Skip to main content

Advertisement

SpringerLink
Log in

Content-based image retrieval for the diagnosis of myocardial perfusion imaging using a deep convolutional autoencoder

  • Original Article
  • Published:
Journal of Nuclear Cardiology Aims and scope

Abstract

Background

Single-photon emission computed tomography (SPECT) myocardial perfusion imaging (MPI) plays a crucial role in the optimal treatment strategy for patients with coronary heart disease. We tested the feasibility of feature extraction from MPI using a deep convolutional autoencoder (CAE) model.

Methods

Eight hundred and forty-three pairs of stress and rest myocardial perfusion images were collected from consecutive patients who underwent cardiac scintigraphy in our hospital between December 2019 and February 2022. We trained a CAE model to reproduce the input paired image data, so as the encoder to output a 256-dimensional feature vector. The extracted feature vectors were further dimensionally reduced via principal component analysis (PCA) for data visualization. Content-based image retrieval (CBIR) was performed based on the cosine similarity of the feature vectors between the query and reference images. The agreement of the radiologist’s finding between the query and retrieved MPI was evaluated using binary accuracy, precision, recall, and F1-score.

Results

A three-dimensional scatter plot with PCA revealed that feature vectors retained clinical information such as percent summed difference score, presence of ischemia, and the location of scar reported by radiologists. When CBIR was used as a similarity-based diagnostic tool, the binary accuracy was 81.0%.

Conclusion

The results indicated the utility of unsupervised feature learning for CBIR in MPI.

Graphical abstract

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

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Figure 1
Figure 2
Figure 3
Figure 4

Similar content being viewed by others

Data availability

The data that support the findings of this study are available in the Mendeley Data Repository (https://doi.org/10.17632/mjhhw3zdwv.1).

Abbreviations

SPECT:

Single-photon emission computed tomography

MPI:

Myocardial perfusion imaging

CHD:

Coronary heart disease

ML:

Machine learning

CBIR:

Content-based image retrieval

CAE:

Convolutional autoencoder

SRS:

Summed rest score

SSS:

Summed stress score

SDS:

Summed difference score

PCA:

Principal component analysis

References

  1. Hachamovitch R, Rozanski A, Shaw LJ, Stone GW, Thomson LE, Friedman JD. Impact of ischaemia and scar on the therapeutic benefit derived from myocardial revascularization vs. medical therapy among patients undergoing stress-rest myocardial perfusion scintigraphy. Eur Heart J 2011;32:1012‐24.

    Article  PubMed  Google Scholar 

  2. Seifert R, Weber M, Kocakavuk E, Rischpler C, Kersting D. Artificial intelligence and machine learning in nuclear medicine: Future perspectives. Semin Nucl Med 2021;51:170‐7.

    Article  PubMed  Google Scholar 

  3. Guner LA, et al. An open-source framework of neural networks for diagnosis of coronary artery disease from myocardial perfusion SPECT. J Nucl Cardiol 2010;17:405‐13.

    Article  PubMed  Google Scholar 

  4. Betancur J, et al. Deep learning for prediction of obstructive disease from fast myocardial perfusion SPECT: A multicenter study. JACC Cardiovasc Imaging 2018;11:1654‐63.

    Article  PubMed  PubMed Central  Google Scholar 

  5. Kaplan Berkaya S, Ak Sivrikoz I, Gunal S. Classification models for SPECT myocardial perfusion imaging. Comput Biol Med 2020;123:103893.

    Article  PubMed  Google Scholar 

  6. Sengupta PP, et al. Proposed requirements for cardiovascular imaging-related machine learning evaluation (PRIME): A checklist. JACC Cardiovasc Imaging 2020;13:2017‐35.

    Article  PubMed  PubMed Central  Google Scholar 

  7. Pasini A. Artificial neural networks for small dataset analysis. J Thorac Dis 2015;7:953‐60.

    PubMed  PubMed Central  Google Scholar 

  8. Müller H, Michoux N, Bandon D, Geissbuhler A. A review of content-based image retrieval systems in medical applications: Clinical benefits and future directions. Int J Med Inform 2004;73:1‐23.

    Article  PubMed  Google Scholar 

  9. Dhara AK, Mukhopadhyay S, Dutta A, Garg M, Khandelwal N. Content-based image retrieval system for pulmonary nodules: assisting radiologists in self-learning and diagnosis of lung cancer. J Digit Imaging 2017;30:63‐77.

    Article  PubMed  Google Scholar 

  10. Daoud MI, Saleh A, Hababeh I, Alazrai R, Content-based image retrieval for breast ultrasound images using convolutional autoencoders: A feasibility study. BioSMART 2019 - Proc. 3rd Int. Conf. Bio-Engineering Smart Technol. (2019). https://doi.org/10.1109/BIOSMART.2019.8734190.

  11. Klocke FJ, et al. ACC/AHA/ASNC guidelines for the clinical use of cardiac radionuclide imaging–executive summary: A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (ACC/AHA/ASNC Committee to Revise the 1995 Guideli. Circulation 2003;108:1404‐18.

    Article  PubMed  Google Scholar 

  12. Cerqueira MD, et al. Standardized myocardial segmentation and nomenclature for tomographic imaging of the heart. J Cardiovasc Magn Reson 2002;4:203‐10.

    Article  Google Scholar 

  13. Nakajima K, et al. Creation and characterization of Japanese standards for myocardial perfusion SPECT: Database from the Japanese Society of Nuclear Medicine Working Group. Ann Nucl Med 2007;21:505‐11.

    Article  PubMed  Google Scholar 

  14. Moroi M, Yamashina A, Tsukamoto K, Nishimura T, J-ACCESS Investigators. Coronary revascularization does not decrease cardiac events in patients with stable ischemic heart disease but might do in those who showed moderate to severe ischemia. Int J Cardiol 2012;158:246‐52.

    Article  PubMed  Google Scholar 

  15. Maron DJ, et al. International study of comparative health effectiveness with medical and invasive approaches (ISCHEMIA) trial: Rationale and design. Am Heart J 2018;201:124‐35.

    Article  PubMed  Google Scholar 

  16. Lever J, Krzywinski M, Altman N. Points of significance: principal component analysis. Nat Methods 2017;14:641‐2.

    Article  CAS  Google Scholar 

  17. Higaki A, Mahmoud AUM, Paradis P, Schiffrin EL. Automated detection and diameter estimation for mouse mesenteric artery using semantic segmentation. J Vasc Res 2021;58:379‐87.

    Article  CAS  PubMed  Google Scholar 

  18. Higaki A, Uetani T, Ikeda S, Yamaguchi O. Co-authorship network analysis in cardiovascular research utilizing machine learning (2009–2019). Int J Med Inform 2020;143:104274.

    Article  PubMed  Google Scholar 

  19. Vollmer S, Mateen BA, Bohner G, Király FJ, Ghani R, Jonsson P, et al. Machine learning and artificial intelligence research for patient benefit: 20 critical questions on transparency, replicability, ethics, and effectiveness. BMJ 2020;368:15.

    Google Scholar 

  20. Latif A, Rasheed A, Sajid U, Ahmed J, Ali N, Ratyal NI et al. Content-based image retrieval and feature extraction: a comprehensive review. Math Probl Eng 2019, (2019).

  21. Higaki A, Kurokawa T, Kazatani T, Kido S, Aono T, Matsuda K et al. Image similarity-based cardiac rhythm device identification from X-rays using feature point matching. Pacing Clin Electrophysiol 2021;44:633‐40.

    Article  PubMed  Google Scholar 

  22. Siradjuddin IA, Wardana WA, Sophan MK, Feature extraction using self-supervised convolutional autoencoder for content based image retrieval. ICICOS 2019 - 3rd Int. Conf. Informatics Comput. Sci. Accel. Informatics Comput. Res. Smarter Soc. Era Ind. 4.0, Proc. (2019). https://doi.org/10.1109/ICICOS48119.2019.8982468.

  23. Choe J, et al. Content-based Image retrieval by using deep learning for interstitial lung disease diagnosis with chest CT. Radiology 2021. https://doi.org/10.1148/radiol.2021204164.

    Article  PubMed  Google Scholar 

  24. Le QV Building high-level features using large scale unsupervised learning. in 2013 IEEE International Conference on Acoustics, Speech and Signal Processing 8595–8598 (IEEE, 2013). https://doi.org/10.1109/ICASSP.2013.6639343.

Download references

Acknowledgments

We would like to thank Editage (www.editage.com) for English language editing.

Funding

None.

Author information

Authors and Affiliations

  1. Department of Cardiology, Ehime Prefectural Central Hospital, 83, Kasuga-machi, Matsuyama, 790-0024, Japan

    Akinori Higaki MD, PhD, Tsukasa Kurokawa MD, Hikaru Okabe MD, Takuro Kazatani MD, Shinsuke Kido MD, Tetsuya Aono MD, Kensho Matsuda MD, MSc, Yuta Tanaka MD, Saki Hosokawa MD, Tetsuya Kosaki MD, Go Kawamura MD, Tatsuya Shigematsu MD, Yoshitaka Kawada MD, Go Hiasa MD, PhD, Tadakatsu Yamada MD & Hideki Okayama MD, PhD

  2. Department of Radiology, Ehime University Graduate School of Medicine, Toon, Japan

    Naoto Kawaguchi MD, PhD

  3. Department of Cardiology, Pulmonology, Hypertension & Nephrology, Ehime University Graduate School of Medicine, Toon, Japan

    Akinori Higaki MD, PhD

Authors
  1. Akinori Higaki MD, PhD
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Naoto Kawaguchi MD, PhD
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Tsukasa Kurokawa MD
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Hikaru Okabe MD
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Takuro Kazatani MD
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Shinsuke Kido MD
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Tetsuya Aono MD
    View author publications

    You can also search for this author in PubMed Google Scholar

  8. Kensho Matsuda MD, MSc
    View author publications

    You can also search for this author in PubMed Google Scholar

  9. Yuta Tanaka MD
    View author publications

    You can also search for this author in PubMed Google Scholar

  10. Saki Hosokawa MD
    View author publications

    You can also search for this author in PubMed Google Scholar

  11. Tetsuya Kosaki MD
    View author publications

    You can also search for this author in PubMed Google Scholar

  12. Go Kawamura MD
    View author publications

    You can also search for this author in PubMed Google Scholar

  13. Tatsuya Shigematsu MD
    View author publications

    You can also search for this author in PubMed Google Scholar

  14. Yoshitaka Kawada MD
    View author publications

    You can also search for this author in PubMed Google Scholar

  15. Go Hiasa MD, PhD
    View author publications

    You can also search for this author in PubMed Google Scholar

  16. Tadakatsu Yamada MD
    View author publications

    You can also search for this author in PubMed Google Scholar

  17. Hideki Okayama MD, PhD
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Akinori Higaki MD, PhD.

Ethics declarations

Disclosure

The authors have indicated that they have no financial conflict of interest.

Additional information

Publisher's Note

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

The authors of this article have provided a PowerPoint file, available for download at SpringerLink, which summarises the contents of the paper and is free for re-use at meetings and presentations. Search for the article DOI on SpringerLink.com.

The authors have also provided an audio summary of the article, which is available to download as ESM, or to listen to via the JNC/ASNC Podcast.

Supplementary Information

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Higaki, A., Kawaguchi, N., Kurokawa, T. et al. Content-based image retrieval for the diagnosis of myocardial perfusion imaging using a deep convolutional autoencoder. J. Nucl. Cardiol. 30, 540–549 (2023). https://doi.org/10.1007/s12350-022-03030-4

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s12350-022-03030-4

Keywords

Search

Navigation