Noninvasive diagnosis of pulmonary hypertension using heart sound analysis

Abstract

Right-heart catheterization is the most accurate method for measuring pulmonary artery pressure (PAP). It is an expensive, invasive procedure, exposes patients to the risk of infection, and is not suited for long-term monitoring situations. Medical researchers have shown that PAP influences the characteristics of heart sounds. This suggests that heart sound analysis is a potential method for the noninvasive diagnosis of pulmonary hypertension. We describe the development of a prototype system, called PHD (pulmonary hypertension diagnoser), that implements this method. PHD uses patient data with machine learning algorithms to build models of how pulmonary hypertension affects heart sounds. Data from 20 patients were used to build the models and data from another 31 patients were used as a validation set. PHD diagnosed pulmonary hypertension in the validation set with 77% accuracy and 0.78 area under the receiver-operating-characteristic curve.