Echocardiographic artificial intelligence for pulmonary hypertension classification

Objective: The classification of pulmonary hypertension (PH) is crucial for determining the appropriate therapeutic strategy. We investigated whether machine learning (ML) algorithms may assist in echocardiographic PH prediction, where current guidelines recommend integrating several different parameters.
Methods: We obtained physical and echocardiographic data from 885 patients who underwent right heart catheterization (RHC). Patients were classified into three groups: non-PH, pre-capillary PH, and post-capillary PH, based on values obtained from RHC. Utilizing 24 parameters, we created predictive models employing four different classifiers and selected the one with the highest area under the curve (AUC). We then calculated the macro-average classification accuracy for PH on the derivation cohort (n=720) and prospective validation dataset (n=165), comparing the results with guideline-based echocardiographic assessment obtained from each cohort.
Results: Logistic regression with elastic net regularization had the highest classification accuracy, with AUCs of 0.789, 0.766, and 0.742 for normal, pre-capillary PH, and post-capillary PH, respectively. The ML model demonstrated significantly better predictive accuracy than the guideline-based echocardiographic assessment in the derivation cohort (59.4% vs. 51.6%, p<0.01). In the independent validation dataset, the ML model's accuracy was comparable to the guideline-based PH classification (59.4% vs. 57.8%, p=0.638).
Conclusions: This preliminary study suggests promising potential for our ML model in predicting echocardiographic PH. Further research and validation are needed to fully assess its clinical utility in PH diagnosis and treatment decision-making.

This article has been accepted for publication in Heart, 2024 following peer review, and the Version of Record can be accessed online at https://doi.org/10.1136/heartjnl-2023-323320