Lung sounds auscultation is often the first noninvasive resource for detection and
discrimination of respiratory pathologies available to the physician through the use of the
stethoscope. Hearing interpretation though was the only means of appreciation of the lung
sounds diagnostic information for many decades. Nevertheless in recent years computerized
auscultation combined with signal processing techniques has boosted the diagnostic capabilities
of lung sounds. The latter were traditionally analyzed and characterized by morphological
changes in the time domain using statistical measures by spectral properties in the frequency
domain using simple spectral analysis or by nonstationary properties in a joint time-frequency
domain using short-time Fourier transform. Advanced signal processing techniques however have
emerged in the last decade broadening the perspective in lung sounds analysis. The scope of
this book is to present up-to-date signal processing techniques that have been applied to the
area of lung sound analysis. It starts with a description of the nature of lung sounds and
continues with the introduction of new domains in their representation new denoising
techniques and concludes with some reflective implications both from engineers' and
physicians' perspective. Issues of nonstationarity nonlinearity non-Gaussianity modeling
and classification of lung sounds are addressed with new methodologies revealing a more
realistic approach to their pragmatic nature. Advanced denoising techniques that effectively
circumvent the noise presence (e.g. heart sound interference background noise) in lung sound
recordings are described providing the physician with high-quality auscultative data. The book
offers useful information both to engineers and physicians interested in bioacoustics clearly
demonstrating the current trends in lung sound analysis. Table of Contents: The Nature of Lung
Sound Signals New Domains in LS Representation Denoising Techniques Reflective
Implications
