactive online calculator for real-time risk assessment. Conclusions: The ECG-based logistic regression model provides an interpretable approach with promising internally validated performance for predicting LAAT in NVAF. This tool enables rapid, non-invasive risk stratification and may help prioritize patients who warrant expedited definitive imaging; its clinical utility, including any role in anticoagulation decisions, requires external prospective validation.

Background: Traditional occluders for atrial septal defect have 2 drawbacks: incomplete endothelialization and difficulty in achieving atrial septum puncture post implantation. The novel occluder ReAces was designed to address these 2 clinical pain points. In this study, we aimed to compare the efficacy and safety of ReAces with a traditional Amplatzer‐like atrial septal defect occluder. Methods: This was a multicenter, randomized, parallel‐controlled, noninferiority clinical trial. Patients with atrial septal defects were enrolled and randomly implanted with a ReAces or a traditional Amplatzer‐like occluder. The primary outcome was complete closure rate at 12 months post implantation. The Migraine Disability Assessment scores before and after the procedure were recorded. Results: Sixty‐two patients were implanted with ReAces and 64 patients with a traditional occluder. At the 12‐month follow‐up, complete closure was achieved in 61 patients in the trial group (98.4%) and 63 patients in the control group (98.4%), which demonstrated noninferiority (95% CI, −4.4% to 4.3%). No severe adverse events occurred during 12 months of follow‐up except a case of device dislodgement in the trial group. The thickness of the central region of ReAces at the 12‐month follow‐up was significantly lower than that of the traditional device (6.8±3.7 mm versus 12.2±5.2, P P =0.042) and a lower proportion of patients with a nonzero Migraine Disability Assessment score (5.6% versus 18.2%, P =0.042) at the 1‐month follow‐up. Conclusions: The efficacy of ReAces was noninferior to that of a traditional Amplatzer‐like occluder for transcatheter atrial septal defect closure. ReAces had a thinner central region during the follow‐up and a lower Migraine Disability Assessment score at the 1‐month follow‐up. Registration: URL: https://www.clinicaltrials.gov ; Unique identifier: NCT05371366.

Hypoxia-induced right ventricular (RV) remodeling and dysfunction present a significant health risk to populations experiencing prolonged hypoxic conditions. Intense light, a noninvasive and easily implemented intervention, has previously been reported to exert cardioprotective effects by improving myocardial ischemia. However, whether intense light provides protective benefits against RV remodeling and the underlying mechanisms remain largely unexplored. In this study, we established mouse models exhibiting RV remodeling and dysfunction through long-term hypoxia to investigate the protective effects of intense light. Echocardiography, hemodynamic parameters measurements, and Fulton index assessments were employed to evaluate RV dysfunction and remodeling. Additionally, single-nuclei RNA sequencing, immunohistochemistry, immunofluorescence, and western blotting analyses were conducted to identify targeted genes in macrophage-associated inflammation within the heart. The results indicate that intense light significantly alleviates hypoxia-induced RV remodeling and dysfunction in mice. Intense light may mediate macrophage-associated inflammation through differentially expressed genes, including PF4, as well as the quantity of macrophages in the right ventricles (RVes). Resident macrophages (Res_Macro) demonstrate cardioprotective effects when intense light is applied, which mitigates RV remodeling. Our findings also suggest that PF4 expression and the presence of PF4⁺ resident macrophages (Res_PF4⁺_Macro) are linked to the attenuation of RV remodeling by intense light. Macrophage PF4 expression and the quantity of PF4⁺ macrophages in the RVes are closely associated with the levels of RV remodeling and dysfunction. This study unveils a novel noninvasive approach for the prevention of RV remodeling and dysfunction induced by hypoxia, and indicates that Res_PF4⁺_Macro and PF4 expression could be potential intervening targets.

Virtual coronary intervention planning (VCIP) aims to optimize the hemodynamic outcomes of percutaneous coronary intervention (PCI) in patients with coronary stenosis. However, its clinical adoption remains constrained by the computational burden associated with evaluating numerous combinatorial intervention strategies, leading to time-consuming workflows and potentially suboptimal decisions in the catheterization laboratory. While conventional deep reinforcement learning (DRL) offers a path to automated VCIP, it often explores state-action-reward space inefficiently. In this study, we propose an Informed-Exploration Reinforcement Learning framework that concentrates the search on clinically meaningful interventions by integrating historical intervention experience with patient-specific anatomical and physiological information to guide the generation of functionally informed stent strategies. Extensive experiments on 172 vessels from 146 patients show that IERL achieves high agreement (r = 0.815) with real interventions and excellent computational efficiency with an average run time of 2.1 seconds. By aligning exploration with both prior experience and patient context, IERL provides objective, reproducible, and near-real-time VCIP decision support, enabling timely and interpretable recommendations compatible with catheterization workflows. The code and models are available at: https://github.com/HIC-SYSU/IERL/tree/main.