Key result
Machine learning model integrating echo and clinical variables predicts pulmonary hypertension with 0.997 AUC.
Why the study?
Pulmonary hypertension leads to right heart failure and early detection is critical for improving patient outcomes, motivating the development of a novel machine learning-based diagnostic model.
Does a machine learning-based diagnostic model combining echocardiographic and clinical variables accurately predict the risk of pulmonary hypertension?
Cohort
Does a machine learning-based diagnostic model combining echocardiographic and clinical variables accurately predict the risk of pulmonary hypertension?
Effect estimate: AUC 0.997
A novel machine learning model combining echocardiographic and clinical variables demonstrated exceptional accuracy (AUC 0.997) for the early diagnosis of pulmonary hypertension.
BACKGROUND Pulmonary hypertension (PH) is a progressive disorder characterized by elevated pulmonary artery pressure and increased pulmonary vascular resistance, ultimately leading to right heart failure. Early detection is critical for improving patient outcomes. OBJECTIVE To establish a novel machine learning-based diagnostic model for PH. METHODS A diagnostic model for the early detection of pulmonary hypertension (PH) was developed through a two-step approach. First, Recursive Feature Elimination (RFE) was employed to select the most relevant echocardiographic variables, which were subsequently integrated into a composite ultrasound index using machine learning techniques such as XGBoost. In the second step, this ultrasound index was integrated with clinical variables identified through LASSO regression. Together, these elements were combined to construct a logistic regression model for diagnosis. The model’s performance was rigorously evaluated using ROC curves, calibration plots, and decision curve analysis (DCA) to ensure its clinical relevance and accuracy. RESULTS Machine learning identified key echocardiographic and clinical predictors, with the XGBoost model achieving high AUC, sensitivity, and specificity. LASSO regression identified critical clinical variables, including prothrombin time activity and serum cystatin C. The diagnostic model demonstrated high predictive accuracy, with an AUC of 0.997. Calibration and decision curve analyses indicated close alignment between predicted and observed outcomes validating the model’s clinical value, especially at higher risk thresholds. CONCLUSIONS This model enhances early pulmonary hypertension (PH) diagnosis through a non-invasive approach and demonstrates strong predictive accuracy. It facilitates early intervention and personalized treatment, with potential applications in broader cardiovascular disease management. CLINICALTRIAL The study was approved by the Research Ethics Commission of Wuhan Zhongnan Hospital and the requirement for informed consent was waived by the Ethics Commission (Approval No.2023185)
No takes yet. Share an insight, caveat, or question.
Cheng et al. (2025) conducted a cohort in Pulmonary hypertension. Machine learning-based diagnostic model was evaluated on Diagnostic accuracy for early detection of pulmonary hypertension (AUC 0.997). A machine learning-based diagnostic model integrating echocardiographic and clinical variables demonstrated high predictive accuracy for pulmonary hypertension with an AUC of 0.997.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: