Limits...
A framework for automatic heart sound analysis without segmentation.

Yuenyong S, Nishihara A, Kongprawechnon W, Tungpimolrut K - Biomed Eng Online (2011)

Bottom Line: Geometric mean was used as performance index.Average classification performance using ten-fold cross-validation was 0.92 for noise free case, 0.90 under white noise with 10 dB signal-to-noise ratio (SNR), and 0.90 under impulse noise up to 0.3 s duration.Further work include building a new training set recorded from actual patients, then further evaluate the method based on this new training set.

View Article: PubMed Central - HTML - PubMed

Affiliation: Department of Communication and Integrated Systems, Tokyo Institute of Technology, Japan 2-12-1-W9-108 Ookayama, Meguro-ku, Tokyo, Japan. toey123@gmail.com

ABSTRACT

Background: A new framework for heart sound analysis is proposed. One of the most difficult processes in heart sound analysis is segmentation, due to interference form murmurs.

Method: Equal number of cardiac cycles were extracted from heart sounds with different heart rates using information from envelopes of autocorrelation functions without the need to label individual fundamental heart sounds (FHS). The complete method consists of envelope detection, calculation of cardiac cycle lengths using auto-correlation of envelope signals, features extraction using discrete wavelet transform, principal component analysis, and classification using neural network bagging predictors.

Result: The proposed method was tested on a set of heart sounds obtained from several on-line databases and recorded with an electronic stethoscope. Geometric mean was used as performance index. Average classification performance using ten-fold cross-validation was 0.92 for noise free case, 0.90 under white noise with 10 dB signal-to-noise ratio (SNR), and 0.90 under impulse noise up to 0.3 s duration.

Conclusion: The proposed method showed promising results and high noise robustness to a wide range of heart sounds. However, more tests are needed to address any bias that may have been introduced by different sources of heart sounds in the current training set, and to concretely validate the method. Further work include building a new training set recorded from actual patients, then further evaluate the method based on this new training set.

Show MeSH
Segmentation of heart sound with S3. Note the error caused by the third peak.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC3045988&req=5

Figure 5: Segmentation of heart sound with S3. Note the error caused by the third peak.

Mentions: This section illustrates problems that arise when an abnormal sound is segmented. Figure 5 shows the envelope signal of a heart sound with S3. Suppose that the three leftmost peaks are analysed, which actually correspond to S1, S2, and S3, respectively. The third peak is an extra peak such that it does not correspond to any FHS. Due to the presence of this peak (marked as "extra peak" in Figure 5), the length of p2-p3 interval has changed. It is no longer the distance between the second and the fourth peak, like it should be without the extra third peak, but the distance between the second and third peak. This is shorter than the p1-p2 interval and leads to the false conclusion that the second peak is S1. Using this peak as a reference to label all other peaks leads to wrong identification of all peaks in the envelope signal and ultimately wrong segmentation results.


A framework for automatic heart sound analysis without segmentation.

Yuenyong S, Nishihara A, Kongprawechnon W, Tungpimolrut K - Biomed Eng Online (2011)

Segmentation of heart sound with S3. Note the error caused by the third peak.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC3045988&req=5

Figure 5: Segmentation of heart sound with S3. Note the error caused by the third peak.
Mentions: This section illustrates problems that arise when an abnormal sound is segmented. Figure 5 shows the envelope signal of a heart sound with S3. Suppose that the three leftmost peaks are analysed, which actually correspond to S1, S2, and S3, respectively. The third peak is an extra peak such that it does not correspond to any FHS. Due to the presence of this peak (marked as "extra peak" in Figure 5), the length of p2-p3 interval has changed. It is no longer the distance between the second and the fourth peak, like it should be without the extra third peak, but the distance between the second and third peak. This is shorter than the p1-p2 interval and leads to the false conclusion that the second peak is S1. Using this peak as a reference to label all other peaks leads to wrong identification of all peaks in the envelope signal and ultimately wrong segmentation results.

Bottom Line: Geometric mean was used as performance index.Average classification performance using ten-fold cross-validation was 0.92 for noise free case, 0.90 under white noise with 10 dB signal-to-noise ratio (SNR), and 0.90 under impulse noise up to 0.3 s duration.Further work include building a new training set recorded from actual patients, then further evaluate the method based on this new training set.

View Article: PubMed Central - HTML - PubMed

Affiliation: Department of Communication and Integrated Systems, Tokyo Institute of Technology, Japan 2-12-1-W9-108 Ookayama, Meguro-ku, Tokyo, Japan. toey123@gmail.com

ABSTRACT

Background: A new framework for heart sound analysis is proposed. One of the most difficult processes in heart sound analysis is segmentation, due to interference form murmurs.

Method: Equal number of cardiac cycles were extracted from heart sounds with different heart rates using information from envelopes of autocorrelation functions without the need to label individual fundamental heart sounds (FHS). The complete method consists of envelope detection, calculation of cardiac cycle lengths using auto-correlation of envelope signals, features extraction using discrete wavelet transform, principal component analysis, and classification using neural network bagging predictors.

Result: The proposed method was tested on a set of heart sounds obtained from several on-line databases and recorded with an electronic stethoscope. Geometric mean was used as performance index. Average classification performance using ten-fold cross-validation was 0.92 for noise free case, 0.90 under white noise with 10 dB signal-to-noise ratio (SNR), and 0.90 under impulse noise up to 0.3 s duration.

Conclusion: The proposed method showed promising results and high noise robustness to a wide range of heart sounds. However, more tests are needed to address any bias that may have been introduced by different sources of heart sounds in the current training set, and to concretely validate the method. Further work include building a new training set recorded from actual patients, then further evaluate the method based on this new training set.

Show MeSH