A method of construction of information images of the acoustic signals of the human bronchopulmonary system

• Format: Conference Proceeding
• Language: English
• Subjects: Acoustic noise; Acoustics; Algorithms; Bronchi; Construction methods; Parameter modification; Signal analysis; Weighting functions
• ISSN: 0094-243X; 1551-7616
• DOI: 10.1063/1.4936043

The present study focuses on the development of a method of identification of respiratory sounds and noises of a human naturally and in various pathological conditions. The existing approaches based on a simple method of frequency and time signal analysis, have insufficient specificity, efficiency and unambiguous interpretation of the results of a clinical study. An algorithm for a phase selection of respiratory cycles and analysis of respiratory sounds resulting from bronchi examination of a patient has been suggested. The algorithm is based on the method of phase timing analysis of bronchi phonograms. The results of the phase-frequency algorithm with high resolution reflects a time position of the traceable signals and the individual structure of recorded signals. This allows using the proposed method for the formation of information images (models) of the diagnostically significant fragments. A weight function, frequency parameters of which can be selectively modified, is used for this purpose. The vision of the weighting function is specific to each type of respiratory noise, traditionally referred to quality characteristics (wet or dry noise, crackling, etc.).