An approach towards a heartbeat sound information retrieval system

Interpretation of heart sounds can be a problematic and difficult task for cardiology specialists. Diagnosis of heart diseases requires special skill and clinical experience along with detailed and expensive tests. However, heart disease diagnosis by heartbeat sounds is preferable and still widely used as the first step of diagnosis. Recently, Computer aided auscultation has emerged as a cost-effective technique to analyze and interpret the heart sounds. In this study we propose a feasible technique for developing a heartbeat sound retrieval system using text based approaches useful towards automated heart disease detection. The feasibility of these text based retrieval approaches are shown from retrieval experiments with around 100 digital heart sound recordings and eight categories of heart diseases. Each recording include ten heartbeat cycles.