Skip to main content
Springer Nature Link
Log in

A compact hybrid fully convolutional and BiLSTM network with squeeze and temporal excitation approach for semantic segmentation of ECG signal

  • Track 2: Medical Applications of Multimedia
  • Published:
Multimedia Tools and Applications Aims and scope Submit manuscript

Abstract

Existing approaches for ECG wave segmentation, only focused on the P, QRS, and T wave localization using machine and deep learning approaches. Although, deep learning-based approaches have achieved satisfactory results, but these approaches generate a large number of trainable parameters which leads to the computational burden and overfitting problem. Moreover, the existing approaches do not consider PR and ST segments which affects the precise localization of fiducial points. Hence, to alleviate these problems a novel approach to compress deep architecture has been proposed to localize key ECG waves along with PR and ST segments. For compressing the deep models, we first time implemented a filter squeeze and temporal excitation approach. The proposed method involves- i) Semantic segmentation of ECG waves and segments using hybrid convolutional and BiLSTM networks to extract temporal dependencies, and ii) Integration of filter squeeze and temporal excitation (FS-TE) which adaptively recalibrates temporal feature responses by exhibiting temporal interdependencies. The experiments on the standard QT database of Physionet with the proposed network achieved ~3% increment in the performance and 2 times reduction in the number of parameters as compared to the existing method. The proposed model learns channel-wise relevant temporal features with less computational cost as compared to existing stacked BiLSTM-based approaches for semantic segmentation of ECG waves.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Subscribe and save

Springer+ Basic
€32.70 /Month
  • Get 10 units per month
  • Download Article/Chapter or eBook
  • 1 Unit = 1 Article or 1 Chapter
  • Cancel anytime
Subscribe now

Buy Now

Price includes VAT (France)

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7

Similar content being viewed by others

References

  1. Abrishami H, Campbell M, Han C, et al (2018) P-QRS-T localization in ECG using deep learning. In: 2018 IEEE EMBS international conference on biomedical and health informatics, BHI 2018. Institute of Electrical and Electronics Engineers Inc., pp 210–213

  2. Abrishami H, Campbell M, Czosek R (2018) Supervised ECG interval segmentation using LSTM neural network. Int Conf Bioinforma Comput Biol BIOCOMP’18

  3. Ali A, Zhu Y, Zakarya M (2021) A data aggregation based approach to exploit dynamic spatio-temporal correlations for citywide crowd flows prediction in fog computing. Multimed Tools Appl 80:31401–31433

    Article  Google Scholar 

  4. Ali A, Zhu Y, Zakarya M (2021) Exploiting dynamic spatio-temporal correlations for citywide traffic flow prediction using attention based neural networks. Inf Sci (NY) 577:852–870

    Article  MathSciNet  Google Scholar 

  5. Ali A, Zhu Y, Zakarya M (2022) Exploiting dynamic spatio-temporal graph convolutional neural networks for citywide traffic flows prediction. Neural Netw 145:233–247

    Article  Google Scholar 

  6. Altuve M, Casanova O, Wong S, et al (2007) Evaluación de dos Métodos para la Segmentación del Ancho de la Onda T en el ECG. In: IV Latin American Congress on Biomedical Engineering 2007, Bioengineering solutions for Latin America Health. Springer, pp 1254–1258

  7. Andreao RV, Dorizzi B, Boudy J (2006) ECG signal analysis through hidden Markov models. IEEE Trans Biomed Eng 53:1541–1549

    Article  Google Scholar 

  8. Bengio Y, Simard P, Frasconi P (1994) Learning long-term dependencies with gradient descent is difficult. IEEE Trans Neural Netw 5:157–166. https://doi.org/10.1109/72.279181

    Article  Google Scholar 

  9. Beraza I, Romero I (2017) Comparative study of algorithms for ECG segmentation. Biomed Signal Process Control 34:166–173. https://doi.org/10.1016/J.BSPC.2017.01.013

    Article  Google Scholar 

  10. Campbell MJ, Zhou X, Han C, Abrishami H, Webster G, Miyake CY, Sower CT, Anderson JB, Knilans TK, Czosek RJ (2017) Pilot study analyzing automated ECG screening of hypertrophic cardiomyopathy. Hear Rhythm 14:848–852. https://doi.org/10.1016/j.hrthm.2017.02.011

    Article  Google Scholar 

  11. Chollet F (2015) Keras: the python deep learning library. KerasIo ascl-1806. https://doi.org/10.1086/316861

  12. de Chazal P, Celler BG (1996) Automatic measurement of the QRS onset and offset in individual ECG leads. Annu Int Conf IEEE Eng Med Biol - Proc 4:1399–1400. https://doi.org/10.1109/IEMBS.1996.647474

    Article  Google Scholar 

  13. Dumont J, Hernández AI, Carrault G (2005) Parameter optimization of a wavelet-based electrocardiogram delineator with an evolutionary algorithm. Comput Cardiol 32:707–710. https://doi.org/10.1109/CIC.2005.1588202

  14. Frénay B, de Lannoy G, Verleysen M (2009) Emission modelling for supervised ECG segmentation using finite differences. In: 4th European conference of the international federation for medical and biological engineering. Springer, pp 1212–1216

  15. Ghaffari A, Homaeinezhad MR, Khazraee M, Daevaeiha MM (2010) Segmentation of Holter ECG waves via analysis of a discrete wavelet-derived multiple skewness-kurtosis based metric. Ann Biomed Eng 38:1497–1510. https://doi.org/10.1007/s10439-010-9919-3

    Article  Google Scholar 

  16. Graja S, Boucher J-M (2003) Multiscale hidden Markov model applied to ECG segmentation. In: IEEE international symposium on intelligent signal processing, 2003. IEEE, pp 105–109

  17. Gupta R, Mitra M, Mondal K, Bhowmick S (2011) A derivative-based approach for QT-segment feature extraction in digitized ECG record. Proc - 2nd Int Conf Emerg Appl Inf Technol EAIT 2011 63–66. https://doi.org/10.1109/EAIT.2011.61

  18. Homaeinezhad MR, Ghaffari A, Toosi HN, et al (2011) A unified framework for delineation of ambulatory Holter ECG events via analysis of a multiple-order derivative wavelet-based measure

  19. Hu J, Shen L, Sun G (2018) Squeeze-and-excitation networks. In: Proceedings of the IEEE computer society conference on computer vision and pattern recognition. IEEE Computer Society, pp 7132–7141

  20. Hughes NP, Tarassenko L, Roberts SJ Markov Models for automated ECG interval analysis

  21. Illanes-Manriquez A (2010) An automatic multi-lead electrocardiogram segmentation algorithm based on abrupt change detection. In: 2010 annual international conference of the IEEE engineering in medicine and biology society, EMBC’10, pp 2334–2337

  22. Illanes-Manriquez A, Zhang Q (2008) An algorithm for robust detection of QRS onset and offset in ECG signals. In: Computers in cardiology, pp 857–860

  23. Jane R, Blasi A, García J, Laguna P (1997) Evaluation of an automatic threshold based detector of waveform limits in Holter ECG with the QT database. In: computers in cardiology 1997. IEEE, pp 295–298

  24. Kaiser W, Faber TS, Findeis M (1996) Automatic learning of rules: a practical example of using artificial intelligence to improve computer-based detection of myocardial infarction and left ventricular hypertrophy in the 12-lead ECG. J Electrocardiol 29:17–20

    Article  Google Scholar 

  25. Karimipour A, Homaeinezhad MR (2014) Real-time electrocardiogram P-QRS-T detection-delineation algorithm based on quality-supported analysis of characteristic templates. Comput Biol Med 52:153–165. https://doi.org/10.1016/j.compbiomed.2014.07.002

    Article  Google Scholar 

  26. Kingma DP, Ba J (2014) Adam: a method for stochastic optimization

  27. Laguna P, Thakor NV, Caminal P, Jané R, Yoon HR, Bayés de Luna A, Marti V, Guindo J (1990) New algorithm for QT interval analysis in 24-hour Holter ECG: performance and applications. Med Biol Eng Comput 28:67–73

    Article  Google Scholar 

  28. Laguna P, Jané R, Caminal P (1994) Automatic detection of wave boundaries in multilead ECG signals: validation with the CSE database. Comput Biomed Res 27:45–60. https://doi.org/10.1006/cbmr.1994.1006

    Article  Google Scholar 

  29. Laguna P, Mark RG, Goldberg A, Moody GB (1997) A database for evaluation of algorithms for measurement of QT and other waveform intervals in the ECG. In: Computers in cardiology 1997. IEEE, pp 673–676

  30. Last T, Nugent CD, Owens FJ (2004) Multi-component based cross correlation beat detection in electrocardiogram analysis. Biomed Eng Online 3:26. https://doi.org/10.1186/1475-925X-3-26

    Article  Google Scholar 

  31. Li C, Zheng C, Tai C (1995) Detection of ECG characteristic points using wavelet transforms. IEEE Trans Biomed Eng 42:21–28. https://doi.org/10.1109/10.362922

    Article  Google Scholar 

  32. Liang X, Li L, Liu Y, Chen D, Wang X, Hu S, Wang J, Zhang H, Sun C, Liu C (2022) ECG_SegNet: an ECG delineation model based on the encoder-decoder structure. Comput Biol Med 145:105445. https://doi.org/10.1016/j.compbiomed.2022.105445

    Article  Google Scholar 

  33. Londhe AN, Atulkar M (2020) Segmentation of ECG waves using LSTM networks. In: modelling and analysis of active biopotential signals in healthcare, vol 2. IOP publishing, pp 12–24

  34. Londhe AN, Atulkar M (2021) Semantic segmentation of ECG waves using hybrid channel-mix convolutional and bidirectional LSTM. Biomed Signal Process Control 63:102162. https://doi.org/10.1016/j.bspc.2020.102162

    Article  Google Scholar 

  35. Madeiro JPV, Cortez PC, Marques JAL, Seisdedos CRV, Sobrinho CRMR (2012) An innovative approach of QRS segmentation based on first-derivative, Hilbert and wavelet transforms. Med Eng Phys 34:1236–1246. https://doi.org/10.1016/j.medengphy.2011.12.011

    Article  Google Scholar 

  36. Martínez JP, Olmos S, Laguna P (2000) Evaluation of a wavelet-based ECG waveform detector on the QT database. Comput Cardiol 81–84. https://doi.org/10.1109/CIC.2000.898460

  37. Martínez A, Alcaraz R, Rieta JJ (2010) Application of the phasor transform for automatic delineation of single-lead ECG fiducial points. Physiol Meas 31:1467–1485. https://doi.org/10.1088/0967-3334/31/11/005

  38. Mukhopadhyay SK, Mitra M, Mitra S (2011) Time plane ECG feature extraction using Hilbert transform, variable threshold and slope reversal approach. In: proceedings of the 2011 international conference on communication and industrial application, ICCIA 2011

  39. Nair V, Hinton GE (2010) Rectified linear units improve restricted boltzmann machines. In: Proceedings of the 27th international conference on machine learning (ICML-10), pp 807–814

  40. Nurmaini S, Darmawahyuni A, Rachmatullah MN, Effendi J, Sapitri AI, Firdaus F, Tutuko B (2021) Beat-to-beat electrocardiogram waveform classification based on a stacked convolutional and bidirectional long short-term memory. IEEE Access 9:92600–92613. https://doi.org/10.1109/ACCESS.2021.3092631

    Article  Google Scholar 

  41. Roy AG, Navab N, Wachinger C (2018) Concurrent spatial and channel ‘squeeze & excitation’in fully convolutional networks. In: International conference on medical image computing and computer-assisted intervention. Springer, pp 421–429

  42. Schreier G, Hayn D, Lobodzinski S (2003) Development of a new QT algorithm with Heterogenous ECG databases. In: Journal of Electrocardiology. Churchill Livingstone Inc., pp 145–150

  43. Sodmann P, Vollmer M, Nath N, Kaderali L (2018) A convolutional neural network for ECG annotation as the basis for classification of cardiac rhythms. Physiol Meas 39:104005. https://doi.org/10.1088/1361-6579/aae304

    Article  Google Scholar 

  44. Stamkopoulos T, Maglaveras N, Bamidis PD, Pappas C (2000) Wave segmentation using nonstationary properties of ECG. In: Computers in cardiology. IEEE, pp 529–532

  45. Sun Y, Chan KL, Krishnan SM (2005) Characteristic wave detection in ECG signal using morphological transform. BMC Cardiovasc Disord 5. https://doi.org/10.1186/1471-2261-5-28

  46. Tutuko B, Rachmatullah MN, Darmawahyuni A et al (2022) Short single-lead ECG signal delineation-based deep learning: implementation in automatic atrial fibrillation identification. Sensors. https://doi.org/10.3390/s22062329

  47. Van Der Walt S, Colbert SC, Varoquaux G (2011) The NumPy array: a structure for efficient numerical computation. Comput Sci Eng 13:22–30. https://doi.org/10.1109/MCSE.2011.37

    Article  Google Scholar 

  48. Vila JA, Gang Y, Presedo JMR et al (2000) A new approach for TU complex characterization. IEEE Trans Biomed Eng 47:764–772

    Article  Google Scholar 

  49. Vullings H, Verhaegen MHG, Verbruggen HB (1998) Automated ECG segmentation with dynamic time warping. In: Proceedings of the 20th annual international conference of the IEEE engineering in medicine and biology society. Vol. 20 biomedical engineering towards the year 2000 and beyond (Cat. No. 98CH36286). IEEE, pp 163–166

  50. Warner RA, Ariel Y, Gasperina MD, Okin PM (2002) Improved electrocardiographic detection of left ventricular hypertrophy. J Electrocardiol 35:111–115. https://doi.org/10.1054/jelc.2002.37163

    Article  Google Scholar 

  51. Xia H, Asif I, Zhao X (2013) Cloud-ECG for real time ECG monitoring and analysis. Comput Methods Prog Biomed 110:253–259. https://doi.org/10.1016/j.cmpb.2012.11.008

    Article  Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Department of Computer Applications, National Institute of Technology Raipur, G. E. Road, Raipur, Chhattisgarh, 492010, India

    Aboli N. Londhe & Mithilesh Atulkar

Authors
  1. Aboli N. Londhe
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Mithilesh Atulkar
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Aboli Londhe. The first draft of the manuscript was written by Aboli Londhe and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Corresponding author

Correspondence to Aboli N. Londhe.

Ethics declarations

Competing interests

The authors have no relevant financial or non-financial interests to disclose.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Springer Nature or its licensor holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Londhe, A.N., Atulkar, M. A compact hybrid fully convolutional and BiLSTM network with squeeze and temporal excitation approach for semantic segmentation of ECG signal. Multimed Tools Appl 82, 12679–12697 (2023). https://doi.org/10.1007/s11042-022-13821-z

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11042-022-13821-z

Keywords

Search

Navigation