Analysis of the second heart sound fordiagnosis of paediatric heart disease

Abstract

The second heart sound (S2) consists of two major components, one due to the closure of the aortic valve (A2) and the other due to the closure of the pulmonary valve (P2). The aortic valve normally closes before the pulmonary valve and leads to a time delay between the two sounds. This delay is known as the ‘split’ in the medical community and is of significant diagnostic importance. The authors aim to develop an automatic technique to measure the split and to compare two common splitting patterns (i.e. variable splitting and fixed splitting), in a quantitative manner. A signal model for S2 is proposed. Accordingly, S2 is decomposed into a number of components with an algorithm based on the time–frequency distribution of the signal. A2 and P2 are selected from the model components and from these the split can be estimated. Groups of patients with the two splitting patterns have been investigated. For each patient, the splits of 20 successive cardiac cycles are measured, their mean and standard deviation are then calculated and used to characterise the two splitting patterns. It is found that the two simple statistical quantities can be used to identify the splitting patterns and hence offer important diagnostic information.