scientific article; zbMATH DE number 1193377
Some applications of Fourier and wavelet transforms to phonocardiogram signal analysis are presented. The short-time Fourier transform is obtained from the usual Fourier transform by multiplying the time signal by an appropriate sliding time window. To define the wavelet transform of a signal, one has to consider an analyzing wavelet satisfying some requirements (continuity, integrability, square integrability, possitivity, admissibility, ...). The most adequate analyzing wavelet for time-frequency analysis is the complex exponential modulated Gaussian function, i.e., the Gabor wavelet. The wavelet transform gives better time resolution than the short-time Fourier transform and Gabor transform for broad band signals. However, short-time Fourier transform gives better frequency resolutions. Using these ideas, an algorithm is presented to study nonstationary signals by combining two transforms in order to get better resolutions in time and frequency. First, the nonstationary signal is decomposed into subband stationary signals using the wavelet transform with adequate scales and then to each of these stationary subband signals the inverse wavelet transform is applied to find its spectrum by using the classical Fourier transform. This paper is a clear exponent of the importance and relevance of mathematics and mathematical techniques in the medical sciences. Medical considerations: The use of phonocardiograms is a medical technique to diagnose heart diseases. The physician should interpret the heart blood flow by auscultation. Abnormal sounds may contain murmurs and aberrations due to some pathologies of the cardiovascular system. For instance, turbulent flow inside an aneurysm is evidenced by reduced bruits [\textit{S.M. Chitanvis}, \textit{G. Hademenos} and \textit{W.J. Powers}, Hemodynamical assessment of the development and rupture of intracraneal aneurysms using computational simulation. Neurol. Res. 17, 426-34 (1995)], and different implications of the sounds due to intracraneal aneurysms are considered by \textit{T.D. Mast} and \textit{A.D. Pierce} [A theory of aneurysms sounds. J. Biomech. 28, 1045-1053 (1995)], and \textit{J.J. Nieto} and \textit{A. Torres} [Approximation of solutions for nonlinear problems with an application to the study of aneurysms of the circle of Willis. Nonlinear Anal. 40, 513-521 (2000); A nonlinear biomathematical model for the study of intracraneal aneurysms. J. Neurol. Sci. (to appear)]. Those aberrations are difficult to interpret and may confuse the physician. In consequence, a mathematical analysis of the sounds should be helpful and that is what is studied in this paper. For the entire collection see [Zbl 0889.00018].
This page was built for publication:
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3841841)