A scientific study published on Nature on October 13, 2025, presents an advanced Artificial Intelligence (AI) system that can automatically detect pulmonary arterial hypertension (PAH) in infants under one year old who underwent Ventricular Septal Defect (VSD) surgery. The researchers analysed data from over 1,300 paediatric patients treated at the Women and Children’s Hospital, Qingdao University and Qingdao Central Hospital from January 2017 to December 2023 and developed a multimodal multipath Artificial Intelligence model that integrates echocardiography and chest radiography images. This model achieved excellent diagnostic accuracy (AUC > 0.9), outperforming single-modality approaches. The Artificial Intelligence system shows promise for early, non-invasive detection of pulmonary arterial hypertension, reducing reliance on invasive cardiac catheterisation and supporting clinicians in making faster, evidence-based treatment decisions.
Read more at this link on Nature
Citation
Luo, G., Liu, H., Li, Z. et al. A multimodal multipath AI system for assessing PAH after VSD correction on echocardiography and chest radiography images. Sci Rep 15, 35676 (2025). https://doi.org/10.1038/s41598-025-19584-5
Summary by Deger Kesimoglu, volunteer for the Alliance for Pulmonary Hypertension