Robust detection of background noise in phonocardiograms

In this paper we present a method for detecting external and internal noises that can corrupt the phonocar-diogram (PCG). Using a reference signal from an accessory microphone placed on the stomach, the method first spectrally enhances this signal and then applies short-term log energy (STLE) to detect the noise. We also propose a robust method to determine a decision threshold to detect noise using the STLE. Our results show that spectral enhancement combined with STLE can efficiently detect noises, external and internal, that can corrupt the PCG.

[1]  Alexander Fischer,et al.  Quantile based noise estimation for spectral subtraction and Wiener filtering , 2000, 2000 IEEE International Conference on Acoustics, Speech, and Signal Processing. Proceedings (Cat. No.00CH37100).

[2]  Peter Hult,et al.  Feature Extraction for Systolic Heart Murmur Classification , 2006, Annals of Biomedical Engineering.

[3]  J. Hertzberg,et al.  Artificial Neural Network-Based Method of Screening Heart Murmurs in Children , 2001, Circulation.

[4]  John Semmlow,et al.  Acoustic detection of coronary artery disease. , 2007, Annual review of biomedical engineering.

[5]  Andrzej Drygajlo,et al.  Entropy based voice activity detection in very noisy conditions , 2001, INTERSPEECH.

[6]  Richard M. Schwartz,et al.  Enhancement of speech corrupted by acoustic noise , 1979, ICASSP.

[7]  S. Boll,et al.  Suppression of acoustic noise in speech using spectral subtraction , 1979 .

[8]  William Pavlicek,et al.  Dynamics of Diastolic Sounds Caused by Partially Occluded Coronary Arteries , 2009, IEEE Transactions on Biomedical Engineering.