Research Article
BibTex RIS Cite

Real Tıme Detection of S1 and S2 Heart Sounds

Year 2020, Volume: 3 Issue: 2, 62 - 68, 31.08.2020

Abstract

Introduction
Automatic detection of S1 and S2 heart sounds is critical for diagnostic decision support systems that use heart sound as a means for decision making. There were previously suggested methods in the literature but offline nature of these analysis is impractical since the output is needed during the auscultation not later.
The aim of this study was to provide an algorithm for real-time detection of S1 and S2.
Materials and Methods
A total of 25 patients were included for the study. Group 1 consisted of healthy individuals. Group 2 consists of patients with systolic murmurs, diastolic murmurs, physiological or paradoxical splitting. Group 3 consisted of pathological atrial and ventricular gallops.
The suggested method first filtered the audio data then an envelope is constructed from the signal energy. The standard deviation of the envelope is employed as a threshold value for peak detection. Consecutive three peak values are utilized to estimate current and future locations of S1 and S2 and heart rate. These future estimates are used to optimize misinterpretations in S1 and S2 locations of the next heart cycle.
Results
Group 1 included 24% of the study group; Group 2 included 64% and Group 3 included 12% of the patients. The detection rate was 92%, 75% and 46% for Group 1, Group 2 and Group 3 patients, respectively. The overall success rate for all study population was 75%.
Conclusion
In this study, the feasibility of real time detection of S1 and S2 is shown. The method achieved 75 % success rate in Group 2 patients, although S1 and S2 sounds were barely visible in most of the cases. The fall in success rate in Group 3 patients is consistent with the findings in literature, since S3 and S4 are usually misinterpreted as S1 and S2 in severe gallop cases.

Thanks

We thank Emre Turgay for his exceptional contributions on creating the algorithm and software of this study.

References

  • 1. Hu XJ, Ma XJ, Zhao QM, et al. Pulse Oximetry and Auscultation for Congenital Heart Disease Detection. Pediatrics. 2017;140(4):e20171154. doi:10.1542/peds.2017-1154
  • 2. Iversen K, Søgaard Teisner A, Dalsgaard M, et al. Effect of teaching and type of stethoscope on cardiac auscultatory performance. Am Heart J. 2006;152(1):85.e1‐85.e857. doi:10.1016/j.ahj.2006.04.013
  • 3. Malarvili MB, Kamarulafizam I, Hussain SZ, et al. Heart sound segmentation algorithm based on instantaneous energy of electrocardiogram. Computers in Cardiology, 2003; 327-330.
  • 4. El-Segaier M, Lilja O, Lukkarinen S, Sörnmo L, Sepponen R, Pesonen E. Computer-based detection and analysis of heart sound and murmur. Ann Biomed Eng. 2005;33(7):937‐942. doi:10.1007/s10439-005-4053-3
  • 5. Liang H, Lukkarinen S, Hartimo I. Heart sound segmentation algorithm based on heart sound envelogram. Computers in Cardiology 1997; 105-108. doi: 10.1109/CIC.1997.647841
  • 6. Kumar D, Carvalho P, Antunes M, et al. Detection of S1 and S2 heart sounds by high frequency signatures. Conf Proc IEEE Eng Med Biol Soc. 2006;2006:1410‐1416. doi:10.1109/IEMBS.2006.260735
  • 7. Hebden JR,Torry JN. Neural network and conventional classifiers to distinguish between first and second heart sounds. IEE Colloquium on Artificial Intelligence Methods for Biomedical Data Processing, London, UK, 1996. 1996;3/1-3/6, doi: 10.1049/ic:19960638.
  • 8. Stasis AC, Loukis E, Pavlopoulos S, et al. Using decision tree algorithms as a basis for a heart sound diagnosis decision support system. 4th International IEEE EMBS Special Topic Conference on Information Technology Applications in Biomedicine, 2003. 2003;354-357.
  • 9. Tamer Ö, Dokur Z. Classification of heart sounds using an artificial neural network. Pattern Recognition Letters. 2003;24(1-3):617-623
  • 10. Kumar D, Carvalho P, Antunes M, et al. A New Algorithm for detection of S1 and S2 heart sounds. 2006 IEEE International Conference on Acoustics Speech and Signal Processing Proceedings, Toulouse, France, 2006. 2006;2:1180-1183. doi: 10.1109/ICASSP.2006.1660559.
  • 11. Carvalho P, Gil P , Henriques M, et al. Low Complexity Algorithm for Heart Sound Segmentation using the Variance Fractal Dimension. IEEE Int. Symp. on Intelligent Signal Processing, Faro, 2005. 2005:194-199, doi: 10.1109/WISP.2005.1531657.
  • 12. Gavrovska A, Goran Z, Vesna B, et al. Identification of S1 and S2 Heart Sound Patterns Based on Fractal Theory and Shape Context. Complexity 2017; 2017: 1580414:1-1580414:9.
  • 13. Hassani K, Bajelani K, Navidbakhsh M, et al. (2014) Detection and Identification of S1 and S2 Heart Sounds Using Wavelet Decomposition and Reconstruction. In: Roa Romero L. (eds) XIII Mediterranean Conference on Medical and Biological Engineering and Computing 2013. IFMBE Proceedings, vol 41. Springer, Cham. Doi: https://doi.org/10.1007/978-3-319-00846-2_126

S1 ve S2 Kalp Seslerinin Gerçek Zamanlı Tespiti

Year 2020, Volume: 3 Issue: 2, 62 - 68, 31.08.2020

Abstract

Giriş
S1 ve S2 kalp seslerinin otomatik algılanması, kalp sesini karar verme aracı olarak kullanan teşhis karar destek sistemleri için kritik öneme sahiptir. Literatürde offline çalışan karar verme yöntemleri mevcuttur fakat bu yöntemlerde analiz daha sonra yapılmaktadırlar. Bu yöntemler oskültasyonla eş zamanlı olmadıklarından da kullanımı pratik değildir.
Bu çalışmanın amacı S1 ve S2'nin gerçek zamanlı tespiti için bir algoritma geliştirilmesidir.
Materyal ve Metod
Toplam 25 hasta çalışmaya dahil edilmiştir. Grup 1 sağlıklı bireylerden oluşmaktadır. Grup 2, sistolik üfürümleri, diyastolik üfürümleri, fizyolojik veya paradoksal bölünmesi olan hastalardan oluşmaktadır. Grup 3, patolojik atriyal ve ventriküler gallop içermektedir.
Önerdiğimiz yöntem önce ses verilerini filtrelemekte, sonrasında sinyal enerjisinden bir zarf oluşturmaktadır. Zarfın standart sapması, tepe tespiti için bir eşik değeri olarak kullanılmaktadır. S1 ve S2'nin şimdiki ve gelecekteki yerlerini ve kalp hızını tahmin etmek için ardışık üç tepe değeri kullanılmaktadır. Bu tahminler, bir sonraki kalp döngüsünün S1 ve S2 konumlarındaki yanlış yorumlamaları optimize etmek için kullanılır.
Bulgular
Grup 1, çalışma grubunun % 24'ünü; Grup 2 hastaların % 64'ünü ve Grup 3 hastaların% 12'sini oluşturmaktadır. Grup 1, Grup 2 ve Grup 3 hastalarında tespit oranı sırasıyla %92, %75 ve %46’dır. Tüm çalışma populasyonu için toplam başarı oranı %75'tir.
Sonuç
Bu çalışmada, S1 ve S2'nin gerçek zamanlı tespitinin fizibilitesi gösterilmiştir. S1 ve S2 sesleri çoğu vakada belirgin olmamasına rağmen, yöntem Grup 2 hastalarında % 75 başarı oranına ulaşmıştır. Grup 3 hastalarda başarı oranındaki düşüş görülmektedir. Bu durum ileri gallop vakalarında S3 ve S4’ün genellikle S1 ve S2 olarak yanlış yorumlanması nedenli olmaktadır ve bu sonuçlar literatürle uyumludur.

References

  • 1. Hu XJ, Ma XJ, Zhao QM, et al. Pulse Oximetry and Auscultation for Congenital Heart Disease Detection. Pediatrics. 2017;140(4):e20171154. doi:10.1542/peds.2017-1154
  • 2. Iversen K, Søgaard Teisner A, Dalsgaard M, et al. Effect of teaching and type of stethoscope on cardiac auscultatory performance. Am Heart J. 2006;152(1):85.e1‐85.e857. doi:10.1016/j.ahj.2006.04.013
  • 3. Malarvili MB, Kamarulafizam I, Hussain SZ, et al. Heart sound segmentation algorithm based on instantaneous energy of electrocardiogram. Computers in Cardiology, 2003; 327-330.
  • 4. El-Segaier M, Lilja O, Lukkarinen S, Sörnmo L, Sepponen R, Pesonen E. Computer-based detection and analysis of heart sound and murmur. Ann Biomed Eng. 2005;33(7):937‐942. doi:10.1007/s10439-005-4053-3
  • 5. Liang H, Lukkarinen S, Hartimo I. Heart sound segmentation algorithm based on heart sound envelogram. Computers in Cardiology 1997; 105-108. doi: 10.1109/CIC.1997.647841
  • 6. Kumar D, Carvalho P, Antunes M, et al. Detection of S1 and S2 heart sounds by high frequency signatures. Conf Proc IEEE Eng Med Biol Soc. 2006;2006:1410‐1416. doi:10.1109/IEMBS.2006.260735
  • 7. Hebden JR,Torry JN. Neural network and conventional classifiers to distinguish between first and second heart sounds. IEE Colloquium on Artificial Intelligence Methods for Biomedical Data Processing, London, UK, 1996. 1996;3/1-3/6, doi: 10.1049/ic:19960638.
  • 8. Stasis AC, Loukis E, Pavlopoulos S, et al. Using decision tree algorithms as a basis for a heart sound diagnosis decision support system. 4th International IEEE EMBS Special Topic Conference on Information Technology Applications in Biomedicine, 2003. 2003;354-357.
  • 9. Tamer Ö, Dokur Z. Classification of heart sounds using an artificial neural network. Pattern Recognition Letters. 2003;24(1-3):617-623
  • 10. Kumar D, Carvalho P, Antunes M, et al. A New Algorithm for detection of S1 and S2 heart sounds. 2006 IEEE International Conference on Acoustics Speech and Signal Processing Proceedings, Toulouse, France, 2006. 2006;2:1180-1183. doi: 10.1109/ICASSP.2006.1660559.
  • 11. Carvalho P, Gil P , Henriques M, et al. Low Complexity Algorithm for Heart Sound Segmentation using the Variance Fractal Dimension. IEEE Int. Symp. on Intelligent Signal Processing, Faro, 2005. 2005:194-199, doi: 10.1109/WISP.2005.1531657.
  • 12. Gavrovska A, Goran Z, Vesna B, et al. Identification of S1 and S2 Heart Sound Patterns Based on Fractal Theory and Shape Context. Complexity 2017; 2017: 1580414:1-1580414:9.
  • 13. Hassani K, Bajelani K, Navidbakhsh M, et al. (2014) Detection and Identification of S1 and S2 Heart Sounds Using Wavelet Decomposition and Reconstruction. In: Roa Romero L. (eds) XIII Mediterranean Conference on Medical and Biological Engineering and Computing 2013. IFMBE Proceedings, vol 41. Springer, Cham. Doi: https://doi.org/10.1007/978-3-319-00846-2_126
There are 13 citations in total.

Details

Primary Language English
Subjects Cardiovascular Surgery
Journal Section Articles
Authors

Özge Turgay Yıldırım 0000-0002-6731-4958

Ayşegül Turgay 0000-0001-6207-1101

Publication Date August 31, 2020
Acceptance Date June 17, 2020
Published in Issue Year 2020 Volume: 3 Issue: 2

Cite

APA Turgay Yıldırım, Ö., & Turgay, A. (2020). Real Tıme Detection of S1 and S2 Heart Sounds. Journal of Cukurova Anesthesia and Surgical Sciences, 3(2), 62-68.

download

You are free to:
Share — copy and redistribute the material in any medium or format The licensor cannot revoke these freedoms as long as you follow the license terms.
Under the following terms: Attribution — You must give appropriate credit, provide a link to the license, and indicate if changes were made. You may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use. NonCommercial — You may not use the material for commercial purposes. NoDerivatives — If you remix, transform, or build upon the material, you may not distribute the modified material. No additional restrictions — You may not apply legal terms or technological measures that legally restrict others from doing anything the license permits.