Machine learning model for predicting left atrial thrombus or spontaneous echo contrast in non-valvular atrial fibrillation patients based on multimodal echocardiographic parameters

Abstract

BACKGROUND: This study sought to develop a robust machine learning (ML)-based predictive model that synthesizes multimodal echocardiographic data and clinical risk factors to assess thrombosis risk in patients with non-valvular atrial fibrillation (NVAF). METHODS AND RESULTS: A total of 402 NVAF patients scheduled for AF radiofrequency ablation and/or left atrial appendage closure at the First Affiliated Hospital of Guangxi Medical University from January 2020 to December 2023 were prospectively collected. Among them, there were 289 males (71.9%) and 113 females (28.1%), with a mean age of 59.7 years. There were 142 patients (35.3%) with left atrial thrombus/spontaneous echocardiographic contrast (LAT/SEC) and 260 patients (64.7%) without LAT/SEC. We constructed seven ML models from the collected clinical data, biochemical markers, and multimodal echocardiographic parameters. Models were evaluated via ROC curves; SHAP values were employed for feature importance analysis.The integration of multimodal echocardiographic parameters with clinical risk factors based on sophisticated ML algorithms significantly improved the precision of thrombus risk prediction in patients with NVAF. Specifically, the XGBoost model (AUC 0.959, 95% CI 0.925?0.993) slightly outperformed the traditional Logistic regression model (AUC 0.949, 95% CI: 0.911-0.987) in predicting thrombus formation risk in NVAF patients, and showed superior predictive ability compared to other ML algorithms. Additionally, XGBoost offered greater clinical net benefit within a threshold probability range of 0.1 to 1.0. SHAP analysis revealed that left atrial structure (left atrial volume index, three-dimensional sphericity index), hemodynamic parameters (left atrial acceleration factor and S/D ratio), and functional parameters (peak atrial longitudinal strain) were important features in predicting the risk of thrombus formation in NVAF patients, with reduced peak atrial longitudinal strain being the most important risk factor for predicting thrombus. Conclusion: Developing a predictive model utilizing ML techniques that incorporate multimodal echocardiographic parameters in conjunction with clinical risk factors has the potential to enhance the predictive accuracy of the thrombosis risk in individuals with NVAF. The XGBoost model shows that decreased PALS, hemodynamic abnormalities and left atrium spherical remodeling are significant factors correlated with increased risk of thrombus in NVAF.