Cardiopulmonary sound analysis is essential for the early detection and diagnosis of respiratory and cardiovascular diseases. This work explores the application of Quantum Machine Learning (QML) models, specifically Quantum Support Vector Machines (QSVM) and Quantum Neural Networks (QNN), for the classification of both lung and heart sounds. Leveraging MFCC-based features and dimensionality reduction techniques, we evaluate the performance of these models on two publicly available benchmark datasets. The experimental results indicate that QML models match or surpass their classical counterparts, particularly under constraints of limited training data and reduced feature sets. These findings underscore the potential of QML as a promising tool for efficient, accurate and unified analysis of cardiopulmonary acoustic signals in next-generation diagnostic systems.