ARTIFICIAL intelligence (AI)–guided echocardiography was shown to accurately predict mortality in patients with pulmonary hypertension and right heart failure, offering a potential new tool for risk stratification in a high-risk clinical population.
Pulmonary hypertension is a progressive condition characterised by elevated pulmonary arterial pressure, leading to right ventricular dysfunction and, ultimately, right heart failure. Despite advances in imaging, accurately assessing right ventricular function remains challenging because of its complex geometry and load dependence. Right ventricular strain has emerged as a sensitive marker of myocardial dysfunction, but its integration into routine prognostic models has been limited.
AI Echocardiography Improves Right Ventricular Risk Prediction
In this large retrospective study, researchers analysed data from 586 patients with pulmonary hypertension and right heart failure who underwent transthoracic echocardiography either before or during hospital admission. Using a multimodal deep-learning framework, the team combined clinical variables, conventional echocardiographic indices, and AI-extracted right ventricular strain features to predict both in-hospital and follow-up mortality.
Patients who died had significantly higher N-terminal pro-B-type natriuretic peptide levels, pulmonary artery systolic pressure, and average right ventricular longitudinal strain compared with survivors. Multivariable analysis showed that pulmonary artery systolic pressure and average right ventricular longitudinal strain independently predicted in-hospital mortality, while average right ventricular longitudinal strain alone predicted longer-term mortality during follow-up.
The AI echocardiography model demonstrated strong discriminative performance, achieving an area under the curve of 0.823 in the testing cohort. These findings suggest that AI-derived right ventricular strain adds clinically meaningful prognostic information beyond conventional echocardiographic measures.
Why This Matters for Clinical Care
Right ventricular failure is the leading cause of death in pulmonary hypertension, yet early identification of patients at highest risk remains difficult. By automating strain analysis and integrating multiple data streams, AI echocardiography may help clinicians make faster and more accurate prognostic assessments, particularly in busy inpatient settings.
Limitations and Future Directions
The study was conducted at a single centre, which may limit generalisability. In addition, echocardiographic data were acquired at different clinical time points, potentially introducing variability. Prospective, multicentre studies are needed to confirm whether AI echocardiography–guided risk stratification can directly improve clinical outcomes.
If validated, this multimodal framework could be incorporated into routine echocardiographic workflows, supporting earlier intervention and more individualised management for patients with pulmonary hypertension and right heart failure.
Reference
Mou H et al. AI–assisted multimodal assessment for right ventricular function from echocardiography predicts mortality in patients with pulmonary hypertension and right heart failure. Sci Rep. 2026; DOI:10.1038/s41598-026-36533-y.
