Author:
Amit Guy,Shukha Khuloud,Gavriely Noam,Intrator Nathan
Abstract
Heart sounds, the acoustic vibrations produced by the mechanical processes of the cardiac cycle, are modulated by respiratory activity. We have used computational techniques of cluster analysis and classification to study the effects of the respiratory phase and the respiratory resistive load on the temporal and morphological properties of the first (S1) and second heart sounds (S2), acquired from 12 healthy volunteers. Heart sounds exhibited strong morphological variability during normal respiration and nearly no variability during apnea. The variability was shown to be periodic, with its estimated period in good agreement with the measured duration of the respiratory cycle. Significant differences were observed between properties of S1 and S2 occurring during inspiration and expiration. S1 was commonly attenuated and slightly delayed during inspiration, whereas S2 was accentuated and its aortic component occurred earlier at late inspiration and early expiration. Typical split morphology was observed for S1 and S2 during inspiration. At high-breathing load, these changes became more prominent and occurred earlier in the respiratory cycle. Unsupervised cluster analysis was able to automatically identify the distinct morphologies associated with different respiratory phases and load. Classification of the respiration phase (inspiration or expiration) from the morphology of S1 achieved an average accuracy of 87 ± 7%, and classification of the breathing load was accurate in 82 ± 7%. These results suggest that quantitative heart sound analysis can shed light on the relation between respiration and cardiovascular mechanics and may be applied to continuous cardiopulmonary monitoring.
Publisher
American Physiological Society
Subject
Physiology (medical),Cardiology and Cardiovascular Medicine,Physiology
Reference29 articles.
1. Robust regression for periodicity detection in non-uniformly sampled time-course gene expression data
2. Cluster analysis and classification of heart sounds
3. Amit G, Gavriely N, Lessick J, Intrator N. Acoustic indices of cardiac functionality. In: International Conference on Bio-inspired Systems and Signal Processing (BIOSIGNALS). Setubal, Portugal: INSTICC, 2008, p. 77–83.
4. Benjamini Y, Hochberg Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J R Stat Soc Ser B 57: 289–300, 1995.
5. Time-frequency and time-scale techniques for the classification of native and bioprosthetic heart valve sounds
Cited by
25 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献