Five-year survival in Pulmonary Hypertension patients: risk stratification by a machine learning approach.

Abstract

<bold>Introduction.</bold> Pulmonary hypertension (PH) is a condition characterized by increased pressure in the pulmonary arteries. Because of its poor prognosis, an optimal management of this disease is necessary. <bold>Objectives.</bold> The study aim was to search for PH phenotypes and develop a predictive model of five-year mortality using machine learning (ML) algorithms. <bold>Methods.</bold> This multicenter study was conducted on 122 PH patients. Clinical and demographic data were collected and, then, used to identify phenotypes through clustering. Subsequently, predictive model was performed by different ML algorithms. <bold>Results.</bold> Three PH clusters were identified: Cluster 1 includes 57% females, mean age of 68.57±10.54 years, 69% from non-respiratory PH groups, and better cardiac (NYHA class 2.61±0.84) and respiratory function (FEV1% 78.78±21.54); Cluster 2 includes 50% females, mean age of 71.36±8.32 years, 44% from PH group 3, worse respiratory function (FEV1% 68.12±10.20), intermediate cardiac function (NYHA class 3.18±0.49) and significantly higher mortality (75%); Cluster 3 represents the youngest cluster (mean age of 61.11±13.50 years) with 65% males, 81% from non-respiratory PH groups, intermediate respiratory function (FEV1% 70.51±17.91) and worse cardiac performance (NYHA class 3.22±0.58). After testing ML models, logistic regression showed the best predictive performance (AUC=0.835 and accuracy=0.744) and identified three mortality-risk factors: age, NYHA class, number of medications taken. <bold>Conclusions</bold>. The results suggest that the integration of ML into clinical practice can improve risk stratification to optimize treatment strategies and enhance the outcome of PH patients.

Ähnliche Arbeiten