[go: up one dir, main page]
More Web Proxy on the site http://driver.im/ Skip to main content
Springer Nature Link
Log in

A deep learning-based acute coronary syndrome-related disease classification method: a cohort study for network interpretability and transfer learning

  • Published:
Applied Intelligence Aims and scope Submit manuscript

Abstract

Accurate and efficient diagnosis of acute coronary syndrome (ACS)-related cardiac diseases is crucial for optimizing medical resources and enhancing clinical treatment efficiency. In this paper, we construct two distinct ECG datasets: a clinical 12-lead original ECG dataset and a clinical 12-lead standard ECG dataset. Our focus is on the automatic diagnosis of diseases clinically associated with acute coronary syndrome, specifically T wave changes, ST-T segment changes, and pathological Q waves. We propose a fine-tuned EfficientNet-based network for model construction and conduct comparative experiments using various state-of-the-art networks. Our experiments demonstrate satisfactory results in the classification of four types of ECGs: normal ECGs and three types of abnormal ECGs, with an overall classification accuracy of 73.333%. Notably, the model achieves a classification precision of 0.963 for pathological Q wave types and a classification specificity of 0.877, 0.859, 0.911, and 0.993 for the four types of models. To evaluate the model's performance, we randomly select a subset of ECGs and invite two groups of doctors, each with 4–5 years of experience in ECG interpretation, to determine the gold standard. A comparison of the proposed model's classification performance with the doctors' annotations reveals that for T wave changes and ST-T segment changes, the individual doctors' accuracy is 70.115%, while the proposed fine-tuning model achieves an accuracy of 71.111%. Thus, the performance of our model matches that of experienced doctors. Furthermore, we apply transfer learning to generalize the model trained on the original 12-lead ECG dataset to the new standard 12-lead ECG dataset. The experimental results demonstrate that the transferred model achieves a classification accuracy of 71.066% on the standard dataset, equivalent to the accuracy obtained on the original ECG dataset. Notably, for the normal sinus rhythm type, the precision and sensitivity of the transferred model improved by 7.4% and 1.8%, respectively. Additionally, the ROC AUC values for normal sinus rhythm, T wave changes, and ST-T segment changes after transfer are 0.857, 0.714, and 0.770, respectively. Moreover, we utilize Grad-CAM technology to provide network interpretability, offering theoretical support for clinical electrocardiogram diagnosis based on ECG records and classification results. Regarding network model efficiency, the average classification time for the transferred learning network on each ECG is approximately 0.0035 s, indicating the model's suitability for clinical practice and its potential to enhance clinical diagnosis efficiency.

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
£29.99 /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 (United Kingdom)

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10
Fig. 11

Similar content being viewed by others

Explore related subjects

Discover the latest articles, news and stories from top researchers in related subjects.

Data availability

ECG data employed in this paper will be shared by the lead contact (qinchengjin@sjtu.edu.cn (C. Qin)) upon request.

Code availability

All original code and code use instructions have been deposited at: https://jbox.sjtu.edu.cn/l/k1wUzd (Password: gtvl).

References

  1. Jahmunah V, Ng EYK, Tan R-S, Oh SL, Acharya UR (2023) Uncertainty quantification in DenseNet model using myocardial infarction ECG signals. Comput Methods Programs Biomed 229:107308. https://doi.org/10.1016/j.cmpb.2022.107308

    Article  Google Scholar 

  2. Rai HM, Chatterjee K (2022) Hybrid CNN-LSTM deep learning model and ensemble technique for automatic detection of myocardial infarction using big ECG data. Appl Intell 52:5366–5384. https://doi.org/10.1007/s10489-021-02696-6

    Article  Google Scholar 

  3. Chaki J, Woźniak M (2023) Deep learning for neurodegenerative disorder (2016 to 2022): a systematic review. Biomed Signal Process Control 80:104223. https://doi.org/10.1016/j.bspc.2022.104223

    Article  Google Scholar 

  4. Woźniak M, Wieczorek M, Siłka J (2023) BiLSTM deep neural network model for imbalanced medical data of IoT systems. Futur Gener Comput Syst 141:489–499. https://doi.org/10.1016/j.future.2022.12.004

    Article  Google Scholar 

  5. Subramanyan L, Ganesan U (2022) A novel deep neural network for detection of Atrial Fibrillation using ECG signals. Knowl-Based Syst 258:109926. https://doi.org/10.1016/j.knosys.2022.109926

    Article  Google Scholar 

  6. Petmezas G, Haris K, Stefanopoulos L, Kilintzis V, Tzavelis A, Rogers JA, Katsaggelos AK, Maglaveras N (2021) automated atrial fibrillation detection using a hybrid CNN-LSTM network on imbalanced ECG datasets. Biomed Signal Process Control 63:102194. https://doi.org/10.1016/j.bspc.2020.102194

    Article  Google Scholar 

  7. Jin Y, Qin C, Huang Y, Zhao W, Liu C (2020) Multi-domain modeling of atrial fibrillation detection with twin attentional convolutional long short-term memory neural networks. Knowl-Based Syst 193:105460. https://doi.org/10.1016/j.knosys.2019.105460

    Article  Google Scholar 

  8. Shi H, Qin C, Xiao D, Zhao L, Liu C (2020) Automated heartbeat classification based on deep neural network with multiple input layers. Knowl-Based Syst 188:105036. https://doi.org/10.1016/j.knosys.2019.105036

    Article  Google Scholar 

  9. Xie H, Liu H, Zhou S, Gao T, Shu M (2022) A lightweight 2-D CNN model with dual attention mechanism for heartbeat classification. Appl Intell. https://doi.org/10.1007/s10489-022-04303-8

    Article  Google Scholar 

  10. Li N, Liu L, Yang Z, Qin S (2023) A self-adjusting ant colony clustering algorithm for ECG arrhythmia classification based on a correction mechanism. Comput Methods Programs Biomed 235:107519. https://doi.org/10.1016/j.cmpb.2023.107519

    Article  Google Scholar 

  11. Boda S, Mahadevappa M, Dutta PK (2023) An automated patient-specific ECG beat classification using LSTM-based recurrent neural networks. Biomed Signal Process Control 84:104756. https://doi.org/10.1016/j.bspc.2023.104756

    Article  Google Scholar 

  12. Liu Y, Jin Y, Liu J, Qin C, Lin K, Shi H, Tao J, Zhao L, Liu C (2021) Precise and efficient heartbeat classification using a novel lightweight-modified method. Biomed Signal Process Control 68:102771. https://doi.org/10.1016/j.bspc.2021.102771

    Article  Google Scholar 

  13. Liu Y, Qin C, Liu J, Jin Y, Li Z, Liu C (2022) An efficient neural network-based method for patient-specific information involved arrhythmia detection. Knowl-Based Syst 250:109021. https://doi.org/10.1016/j.knosys.2022.109021

    Article  Google Scholar 

  14. Hannun AY, Rajpurkar P, Haghpanahi M, Tison GH, Bourn C, Turakhia MP, Ng AY (2019) Cardiologist-level arrhythmia detection and classification in ambulatory electrocardiograms using a deep neural network. Nat Med 25:65. https://doi.org/10.1038/s41591-018-0268-3

    Article  Google Scholar 

  15. Zhu H, Cheng C, Yin H, Li X, Zuo P, Ding J, Lin F, Wang J, Zhou B, Li Y, Hu S, Xiong Y, Wang B, Wan G, Yang X, Yuan Y (2020) Automatic multilabel electrocardiogram diagnosis of heart rhythm or conduction abnormalities with deep learning: a cohort study. Lancet Digit Health 2:e348–e357. https://doi.org/10.1016/S2589-7500(20)30107-2

    Article  Google Scholar 

  16. Ribeiro AH, Ribeiro MH, Paixão GMM, Oliveira DM, Gomes PR, Canazart JA, Ferreira MPS, Andersson CR, Macfarlane PW, Meira W Jr, Schön TB, Ribeiro ALP (2020) Automatic diagnosis of the 12-lead ECG using a deep neural network. Nat Commun 11:1760. https://doi.org/10.1038/s41467-020-15432-4

    Article  Google Scholar 

  17. Sangha V, Mortazavi BJ, Haimovich AD, Ribeiro AH, Brandt CA, Jacoby DL, Schulz WL, Krumholz HM, Ribeiro ALP, Khera R (2022) Automated multilabel diagnosis on electrocardiographic images and signals. Nat Commun 13:1583. https://doi.org/10.1038/s41467-022-29153-3

    Article  Google Scholar 

  18. Al-Zaiti S, Besomi L, Bouzid Z, Faramand Z, Frisch S, Martin-Gill C, Gregg R, Saba S, Callaway C, Sejdić E (2020) Machine learning-based prediction of acute coronary syndrome using only the pre-hospital 12-lead electrocardiogram. Nat Commun 11:3966. https://doi.org/10.1038/s41467-020-17804-2

    Article  Google Scholar 

  19. Sakr AS, Pławiak P, Tadeusiewicz R, Pławiak J, Sakr M, Hammad M (2023) ECG-COVID: an end-to-end deep model based on electrocardiogram for COVID-19 detection. Inf Sci (Ny) 619:324–339. https://doi.org/10.1016/j.ins.2022.11.069

    Article  Google Scholar 

  20. Cai W, Chen Y, Guo J, Han B, Shi Y, Ji L, Wang J, Zhang G, Luo J (2020) Accurate detection of atrial fibrillation from 12-lead ECG using deep neural network. Comput Biol Med 116:103378. https://doi.org/10.1016/j.compbiomed.2019.103378

    Article  Google Scholar 

  21. Yao Q, Wang R, Fan X, Liu J, Li Y (2020) Multi-class Arrhythmia detection from 12-lead varied-length ECG using attention-based time-incremental convolutional neural network. Inf Fusion 53:174–182. https://doi.org/10.1016/j.inffus.2019.06.024

    Article  Google Scholar 

  22. Zhang D, Yang S, Yuan X, Zhang P (2021) Interpretable deep learning for automatic diagnosis of 12-lead electrocardiogram. IScience 24:102373. https://doi.org/10.1016/j.isci.2021.102373

    Article  Google Scholar 

  23. Lai CX, Zhou SJ, Trayanova NA (2021) Optimal ECG-lead selection increases generalizability of deep learning on ECG abnormality classification. Philos Trans R Soc A 379:2212. https://doi.org/10.1098/rsta.2020.0258

    Article  Google Scholar 

  24. Zhang J, Liang D, Liu A, Gao M, Chen X, Zhang X, Chen X (2021) MLBF-Net: a multi-lead-branch fusion network for multi-class arrhythmia classification using 12-lead ECG. IEEE J Transl Eng Heal Med 9:1–11. https://doi.org/10.1109/JTEHM.2021.3064675

    Article  Google Scholar 

  25. Tan M, Le QV (2019) EfficientNet: rethinking model scaling for convolutional neural networks. International conference on machine learning, PMLR. http://refhub.elsevier.com/S1746-8094(23)00210-0/h0075. 2023-06-01

  26. Zheng Y, Feng X, Xia Z, Jiang X, Demontis A, Pintor M, Biggio B, Roli F (2023) Why adversarial reprogramming works, when it fails, and how to tell the difference. Inf Sci (Ny) 632:130–143. https://doi.org/10.1016/j.ins.2023.02.086

    Article  Google Scholar 

  27. Xia H, Lan Y, Song S, Li H (2021) A multi-scale segmentation-to-classification network for tiny microaneurysm detection in fundus images. Knowl-Based Syst 226:107140. https://doi.org/10.1016/j.knosys.2021.107140

    Article  Google Scholar 

  28. Qin C, Jin Y, Zhang Z, Yu H, Tao J, Sun H, Liu C (2023) Anti-noise diesel engine misfire diagnosis using a multi-scale CNN-LSTM neural network with denoising module. CAAI Trans Intell Technol. https://doi.org/10.1049/cit2.12170

    Article  Google Scholar 

  29. Qin C, Wu R, Huang G, Tao J, Liu C (2023) A novel LSTM-autoencoder and enhanced transformer-based detection method for shield machine cutterhead clogging. Sci China Technol Sci 66:512–527. https://doi.org/10.1007/s11431-022-2218-9

    Article  Google Scholar 

  30. Qin C, Huang G, Yu H, Wu R, Tao J, Liu C (2023) Geological information prediction for shield machine using an enhanced multi-head self-attention convolution neural network with two-stage feature extraction. Geosci Front 14:101519. https://doi.org/10.1016/j.gsf.2022.101519

    Article  Google Scholar 

  31. Yu H, Qin C, Tao J, Liu C, Liu Q (2023) A multi-channel decoupled deep neural network for tunnel boring machine torque and thrust prediction. Tunn Undergr Space Technol 133:104949. https://doi.org/10.1016/j.tust.2022.104949

    Article  Google Scholar 

  32. Yu H, Sun H, Tao J, Qin C et al (2023) A multi-stage data augmentation and ABi-ResNet-based method for EPB utilization factor prediction. Autom Constr 147:104734

    Article  Google Scholar 

  33. Selvaraju RR, Cogswell M, Das A, Vedantam R, Parikh D, Batra D (2017) Grad-CAM: visual explanations from deep networks via gradient-based localization. In: 2017 IEEE Int. Conf. Comput. Vis., pp 618–626.https://doi.org/10.1109/ICCV.2017.74

  34. Hoe Kim S, Shin Park J, Soo Lee H, HyukYoo S, Joo Oh K (2023) Combining CNN and Grad-CAM for profitability and explainability of investment strategy: application to the KOSPI 200 futures. Expert Syst Appl 225:120086. https://doi.org/10.1016/j.eswa.2023.120086

    Article  Google Scholar 

  35. Lin Q-H, Niu Y-W, Sui J, Zhao W-D, Zhuo C, Calhoun VD (2022) SSPNet: an interpretable 3D-CNN for classification of schizophrenia using phase maps of resting-state complex-valued fMRI data. Med Image Anal 79:102430. https://doi.org/10.1016/j.media.2022.102430

    Article  Google Scholar 

Download references

Acknowledgements

This work was supported by the Shanghai Municipal Science and Technology Major Project (Grant No. 2021SHZDZX0102).

Author information

Authors and Affiliations

  1. School of Mechanical Engineering, Shanghai Jiao Tong University, 800, Dongchuan Road, Shanghai, 200240, China

    Yunqing Liu, Jinlei Liu, Chengjin Qin, Yanrui Jin, Zhiyuan Li & Chengliang Liu

  2. MoE Key Lab of Artificial Intelligence, AI Institute, Shanghai Jiao Tong University, Shanghai, China

    Yunqing Liu, Jinlei Liu, Chengjin Qin, Yanrui Jin, Zhiyuan Li & Chengliang Liu

  3. Department of Cardiology, Shanghai First People’s Hospital Affiliated with Shanghai Jiao Tong University, 100, Haining Road, Shanghai, 200080, China

    Liqun Zhao

Authors
  1. Yunqing Liu
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Jinlei Liu
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Chengjin Qin
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Yanrui Jin
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Zhiyuan Li
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Liqun Zhao
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Chengliang Liu
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding authors

Correspondence to Chengjin Qin or Chengliang Liu.

Ethics declarations

Conflict of interests

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Additional information

Publisher's note

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

Appendix

Appendix

Table 9 Experimental details of doctor annotation
Full size table

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) 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

Liu, Y., Liu, J., Qin, C. et al. A deep learning-based acute coronary syndrome-related disease classification method: a cohort study for network interpretability and transfer learning. Appl Intell 53, 25562–25580 (2023). https://doi.org/10.1007/s10489-023-04889-7

Download citation

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10489-023-04889-7

Keywords

Search

Navigation