What does this research mean for the field?

A deep learning model applied to heart sounds acquired via an electronic stethoscope can non-invasively detect abnormal hemodynamic parameters for pulmonary hypertension screening. Novelty: ClaimNovelty.METHODOLOGICAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

What question did this study set out to answer?

To develop a non-invasive deep learning model for screening pulmonary hypertension using heart sounds validated against right heart catheterization.

May 20, 2026American Journal of Respiratory and Critical Care Medicine

C18-03 Developing and Evaluating Deep Learning Model for Pulmonary Hypertension Screening via Heart Sounds—With RHC as Gold Standard: A Multi-Center Prospective Study

View Full Paper

Ask AI

Bookmark

To develop a non-invasive deep learning model for screening pulmonary hypertension using heart sounds validated against right heart catheterization.
Conducted a multi-center prospective study across 5 hospitals in China.
Enrolled 384 patients undergoing right heart catheterization and collected clinical data along with heart sound recordings.
Developed Masked Autoencoder pre-trained classification/regression models with 5-fold cross-validation.
The model achieved an F1 score of 83.09±2.06 and an AUC of 77.66±3.89 for detecting mPAP >20 mmHg.
Adjusting the mPAP classification threshold improved specificity and negative predictive value.
The model also outputs pulmonary vascular resistance, aiding in pulmonary hypertension subtyping.

Discussion

No takes yet. Share an insight, caveat, or question.

Implication

Multi-center prospective study develops deep learning model for PH screening using heart sounds, indicating it may enhance access to diagnostics.

Cite This Study

Liao et al. (2026) studied this question.

synapsesocial.com/papers/6a0d4f19f03e14405aa9a57b https://doi.org/10.1093/ajrccm/aamag162.5757

View Full Paper

Ask AI

Bookmark

Implication

Multi-center prospective study develops deep learning model for PH screening using heart sounds, indicating it may enhance access to diagnostics.

C18-03 Developing and Evaluating Deep Learning Model for Pulmonary Hypertension Screening via Heart Sounds—With RHC as Gold Standard: A Multi-Center Prospective Study

Key Points

Discussion

Implication

Cite This Study

C18-03 Developing and Evaluating Deep Learning Model for Pulmonary Hypertension Screening via Heart Sounds—With RHC as Gold Standard: A Multi-Center Prospective Study

Key Points

Discussion

Implication

Cite This Study

Keep reading

Keep reading