A Multistage Deep Belief Networks Application on Arrhythmia Classification

Abstract

An electrocardiogram (ECG) is a biomedical signal type that determines the normality and abnormality of heart beats using the electrical activity of the heart and has a great importance for cardiac disorders. The computer-aided analysis of biomedical signals has become a fabulous utilization method over the last years. This study introduces a multistage deep learning classification model for automatic arrhythmia classification. The proposed model includes a multi-stage classification system that uses ECG waveforms and the Second Order Difference Plot (SODP) features using a Deep Belief Network (DBN) classifier which has a greedy layer wise training with Restricted Boltzmann Machines algorithm. The multistage DBN model classified the MIT-BIH Arrhythmia Database heartbeats into 5 main groups defined by ANSI/AAMI standards. All ECG signals are filtered with median filters to remove the baseline wander. ECG waveforms were segmented from long-term ECG signals using a window with a length of 501 data points (R wave centered). The extracted waveforms and elliptical features from the SODP are utilized as the input of the model.  The proposed DBN-based multistage arrhythmia classification model has discriminated five types of heartbeats with a high accuracy rate of 96.10%.

Keywords

• Arrhythmia,Deep Belief Networks,DBN,Deep Learning,AAMI,ECG Waveform,Second Order Difference Plot,SODP

References

1. WHO. The top 10 causes of death y.y. http://www.who.int/mediacentre/factsheets/fs310/en/ (access: 07 August 2016).
2. Webster JG. Medical Instrumentation, Application and Design. 4th baskı. Boston: Houghtoon Mifflin Company; 1978.
3. Yeh YC, Chiou CW, Lin HJ. Analyzing ECG for cardiac arrhythmia using cluster analysis. Expert Syst Appl 2012;39:1000–10. doi:10.1016/j.eswa.2011.07.101.
4. Hamilton P. Open source ECG analysis. Comput Cardiol 2002;29:101–4. doi:10.1109/CIC.2002.1166717.
5. Ge D, Srinivasan N, Krishnan SM. Cardiac arrhythmia classification using autoregressive modeling. Biomed Eng Online 2002;1:5. doi:10.1186/1475-925X-1-5.
6. Israel SA, Irvine JM, Cheng A, Wiederhold MD, Wiederhold BK. ECG to identify individuals. Pattern Recognit 2005;38:133–42. doi:10.1016/j.patcog.2004.05.014.
7. Plataniotis KN, Hatzinakos D, Lee JKM. ECG Biometric Recognition Without Fiducial Detection. 2006 Biometrics Symp. Spec. Sess. Res. Biometric Consort. Conf., IEEE; 2006, s. 1–6. doi:10.1109/BCC.2006.4341628.
8. Bengio Y, Delalleau O. Justifying and generalizing contrastive divergence. Neural Comput 2009;21:1601–21. doi:10.1162/neco.2008.11-07-647.

9. Moody GB, Mark RG. The impact of the MIT-BIH arrhythmia database. IEEE Eng Med Biol Mag y.y.;20:45–50.
10. Kutlu Y, Kuntalp D. A multi-stage automatic arrhythmia recognition and classification system. Comput Biol Med 2011;41:37–45. doi:10.1016/j.compbiomed.2010.11.003.
11. Rajendra Acharya U, Suri JS, Spaan JAE, Krishnan SM. Advances in cardiac signal processing. 2007. doi:10.1007/978-3-540-36675-1.
12. M. Gabriel Khan. Rapid ECG Interpretation(Contemporary Cardiology). 3rd editio. Humana Press; 2007.
13. Dr Patrick Davey. ECG (electrocardiogram). NetDoctor 2011:1–4.
14. Kara S. Sensing of ECG signals and Imaging at the Computer in Real Time. Erciyes University, 1991.
15. Yayik A, Kutlu Y. Konjestif kalp yetmezliǧinin ikinci-derece fark harita grafiǧi ile topografik analizi ve teşhisi. 2014 22nd Signal Process. Commun. Appl. Conf. SIU 2014 - Proc., 2014, s. 540–3. doi:10.1109/SIU.2014.6830285.
16. Altan G, Kutlu Y. ECG based Human identification using Logspace Grid Analysis of Second Order Difference Plot. Signal Process Commun Appl Conf 2015:1288–91.
17. Bengio Y, Lamblin P, Popovici D, Larochelle H. Greedy Layer-Wise Training of Deep Networks. Adv Neural Inf Process Syst 2007;19:153. doi:citeulike-article-id:4640046.
18. Yan Y, Qin X, Wu Y, Zhang N, Fan J, Wang L. A restricted Boltzmann machine based two-lead electrocardiography classification. 2015 IEEE 12th Int Conf Wearable Implant Body Sens Networks 2015:1–9. doi:10.1109/BSN.2015.7299399.
19. Hinton GE, Osindero S, Teh Y-W. A fast learning algorithm for deep belief nets. Neural Comput 2006;18:1527–54. doi:10.1162/neco.2006.18.7.1527.
20. Lee HS, Cheng Q lan, Thakor N V. ECG waveform analysis by significant point extraction. I. Data reduction. Comput Biomed Res 1987;20:410–27. doi:10.1016/0010-4809(87)90030-9.
21. Ververidis D, Kotropoulos C. Sequential forward feature selection with low computational cost. Signal Process. Conf. 2005 13th Eur., vol. 13, 2005, s. 1–4.
22. Zhang Z, Dong J, Luo X, Choi KS, Wu X. Heartbeat classification using disease-specific feature selection. Comput Biol Med 2014;46:79–89. doi:10.1016/j.compbiomed.2013.11.019.
23. Alajlan N, Bazi Y, Melgani F, Malek S, Bencherif MA. Detection of premature ventricular contraction arrhythmias in electrocardiogram signals with kernel methods. Signal, Image Video Process 2014;8:931–42. doi:10.1007/s11760-012-0339-8.
24. Batra A, Jawa V. Classification of Arrhythmia Using Conjunction of Machine Learning Algorithms and ECG Diagnostic Criteria. Int J Biol Biomed 2016;1:1–7.
25. Melin P, Amezcua J, Valdez F, Castillo O. A new neural network model based on the LVQ algorithm for multi-class classification of arrhythmias. Inf Sci (Ny) 2014;279:483–97. doi:10.1016/j.ins.2014.04.003.
26. Thomas M, Das MK, Ari S. Automatic ECG arrhythmia classification using dual tree complex wavelet based features. AEU - Int J Electron Commun 2015;69:715–21. doi:10.1016/j.aeue.2014.12.013.
27. Leutheuser H, Gradl S, Kugler P, Anneken L, Arnold M, Achenbach S, vd. Comparison of real-time classification systems for arrhythmia detection on Android-based mobile devices. IEEE Eng Med Biol Soc Annu Conf 2014;2014:2690–3. doi:10.1109/EMBC.2014.6944177.
28. Yayik A, Altan G, Kutlu Y, Yildirim E, Yildirim S. Görgül Mod Fonksiyonların Eliptik Analizi ile Kongestif Kalp Yetmezliği Teşhisi. Int Conf Electr Electron Eng 2014:632–5.
29. Rahhal MM Al, Bazi Y, AlHichri H, Alajlan N, Melgani F, Yager RR. Deep Learning Approach for Active Classification of Electrocardiogram Signals. Inf Sci (Ny) 2016;345:340–54. doi:10.1016/j.ins.2016.01.082.
30. Huanhuan M, Yue Z. Classification of Electrocardiogram Signals with Deep Belief Networks. Comput Sci Eng (CSE), 2014 IEEE 17th Int Conf 2014:7–12. doi:10.1109/CSE.2014.36.
31. Allahverdi N, Altan G, Kutlu Y. Diagnosis of Coronary Artery Disease Using Deep Belief Networks. 2 Int Conf Eng Nat Sci 2016, Sarajevo, Bosnia, pp:40-46.
32. Owis MI, Abou-Zied AH, Youssef a. BM, Kadah YM. Study of features based on nonlinear dynamical modeling in ECG arrhythmia detection and classification. IEEE Trans Biomed Eng 2002;49:733–6. doi:10.1109/TBME.2002.1010858.
33. Martis RJ, Acharya UR, Lim CM, Suri JS. Characterization of ECG beats from cardiac arrhythmia using discrete cosine transform in PCA framework. Knowledge-Based Syst 2013;45:76–82. doi:10.1016/j.knosys.2013.02.007.
34. Kim J, Min SD, Lee M. An arrhythmia classification algorithm using a dedicated wavelet adapted to different subjects. Biomed Eng Online 2011;10:56. doi:10.1186/1475-925X-10-56.
35. Tadejko P, Rakowski W. Hybrid wavelet-mathematical morphology feature extraction for heartbeat classification. EUROCON 2007 - Int. Conf. Comput. as a Tool, 2007, s. 127–32. doi:10.1109/EURCON.2007.4400676.
36. Llamedo M, Martinez JP. Heartbeat classification using feature selection driven by database generalization criteria. IEEE Trans Biomed Eng 2011;58:616–25. doi:10.1109/TBME.2010.2068048.
37. Alvarado AS, Lakshminarayan C, Príncipe JC. Time-based compression and classification of heartbeats. IEEE Trans Biomed Eng 2012;59:1641–8. doi:10.1109/TBME.2012.2191407.
38. Ye C, Vijaya Kumar BVK, Coimbra MT. Heartbeat classification using morphological and dynamic features of ECG signals. IEEE Trans Biomed Eng 2012;59:2930–41. doi:10.1109/TBME.2012.2213253.
39. Kutlu Y, Altan G, Allahverdi N. ARRHYTHMIA CLASSIFICATION USING WAVEFORM ECG SIGNALS. 3rd Int. Conf. Adv. Technol. Sci., Konya: 2016, s. 233–239.