Detection and Adaptive Cancellation of Heart Sound Interference in Tracheal Sounds

Heart sound is often a serious source of noise in the analysis of respiratory sounds. The interference of heart sound is important when breath sounds are analyzed, since the intensity of heart sound is 1 to 10 times greater than the intensity of breath sound. Methods to reduce heart sound interference in breath sound analysis had been suggested. However, the majority of those methods require acquisition of an additional signal and knowledge of the spectral content of heart sound. In this paper it has been proposed an algorithm for the detection and adaptive filtering of heart sounds using only the sound signal and, minimizing distortion in the respiratory sound. The main components of the power spectral density (PSD) of the filtered tracheal sound are comparable with the PSD of the original signal. The experimental results shown reduction of the heart sound interference when the adaptive filter is applied. This kind of filter could facilitate the analysis and a better interpretation of the respiratory sound during spontaneous ventilation