Detection of the first and second heart sound using probabilistic models

In this study, a methodology based on hidden Markov models (HMM) as a probabilistic finite state-machine to model systolic and diastolic interval duration is proposed. The detection of the first (S1) and second (S2) heart sound is performed using a network of two HMM's with grammar constraints to parse sequence of systolic and diastolic intervals. Duration modeling was considered in the HMM model architecture selection based on experimental measurements of systolic and diastolic intervals in normal subjects. Feature extraction of heart sound signals was based on time-cepstral features. Results are presented in terms of detection performance compared with QRS peak annotations of the simultaneous ECG recording. The performance of the proposed approach has been evaluated in 80 subjects. The results showed that the system was effective to detect the first and second heart sounds with sensitivity of 95% and a positive predictive value of 97% and thus provides a promising methodology for heart sound analysis.

[1]  Mohammad S. Obaidat,et al.  Performance of the short-time Fourier Transform and Wavelet Transform to phonocardiogram signal analysis , 1992, SAC '92.

[2]  J. N. Torry,et al.  Neural network and conventional classifiers to distinguish between first and second heart sounds , 1996 .

[3]  R. Doraiswami,et al.  Wavelet based bank of correlators approach for phonocardiogram signal classification , 1998, Proceedings of the IEEE-SP International Symposium on Time-Frequency and Time-Scale Analysis (Cat. No.98TH8380).

[4]  I. Hartimo,et al.  Heart sound segmentation algorithm based on heart sound envelogram , 1997, Computers in Cardiology 1997.

[5]  D. Haussler,et al.  Hidden Markov models in computational biology. Applications to protein modeling. , 1993, Journal of molecular biology.

[6]  Biing-Hwang Juang,et al.  Hidden Markov Models for Speech Recognition , 1991 .

[7]  Darren J. Wilkinson,et al.  Detecting homogeneous segments in DNA sequences by using hidden Markov models , 2000 .

[8]  A. Poritz,et al.  Hidden Markov models: a guided tour , 1988, ICASSP-88., International Conference on Acoustics, Speech, and Signal Processing.

[9]  Horst Bunke,et al.  Hidden Markov models: applications in computer vision , 2001 .

[10]  Minfen Shen,et al.  The analysis and classification of phonocardiogram based on higher-order spectra , 1997, Proceedings of the IEEE Signal Processing Workshop on Higher-Order Statistics.

[11]  Raymond L. Watrous,et al.  Detection of the first heart sound using a time-delay neural network , 2002, Computers in Cardiology.

[12]  A. Haghighi-Mood,et al.  A sub-band energy tracking algorithm for heart sound segmentation , 1995, Computers in Cardiology 1995.

[13]  Seung Hong Hong,et al.  Comparison between short time Fourier and wavelet transform for feature extraction of heart sound , 1999, Proceedings of IEEE. IEEE Region 10 Conference. TENCON 99. 'Multimedia Technology for Asia-Pacific Information Infrastructure' (Cat. No.99CH37030).

[14]  Joseph D. Bronzino,et al.  The Biomedical Engineering Handbook , 1995 .