Nothing Special   »   [go: up one dir, main page]
Skip to main content

Advertisement

Springer Nature Link
Log in

Study of CNN Capacity Applied to Left Ventricle Segmentation in Cardiac MRI

  • Original Research
  • Published:
SN Computer Science Aims and scope Submit manuscript

Abstract

CNN (Convolutional Neural Network) models have been successfully used for segmentation of the left ventricle (LV) in cardiac MRI (Magnetic Resonance Imaging), providing clinical measurements. In practice, two questions arise with deployment of CNNs: (1) when is it better to use a shallow model instead of a deeper one? (2) how the size of a dataset might change the network performance? We propose a framework to answer them, by experimenting with deep and shallow versions of three U-Net families, trained from scratch in six subsets varying from 100 to 10,000 images, different network sizes, learning rates and regularization values. 1620 models were evaluated using five-fold cross-validation by loss and DICE. The results indicate that: sample size affects performance more than architecture or hyper-parameters; in small samples the performance is more sensitive to hyper-parameters than architecture; the performance difference between shallow and deeper networks is not the same across families.

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
$34.99 /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 excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3

Similar content being viewed by others

Research Data

The data used in this research is available upon registration on the LVSC website: https://www.cardiacatlas.org/challenges/lv-segmentation-challenge/.

Code Availability

The code developed in this research is intellectual property of Zerbini Foundation and FoxConn and cannot be disclosed by the authors at this moment.

References

  1. Akaike H. A new look at the statistical model identification. IEEE Trans Autom Control. 1974;19(6):716–23.

    Article  MathSciNet  Google Scholar 

  2. Aziz YFA, Fadzli F, Rizal Azman R, Mohamed Sani F, Vijayananthan A, Nazri M. State of the heart: Cmr in coronary artery disease. Curr Med Imaging. 2013;9(3):201–13.

    Article  Google Scholar 

  3. Bartlett PL, Foster DJ, Telgarsky M. Spectrally-normalized margin bounds for neural networks. Adv Neur Inf Proc Sys. 2017;30:6241–50.

    Google Scholar 

  4. Benkendorf DJ, Hawkins CP. Effects of sample size and network depth on a deep learning approach to species distribution modeling. Ecol Inf. 2020;60:101137.

    Article  Google Scholar 

  5. Chaurasia A, Culurciello E. Linknet: exploiting encoder representations for efficient semantic segmentation. 2017. p. 1–4. https://doi.org/10.1109/VCIP.2017.8305148.

  6. Chen C, Qin C, Qiu H, Tarroni G, Duan J, Bai W, Rueckert D. Deep learning for cardiac image segmentation: a review. Front Cardiovasc Med. 2020;7:25.

    Article  Google Scholar 

  7. Dice LR. Measures of the amount of ecologic association between species. Ecology. 1945;26(3):297–302.

    Article  Google Scholar 

  8. Ghavami N, Hu Y, Gibson E, Bonmati E, Emberton M, Moore CM, Barratt DC. Automatic segmentation of prostate mri using convolutional neural networks: Investigating the impact of network architecture on the accuracy of volume measurement and mri-ultrasound registration. Med Image Anal. 2019;58:101558.

    Article  Google Scholar 

  9. Hastie TJ, Tibshirani RJ. Generalized additive models, vol. 43. Boca Raton: Chapman & Hall/CRC; 1990.

    MATH  Google Scholar 

  10. He K, Zhang X, Ren S, Sun J. Deep residual learning for image recognition. In: Proc. IEEE CVPR, 2016. p. 770–8.

  11. Hu J, Shen L, Sun G. Squeeze-and-excitation networks. In: Proc. IEEE CVPR, 2018. p. 7132–41.

  12. Lin TY, Dollár P, Girshick R, He K, Hariharan B, Belongie S. Feature pyramid networks for object detection. In: Proc. IEEE CVPR, 2017. pp. 2117–25.

  13. Martin CH, Mahoney MW. Rethinking generalization requires revisiting old ideas: statistical mechanics approaches and complex learning behavior. arXiv preprint arXiv:1710.09553, 2018.

  14. Neyshabur B, Li Z, Bhojanapalli S, LeCun Y, Srebro N. The role of over-parametrization in generalization of neural networks. In: Proc. ICLR. 2019. p. 1–20.

  15. Oktay O, Schlemper J, Folgoc LL, Lee M, Heinrich M, Misawa K, Mori K, McDonagh S, Hammerla NY, Kainz B, et al. Attention u-net: Learning where to look for the pancreas. In: Proc. Conf. MIDL–medical imaging with deep learning. 2018. p. 1–10.

  16. Ronneberger O, Fischer P, Brox T. U-net: convolutional networks for biomedical image segmentation. In: MICCAI. LNCS. 2015;9351:234–41.

    Google Scholar 

  17. Sandler M, Howard AG, Zhu M, Zhmoginov A, Chen L. Mobilenetv2: inverted residuals and linear bottlenecks. In: Proc. IEEE CVPR, 2018; p. 4510–20. https://doi.org/10.1109/CVPR.2018.00474.

  18. Shahinfar S, Meek P, Falzon G. how many images do i need? understanding how sample size per class affects deep learning model performance metrics for balanced designs in autonomous wildlife monitoring. Ecol Inf. 2020;57:101085.

    Article  Google Scholar 

  19. Simonyan K, Zisserman A. Very deep convolutional networks for large-scale image recognition. In: Proc. ICLR. 2015;1–14.

  20. Spencer JH, Quill JL, Bateman MG, Eggen MD, Howard SA, Goff RP, Howard BT, Quallich SG, Iaizzo PA. The benefits of the atlas of human cardiac anatomy website for the design of cardiac devices. Expert Rev Med Devices. 2013;10(6):729–34. https://doi.org/10.1586/17434440.2013.843449.

    Article  Google Scholar 

  21. Suinesiaputra A, Cowan BR, Al-Agamy AO, Elattar MA, Ayache N, Fahmy AS, Khalifa AM, Medrano-Gracia P, Jolly MP, Kadish AH, et al. A collaborative resource to build consensus for automated left ventricular segmentation of cardiac mr images. Med Image Anal. 2014;18(1):50–62.

    Article  Google Scholar 

  22. Tabachnick BG, Fidell LS. Experimental Design Using ANOVA. Belmont: Duxbury; 2007.

    Google Scholar 

  23. Tan M, Chen B, Pang R, Vasudevan V, Sandler M, Howard A, Le QV. Mnasnet: platform-aware neural architecture search for mobile. In: Proc. IEEE/CVF CVPR, 2019; p. 2820–8.

  24. Tan M., Le, Q. Efficientnet: rethinking model scaling for convolutional neural networks. In: Proc. ICML, PMLR 2019; p. 6105–14.

  25. Zhao H, Shi J, Qi X, Wang X, Jia J. Pyramid scene parsing network. In: Proc. IEEE CVPR, 2017; p. 2881–90.

  26. Zhou Z, Siddiquee MMR, Tajbakhsh N, Liang J. Unet++: a nested u-net architecture for medical image segmentation. In: Stoyanov D, Taylor Z, Carneiro G, Syeda-Mahmood T, et al., editors. Deep learning in medical image analysis and multimodal learning for clinical decision support. Springer; 2018. p. 3–11.

    Chapter  Google Scholar 

Download references

Acknowledgements

The authors thank Zerbini Foundation and FoxConn for supporting this research; A. Marco, C. Graves, D. Cardenas, J. Ferreira Jr. and R. Pereira for discussing ideas about experiments in practical scenarios; R. Moreno and M. Rebelo for initial manuscript revision and feedback.

Funding

This research was supported by Zerbini Foundation and FoxConn projects AIMED-CATI 030/2007 and FOXCONN-001/2019.

Author information

Authors and Affiliations

  1. Laboratorio de Informatica Biomedica, Faculdade de Medicina, Instituto do Coracao, Hospital das Clinicas HCFMUSP, Universidade de Sao Paulo, São Paulo, SP, Brazil

    Marcelo A. F. Toledo, Daniel M. Lima & Marco A. Gutierrez

  2. Laboratorio de Genetica e Cardiologia Molecular, Faculdade de Medicina, Instituto do Coracao, Hospital das Clinicas HCFMUSP, Universidade de Sao Paulo, São Paulo, SP, Brazil

    Jose E. Krieger

Authors
  1. Marcelo A. F. Toledo
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Daniel M. Lima
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Jose E. Krieger
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Marco A. Gutierrez
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Marcelo A. F. Toledo.

Ethics declarations

Conflict of interest/Competing interests

The authors declare that they have no conflict of interest.

Additional information

Publisher's Note

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

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Toledo, M.A.F., Lima, D.M., Krieger, J.E. et al. Study of CNN Capacity Applied to Left Ventricle Segmentation in Cardiac MRI. SN COMPUT. SCI. 2, 480 (2021). https://doi.org/10.1007/s42979-021-00897-x

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1007/s42979-021-00897-x

Keywords

Search

Navigation