Abstract:This study introduces the P‐echo model, a noninvasive tool for predicting shunt closure feasibility in adult congenital heart disease patients with post‐tricuspid valve shunt defects. Developed using retrospective analysis of 1474 VSD or PDA patients and key echocardiographic parameters, the P‐echo model boasts excellent predictive performance and high accuracy, potentially reducing the need for invasive procedures. It categorizes patients into low, medium, and high‐risk groups for shunt closure feasibility, providing critical guidance for clinical decision‐making and individualized treatment strategies. The P‐echo model offers a promising alternative to invasive right heart catheterization, highlighting its potential for wide clinical use. Background As right heart catheterization (RHC) is invasive and not always accessible, this study developed a noninvasive model (P‐echo) to predict shunt closure feasibility in adult congenital heart disease (ACHD) patients with post‐tricuspid valve shunt defects (PTD), specifically isolated ventricular septal defects (VSD) and patent ductus arteriosus (PDA). Methods A retrospective analysis of 1474 VSD or PDA patients from 2012 to 2022 was conducted. Echocardiographic parameters were assessed, and key variables identified via LASSO regression. The P‐echo model incorporated left to right velocity (LRv), right to left velocity (RLv), tricuspid regurgitation (TR), pulmonary artery diameter (PA), and RV/LV ratio. Its predictive performance was evaluated using ROC curve analysis. Results The P‐echo model demonstrated excellent predictive performance with AUC values of 0.975 (95% CI: 0.965–0.984) in the derivation set, 0.963 (95% CI: 0.937–0.989) in the validation set, and high accuracy in both PDA (AUC 0.975, 95% CI: 0.965–0.984) and VSD (AUC 0.958, 95% CI: 0.936–0.980) subsets. In the derivation set, the model categorized patients into low (9.1% closure rate), medium (70.9% closure rate), and high‐risk groups (99.7% closure rate) for shunt closure feasibility. Calibration plots confirmed the model's accuracy. Decision curve analysis showed a higher net benefit across a range of threshold probabilities, indicating the clinical usefulness of the model. Conclusions The P‐echo model is a robust and reliable tool for predicting the feasibility of shunt closure in patients with PTD, offering a noninvasive alternative to RHC. This model can guide clinical decision‐making and support individualized treatment strategies in ACHD management.

An explainable deep learning model to predict partial anomalous pulmonary venous connection for patients with atrial septal defect

Explanatory deep learning to predict elevated pulmonary artery pressure in children with ventricular septal defects using standard chest x-rays: a novel approach

Intelligent Diagnosis of Atrial Septal Defect in Children Using Echocardiography with Deep Learning

Explainable Machine-Learning Predictions for Complications after Pediatric Congenital Heart Surgery

Intelligent Auxiliary Diagnosis of Atrial Septal Defect Based on View Classification

A prediction model of echocardiographic variables to screen for potentially correctable shunts in adult ASD-PAH patients

Atrial Septal Defect Detection in Children Based on Ultrasound Video Using Multiple Instances Learning

Echocardiographic Screening Model for Improved Assessment of Atrial Septal Defect Closure: A Multicenter Retrospective Study

Differential diagnosis of congenital ventricular septal defect and atrial septal defect in children using deep learning–based analysis of chest radiographs

Echocardiography-based AI for detection and quantification of atrial septal defect

Residual networks models detection of atrial septal defect from chest radiographs

Artificial intelligence-enabled 8-lead ECG detection of atrial septal defect among adults: a novel diagnostic tool

Anatomical attention-based prediction of postoperative pulmonary venous obstruction via CTA images

Building A Risk Prediction Model for Postoperative Pulmonary Vein Obstruction via Quantitative Analysis of CTA Images

Prediction of Central Venous Catheter-Associated Deep Venous Thrombosis in Pediatric Critical Care Settings.

A Machine Learning-Based Pulmonary Venous Obstruction Prediction Model Using Clinical Data and CT Image

Deep learning-based computer-aided heart sound analysis in children with left-to-right shunt congenital heart disease

Diagnosis of Fetal Total Anomalous Pulmonary Venous Connection Based on the Post‐left Atrium Space Ratio Using Artificial Intelligence

Prediction of atrial fibrillation in patients with embolic stroke with undetermined source using electrocardiogram deep learning algorithm and clinical risk factors

A Noninvasive Prediction Model With Simple Echocardiographic Variables for Shunts Closure Possibility in Patients With Posttricuspid Valve Shunt Defect