Methods Inf Med 2014; 53(04): 291-295
DOI: 10.3414/ME13-02-0041
Focus Theme – Original Articles
Schattauer GmbH

Computerized Diagnosis of Respira tory Disorders

SVM Based Classification of VAR Model Parameters of Respiratory Sounds
I. Sen
1   Electrical and Electronics Engineering Department, Bogazici University, Istanbul, Turkey
,
M. Saraclar
1   Electrical and Electronics Engineering Department, Bogazici University, Istanbul, Turkey
,
Y. P. Kahya
1   Electrical and Electronics Engineering Department, Bogazici University, Istanbul, Turkey
› Institutsangaben
Weitere Informationen

Publikationsverlauf

received:15. Oktober 2013

accepted:19. Mai 2014

Publikationsdatum:
20. Januar 2018 (online)

Summary

Introduction: This article is part of the Focus Theme of Methods of Information in Medicine on “Biosignal Interpretation: Advanced Methods for Studying Cardiovascular and Respiratory Systems”.

Objectives: This work proposes an algorithm for diagnostic classification of multi-channel respiratory sounds.

Methods: 14-channel respiratory sounds are modeled assuming a 250-point second order vector autoregressive (VAR) process, and the estimated model parameters are used to feed a support vector machine (SVM) classifier. Both a three-class classifier (healthy, bronchi ectasis and interstitial pulmonary disease) and a binary classifier (healthy versus pathological) are considered.

Results: In the binary scheme, the sensitivity and specificity for both classes are 85% ± 8.2%. In the three-class classification scheme, the healthy recall (95% ± 5%) and the interstitial pulmonary disease recall and precision (100% ± 0% both) are rather high. However, bronchiectasis recall is very low (30% ± 15.3%), resulting in poor healthy and bronchiectasis precision rates (76% ± 8.7% and 75% ± 25%, respectively). The main reason behind these poor rates is that the bronchiectasis is confused with the healthy case.

Conclusions: The proposed method is promising, nevertheless, it should be improved such that other mathematical models, additional features, and/or other classifiers are to be experimented in future studies.

 
  • References

  • 1 Charleston-Villalobos S, Albuerne-Sanchez L, Gonzalez-Camarena R, Mejia-Avila M, Carrillo-Rodriguez G, Aljama-Corrales T. Linear and nonlinear analysis of base lung sound in extrinsic allergic alveolitis patients in comparison to healthy subjects. Methods Inf Med 2013; 5 (03) 266-276.
  • 2 Charleston-Villalobos S, Dorantes-Méndez G, González-Camarena R, Chi-Lem G, Carrillo J, Aljama-Corrales T. Acoustic thoracic image of crackle sounds using linear and nonlinear processing techniques. Med Biol Eng Comput 2011; 49 (01) 15-24.
  • 3 Sankur B, Kahya YP, Guler EC, Engin T. Comparison of AR based algorithms for respiratory sounds classification. Comput Biol Med 1994; 24 (01) 67-76.
  • 4 Ciftci K, Kahya YP. Respiratory airflow estimation by time varying autoregressive modeling. In: Proceedings of the 30th Annual International Conference of the IEEE-EMBS. 2008: 347-350.
  • 5 Alsmadi S, Kahya YP. Design of a DSP-based instrument for real-time classification of pulmonary sounds. Comput Biol Med 2008; 38 (01) 53-61.
  • 6 Sen I, Saraclar M, Kahya YP. Exploring an optimal vector autoregressive model for multi-channel pulmonary sound data. Comput Meth Prog Bio 2013; 111 (03) 550-560.
  • 7 Sen I, Kahya YP. A multi-channel device for respiratory sound data acquisition and transient detection. In: Proceedings of the 27th Annual International Conference of the IEEE-EMBS. 2005: 360-363.
  • 8 Lutkepohl H. New introduction to multiple time series analysis. Germany: Springer; 2005
  • 9 Alpaydin E. Introduction to machine learning. 2nd Edition. Cambridge, MA: The MIT Press; 2010
  • 10 Duda RO, Hart PE, Stork DG. Pattern Classification. 2nd Edition. Wiley: 2000
  • 11 Chang C-C, Lin C-J. LIBSVM: a library for support vector machines. ACM Trans Intell Syst Technol. 2011; 2 (03) 27 1-27: 27. Software available at. www.csie.ntu.edu.tw/cjlin/libsvm