access icon openaccess Invert-U-Net DNN segmentation model for MRI cardiac left ventricle segmentation

This is an open access article published by the IET under the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0/)
Received 18/07/2018, Accepted 02/08/2018, Revised 30/07/2018, Published 16/08/2018

In this study, a deeply supervised end-to-end model is presented for fully automated segmentation for cardiac magnetic resonance imaging (MRI) images. Firstly, the mechanism of deep neural network (DNN)-based segmentation is discussed with the relationship of channels and their distribution in network in depth. Following this idea, a U-Net-based model, namely an invert-U-Net model is presented with an innovative filter number structure. Based on the invert-U-Net model, the experiment is carefully designed, and the hyper-parameters are considerately arranged. Finally, the model is applied and evaluated using Sunnybrook MR datasets from the MICCAI 2009 LV segmentation challenge and the experimental result shows that it outperforms the state-of-the-art methods.

Inspec keywords: biomedical MRI; cardiology; neural nets; image segmentation; medical image processing; image filtering

Other keywords: MICCAI 2009 LV segmentation challenge; deeply supervised end-to-end model; fully automated segmentation; Sunnybrook MR datasets; MRI cardiac left ventricle segmentation; deep neural network-based segmentation; invert-U-Net DNN segmentation model; MRI; innovative filter number structure; cardiac magnetic resonance imaging images; channel distribution

Subjects: Optical, image and video signal processing; Biomedical magnetic resonance imaging and spectroscopy; Neural computing techniques; Medical magnetic resonance imaging and spectroscopy; Patient diagnostic methods and instrumentation; Biology and medical computing; Computer vision and image processing techniques

References

    1. 1)
      • 7. Wolterink, J.M., Leiner, T., Viergever, M.A., et al: ‘Automatic segmentation and disease classification using cardiac cine MR images’, arXiv preprint arXiv:1708.01141, 2017.
    2. 2)
      • 14. ‘LV SEGMENTATION CHALLENGE’. Available at https://smial.sri.utoronto.ca/LV_Challenge/Evaluation.html, accessed 16 March 2018.
    3. 3)
      • 9. Baumgartner, C.F., Koch, L.M., Pollefeys, M., et al: ‘An exploration of 2D and 3D.D.eep learning techniques for cardiac MR image segmentation’, arXiv preprint arXiv:1709.04496, 2017.
    4. 4)
      • 12. Petitjean, C., Dacher, J.N.: ‘A review of segmentation methods in short axis cardiac MR images’, Med. Image Anal., 2011, 15, (2), pp. 169–184.
    5. 5)
      • 6. Ronneberger, O., Fischer, P., Brox, T.: ‘U-net: convolutional networks for biomedical image segmentation’. Int. Conf. Medical image computing and computer-assisted intervention, Springer, Cham, 2015, pp. 234–241.
    6. 6)
      • 11. Radau, P., Li, Y., Connelly, K., et al: ‘Evaluation framework for algorithms segmenting short axis cardiac MRI’, The MIDAS Journal – Cardiac MR Left Ventricle Segmentation Challenge, 2009, 49.
    7. 7)
      • 4. Avendi, M.R., Kheradvar, A., Jafarkhani, H.: ‘A combined deep-learning and deformable-model approach to fully automatic segmentation of the left ventricle in cardiac MRI’, Med. Image Anal., 2016, 30, pp. 108–119.
    8. 8)
      • 10. Tan, L.K., Liew, Y.M., Lim, E., et al: ‘Convolutional neural network regression for short-axis left ventricle segmentation in cardiac cine MR sequences’, Med. Image Anal., 2017, 39, pp. 78–86.
    9. 9)
      • 15. Lu, Y., Radau, P., Connelly, K., et al: ‘Automatic image-driven segmentation of left ventricle in cardiac cine MRI’, The MIDAS J., 2009, 49, p. 2.
    10. 10)
      • 8. Isensee, F., Jaeger, P., Full, P.M., et al: ‘Automatic cardiac disease assessment on cine-MRI via time-series segmentation and domain specific features’, arXiv preprint arXiv:1707.00587, 2017.
    11. 11)
      • 16. Poudel, R.P.K., Lamata, P., Montana, G.: ‘Recurrent fully convolutional neural networks for multi-slice MRI cardiac segmentation’ Reconstruction, Segmentation, and Analysis of Medical Images. (Springer, Cham, 2016), pp. 83–94.
    12. 12)
      • 3. Long, J., Shelhamer, E., Darrell, T.: ‘Fully convolutional networks for semantic segmentation’. 2015 IEEE Conf. on Computer Vision and Pattern Recognition, Boston, USA, June 2015, pp. 3431–3440.
    13. 13)
      • 1. Mozaffarian, D., Benjamin, E.J., Go, A.S., et al: ‘Heart disease and stroke statistics-2016 update: a report from the American Heart Association’, Circulation, 2016, 133, (4), pp. e38–e360.
    14. 14)
      • 13. Ngo, T.A., Lu, Z., Carneiro, G.: ‘Combining deep learning and level set for the automated segmentation of the left ventricle of the heart from cardiac cine magnetic resonance’, Med. Image Anal., 2017, 35, pp. 159–171.
    15. 15)
      • 5. Tran, P.V.: ‘A fully convolutional neural network for cardiac segmentation in short-axis MRI’, arXiv preprint arXiv:1604.00494, 2016.
    16. 16)
      • 2. Tee, M., Noble, J.A., Bluemke, D.A.: ‘Imaging techniques for cardiac strain and deformation: comparison of echocardiography, cardiac magnetic resonance and cardiac computed tomography’, Expert Rev. Cardiovasc. Ther., 2013, 11, (2), pp. 221–231.
http://iet.metastore.ingenta.com/content/journals/10.1049/joe.2018.8302
Loading

Related content

content/journals/10.1049/joe.2018.8302
pub_keyword,iet_inspecKeyword,pub_concept
6
6
Loading