Analyzing approaches to mathematic modeling of heart sounds

The article introduces the findings of the analysis of the existing approaches to the development of mathematical models of acoustic heart phenomena. The analysis of mathematical methods that can be used to model heart sounds has been performed with the use of reference signals from the 3M Open Library (Littmann Library) and a set of signals obtained by the authors during their previous scientific efforts. The analysis findings have allowed revealing the approaches and methods that are most suitable for developing the mathematical models of human phonocardiograms (normal and pathological) for further research efforts meant to develop methods to single out heart beats against the high level of interference and creating intervalograms to characterize the heart rate at the current moments of time. In addition to the generation of model phonocardiograms, the article reviews the methods to analyze model and real-life phonocardiograms with the assessment of an input from random and deterministic components.

Phonocardiogram classification using deep neural networks and weighted probability comparisons.

Cardiac auscultation is one of the most conventional approaches for the initial assessment of heart disease, however the technique is highly user-dependent and with low repeatability. Several computational approaches based on the analysis of the phonocardiograms (PCG) have been proposed to classify heart sounds into normal or abnormal, but most often do not achieve acceptable levels of sensitivity (Se) and specificity (Sp) or require the use of special hardware. We propose a novel approach for classification of PCG. First, the system makes use of deep neural networks for computing individual cardiac cycle probabilities, followed by classification using weighted probability comparisons. The system was tested on an extended dataset consisting of a balanced sample of 18179 normal and abnormal cycles, achieving Se and Sp values of 91.3% and 93.8% respectively. In addition, the system overcomes previous limitations since it was trained with a balanced sample; also, the decision factor used during the classification stage allows to control the trade-off between Se and Sp, making the proposed system suitable for clinical applications.