Skip to main content

Advertisement

SpringerLink
Log in

Aortic wall segmentation in 18F-sodium fluoride PET/CT scans: Head-to-head comparison of artificial intelligence-based versus manual segmentation

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

Abstract

Background

We aimed to establish and test an automated AI-based method for rapid segmentation of the aortic wall in positron emission tomography/computed tomography (PET/CT) scans.

Methods

For segmentation of the wall in three sections: the arch, thoracic, and abdominal aorta, we developed a tool based on a convolutional neural network (CNN), available on the Research Consortium for Medical Image Analysis (RECOMIA) platform, capable of segmenting 100 different labels in CT images. It was tested on 18F-sodium fluoride PET/CT scans of 49 subjects (29 healthy controls and 20 angina pectoris patients) and compared to data obtained by manual segmentation. The following derived parameters were compared using Bland–Altman Limits of Agreement: segmented volume, and maximal, mean, and total standardized uptake values (SUVmax, SUVmean, SUVtotal). The repeatability of the manual method was examined in 25 randomly selected scans.

Results

CNN-derived values for volume, SUVmax, and SUVtotal were all slightly, i.e., 13-17%, lower than the corresponding manually obtained ones, whereas SUVmean values for the three aortic sections were virtually identical for the two methods. Manual segmentation lasted typically 1-2 hours per scan compared to about one minute with the CNN-based approach. The maximal deviation at repeat manual segmentation was 6%.

Conclusions

The automated CNN-based approach was much faster and provided parameters that were about 15% lower than the manually obtained values, except for SUVmean values, which were comparable. AI-based segmentation of the aorta already now appears as a trustworthy and fast alternative to slow and cumbersome manual segmentation.

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

Similar content being viewed by others

Abbreviations

AI:

Artificial intelligence

CNN:

Convolutional neural network

CT:

Computed tomography

NaF:

18F- sodium fluoride

PET:

Positron emission tomography

ROI:

Region of interest

SUV:

Standardized uptake value

VOI:

Volume of interest

References

  1. Doherty TM, Asotra K, Fitzpatrick LA, Qiao J-H, Wilkin DJ, Detrano RC, et al. Calcification in atherosclerosis: Bone biology and chronic inflammation at the arterial crossroads. Proc Natl Acad Sci 2003;100:11201-6.

    Article  CAS  Google Scholar 

  2. Tarkin JM, Dweck MR, Evans NR, Takx RA, Brown AJ, Tawakol A, et al. Imaging atherosclerosis. Circ Res 2016;118:750-69.

    Article  CAS  Google Scholar 

  3. McKenney-Drake ML, Moghbel MC, Paydary K, Alloosh M, Houshmand S, Moe S, et al. 18 F-NaF and 18 F-FDG as molecular probes in the evaluation of atherosclerosis. Eur J Nucl Med Mol Imaging 2018;45:2190-200.

    Article  Google Scholar 

  4. Libby P. The molecular mechanisms of the thrombotic complications of atherosclerosis. J Intern Med 2008;263:517-27.

    Article  CAS  Google Scholar 

  5. Derlin T, Janssen T, Salamon J, Veldhoen S, Busch J, Schön G, et al. Age-related differences in the activity of arterial mineral deposition and regional bone metabolism: A 18 F-sodium fluoride positron emission tomography study. Osteoporos Int 2015;26:199-207.

    Article  CAS  Google Scholar 

  6. Blomberg BA, de Jong PA, Thomassen A, Lam MG, Vach W, Olsen MH, et al. Thoracic aorta calcification but not inflammation is associated with increased cardiovascular disease risk: Results of the CAMONA study. Eur J Nucl Med Mol Imaging 2017;44:249-58.

    Article  Google Scholar 

  7. Paydary K, Revheim M-E, Emamzadehfard S, Gholami S, Pourhassan S, Werner TJ et al. Quantitative thoracic aorta calcification assessment by 18F-NaF PET/CT and its correlation with atherosclerotic cardiovascular disorders and increasing age. Eur Radiol 2020.

  8. Dou Q, Yu L, Chen H, Jin Y, Yang X, Qin J, et al. 3D deeply supervised network for automated segmentation of volumetric medical images. Med Image Anal 2017;41:40-54.

    Article  Google Scholar 

  9. Lindgren Belal S, Sadik M, Kaboteh R, Enqvist O, Ulén J, Poulsen MH, et al. Deep learning for segmentation of 49 selected bones in CT scans: First step in automated PET/CT-based 3D quantification of skeletal metastases. Eur J Radiol 2019;113:89-95.

    Article  Google Scholar 

  10. Mortensen MA, Borrelli P, Poulsen MH, Gerke O, Enqvist O, Ulén J, et al. Artificial intelligence-based versus manual assessment of prostate cancer in the prostate gland: A method comparison study. Clin Physiol Funct Imaging 2019;39:399-406.

    Article  CAS  Google Scholar 

  11. Polymeri E, Sadik M, Kaboteh R, Borrelli P, Enqvist O, Ulén J, et al. Deep learning-based quantification of PET/CT prostate gland uptake: Association with overall survival. Clin Physiol Funct Imaging 2020;40:106-13.

    Article  CAS  Google Scholar 

  12. Blomberg BA, Thomassen A, Takx RA, Vilstrup MH, Hess S, Nielsen AL, et al. Delayed sodium 18F-fluoride PET/CT imaging does not improve quantification of vascular calcification metabolism: Results from the CAMONA study. J Nucl Cardiol 2014;21:293-304.

    Article  Google Scholar 

  13. Trägårdh E, Borrelli P, Kaboteh R, Gillberg T, Ulén J, Enqvist O, et al. RECOMIA: A cloud-based platform for artificial intelligence research in nuclear medicine and radiology. EJNMMI Phys 2020;7:1-12.

    Article  Google Scholar 

  14. Ronneberger O, Fischer P, Brox T. U-net: Convolutional networks for biomedical image segmentation. International Conference on Medical image computing and computer-assisted intervention; 2015. p. 234-41.

  15. Kingma DP, Ba J. Adam: A method for stochastic optimization. arXiv preprint arXiv:14126980 2014.

  16. Carkeet A. Exact parametric confidence intervals for Bland-Altman limits of agreement. Optom Vis Sci 2015;92:e71-80.

    Article  Google Scholar 

  17. Gerke O. Reporting standards for a Bland-Altman agreement analysis: A review of methodological reviews. Diagnostics 2020;10:334.

    Article  Google Scholar 

  18. Høilund-Carlsen PF, Edenbrandt L, Alavi A. Global disease score (GDS) is the name of the game! Eur J Nucl Med Mol Imaging 2019;46:1768-72.

    Article  Google Scholar 

  19. Høilund-Carlsen PF, Sturek M, Alavi A, Gerke O. Atherosclerosis imaging with 18F-sodium fluoride PET: State-of-the-art review. Eur J Nucl Med Mol Imaging 2020;47:1538-51.

    Article  Google Scholar 

  20. Kinahan PE, Fletcher JW. Positron emission tomography-computed tomography standardized uptake values in clinical practice and assessing response to therapy. Seminars in Ultrasound, CT and MRI 2010;31:496-505.

    Article  Google Scholar 

  21. Natsis KI, Tsitouridis IA, Didagelos MV, Fillipidis AA, Vlasis KG, Tsikaras PD. Anatomical variations in the branches of the human aortic arch in 633 angiographies: Clinical significance and literature review. Surg Radiol Anat 2009;31:319.

    Article  Google Scholar 

  22. Høilund-Carlsen P, Lauritzen S, Marving J, Rasmussen S, Hesse B, Folke K, et al. The reliability of measuring left ventricular ejection fraction by radionuclide cardiography: Evaluation by the method of variance components. Heart 1988;59:653-62.

    Article  Google Scholar 

  23. Arbab-Zadeh A, Fuster V. The myth of the “vulnerable plaque”. J Am Coll Cardiol 2015;65:846.

    Article  Google Scholar 

  24. Duquette AA, Jodoin P-M, Bouchot O, Lalande A. 3D segmentation of abdominal aorta from CT-scan and MR images. Comput Med Imaging Graph 2012;36:294-303.

    Article  Google Scholar 

  25. Xie Y, Padgett J, Biancardi AM, Reeves AP. Automated aorta segmentation in low-dose chest CT images. Int J Comput Assist Radiol Surg 2014;9:211-9.

    Article  Google Scholar 

Download references

Disclosures

None declared.

Funding

The study was partly funded through a PhD scholarship to Reza Piri by the University of Southern Denmark, Odense, Denmark.

Author information

Authors and Affiliations

  1. Department of Nuclear Medicine, Odense University Hospital, 5000, Odense, Denmark

    Reza Piri MD, Amalie Horstmann Nøddeskou-Fink CS, Oke Gerke MSc, PhD & Poul Flemming Høilund-Carlsen MD, DMSc, Prof (Hon)

  2. Department of Clinical Research, University of Southern Denmark, Odense, Denmark

    Reza Piri MD, Oke Gerke MSc, PhD & Poul Flemming Høilund-Carlsen MD, DMSc, Prof (Hon)

  3. Department of Molecular and Clinical Medicine, Institute of Medicine, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden

    Lars Edenbrandt MD, DMSc

  4. Department of Clinical Physiology, Sahlgrenska University Hospital, Region Västra Götaland, Gothenburg, Sweden

    Lars Edenbrandt MD, DMSc

  5. Eigenvision AB, Malmö, Sweden

    Måns Larsson PhD & Olof Enqvist

  6. Department of Electrical Engineering, Chalmers University of Technology, Gothenburg, Sweden

    Olof Enqvist

Authors
  1. Reza Piri MD
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Lars Edenbrandt MD, DMSc
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Måns Larsson PhD
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Olof Enqvist
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Amalie Horstmann Nøddeskou-Fink CS
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Oke Gerke MSc, PhD
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Poul Flemming Høilund-Carlsen MD, DMSc, Prof (Hon)
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Reza Piri MD.

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

Piri, R., Edenbrandt, L., Larsson, M. et al. Aortic wall segmentation in 18F-sodium fluoride PET/CT scans: Head-to-head comparison of artificial intelligence-based versus manual segmentation. J. Nucl. Cardiol. 29, 2001–2010 (2022). https://doi.org/10.1007/s12350-021-02649-z

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s12350-021-02649-z

Keywords

Search

Navigation