Unravelling the potential of phase portrait in the auscultation of mitral valve dysfunction
Department of Optoelectronics, University of Kerala, 695581, Trivandrum, Kerala, India
Accepted: 1 February 2021
Published online: 5 February 2021
The manuscript elucidates the potential of phase portrait, fast Fourier transform, wavelet, and time-series analyses of the heart murmur (HM) of normal (healthy) and mitral regurgitation (MR) in the diagnosis of valve-related cardiovascular diseases. The temporal evolution study of phase portrait and the entropy analyses of HM unveil the valve dysfunction-induced haemodynamics. A tenfold increase in sample entropy in MR from that of normal indicates the valve dysfunction. The occurrence of a large number of frequency components between lub and dub in MR, compared to the normal, is substantiated through the spectral analyses. The machine learning techniques, K-nearest neighbour, support vector machine, and principal component analyses give 100% predictive accuracy. Thus, the study suggests a surrogate method of auscultation of HM that can be employed cost-effectively in rural health centres.
© The Author(s), under exclusive licence to Società Italiana di Fisica and Springer-Verlag GmbH Germany, part of Springer Nature 2021