Architecture of Multiple Convolutional Neural Networks to Construct a Subject-Specific Knee Model for Estimating Local Specific Absorption Rate

Abstract

Electromagnetic simulation is a credible way to estimate the local specific absorption rate (SAR), which is a key consideration in high-field magnetic resonance imaging of the knee joint. To construct a subject-specific knee model, which is critical for SAR simulation, we proposed an architecture comprising multiple convolutional neural networks. Knee tissues were segmented by three U-Nets. Each network was responsible for the segmentation of two tissues that have relatively similar volumes and distinct intensity distributions (muscle and fat, cancellous and cortical bone, and cartilage and meniscus). Additionally, a weighted loss function was used to further alleviate the effect of class imbalance of the segmented tissues. The outputs of these three networks were merged and morphological filtering was used as the post-processing to eliminate holes and isolated voxels. This method was compared with three other segmentation methods. Good segmentation performance was demonstrated on the test set, and the proposed method was found to be superior to the other methods according to several quantitative measures. Meanwhile, local SAR in a 3 T coil using models constructed with the proposed method, manual delineation, and the comparison methods were also evaluated on the test set. On the whole, the maximum values of SAR10g of the models constructed by the proposed method were closer to the results of manual delineation. Overall, the proposed method exhibits promising potential for precisely constructing knee models for SAR simulation.

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

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

Data Availability

Yes (We will share them if needed).

Code Availability

Yes (We will share them if needed).

References

  1. 1.

    S.K. Pakin, J. Xu, M.E. Schweitzer, R.R. Regatte, Magn. Reson. Med. 56(3), 563–571 (2006)

    Article  Google Scholar 

  2. 2.

    A. Watts, R.W. Stobbe, C. Beaulieu, Magn. Reson. Med. 66(3), 697–705 (2011)

    Article  Google Scholar 

  3. 3.

    J. Jin, E. Weber, A. Destruel, K. O’Brien, B. Henin, C. Engstrom, S. Crozier, Magn. Reson. Med. 79(3), 1804–1816 (2018)

    Article  Google Scholar 

  4. 4.

    V. Gagliardi, A. Retico, L. Biagi, G. Aringhieri, V. Zampa, M.R. Symms, G. Tiberi, M. Tosetti, in Proceedings of the IEEE International Symposium on Medical Measurements and Applications (Rome, Italy, 11-13 June 2018) https://doi.org/10.1109/MeMeA.2018.8438709

  5. 5.

    U. Katscher, T. Voigt, C. Findeklee, P. Vernickel, K. Nehrke, O. Doessel, IEEE Trans. Med. Imag. 28(9), 1365–1374 (2009)

    Article  Google Scholar 

  6. 6.

    U. Katscher, C. Findeklee, T. Voigt, Magn. Reson. Med. 68(6), 1911–1918 (2012)

    Article  Google Scholar 

  7. 7.

    H. Homann, I. Graesslin, H. Eggers, K. Nehrke, P. Vernickel, U. Katscher, O. Dössel, P. Börnert, Magn. Reson. Mater. Phy. 25(3), 193–204 (2012)

    Article  Google Scholar 

  8. 8.

    M. Murbach, E. Neufeld, E. Cabot, E. Zastrow, J. Córcoles, W. Kainz, N. Kuster, Magn. Reson. Med. 76(3), 986–997 (2016)

    Article  Google Scholar 

  9. 9.

    V. Hartwig, G. Giovannetti, N. Vanello, L. Landini, M.F. Santarelli, Appl. Magn. Reson. 38(3), 337–348 (2010)

    Article  Google Scholar 

  10. 10.

    J. Jin, F. Liu, E. Weber, S. Crozier, Phys. Med. Biol. 57(24), 8153–8171 (2012)

    Article  Google Scholar 

  11. 11.

    T. Voigt, H. Homann, U. Katscher, O. Doessel, Magn. Reson. Med. 68(4), 1117–1126 (2012)

    Article  Google Scholar 

  12. 12.

    F.F.J. Simonis, A.J.E. Raaijmakers, J.J.W. Lagendijk, C.A.T. van den Berg, Magn. Reson. Med. 77(4), 1691–1700 (2017)

    Article  Google Scholar 

  13. 13.

    M.S. Mallikarjunaswamy, M.S. Holi, R. Raman, J. Med. Imag. Health In. 5(3), 552–560 (2015)

    Google Scholar 

  14. 14.

    C.N. Öztürk, S. Albayrak, Comput. Biol. Med. 72, 90–107 (2016)

    Article  Google Scholar 

  15. 15.

    J. Tang, S. Millington, S.T. Acton, J. Crandall, S. Hurwitz, IEEE Trans. Biomed. Eng. 53(5), 896–907 (2006)

    Article  Google Scholar 

  16. 16.

    J. Fripp, S. Crozier, S.K. Warfield, S. Ourselin, IEEE Trans. Med. Imag. 29(1), 55–64 (2010)

    Article  Google Scholar 

  17. 17.

    T.G. Williams, A.P. Holmes, J.C. Waterton, R.A. Maciewicz, C.E. Hutchinson, R.J. Moots, A.F. Nash, C.J. Taylor, IEEE Trans. Med. Imag. 29(8), 1541–1559 (2010)

    Article  Google Scholar 

  18. 18.

    J.G. Tamez-Peña, J. Farber, P.C. Gonzalez, E. Schreyer, E. Schneider, S. Totterman, IEEE Trans. Biomed. Eng. 59(4), 1177–1186 (2012)

    Article  Google Scholar 

  19. 19.

    L. Shan, C. Zach, C. Charles, M. Niethammer, Med. Image Anal. 18(7), 1233–1246 (2014)

    Article  Google Scholar 

  20. 20.

    K. Zhang, W. Lu, P. Marziliano, Magn. Reson. Imaging. 31(10), 1731–1743 (2013)

    Article  Google Scholar 

  21. 21.

    P. Wang, X. He, Y. Li, X. Zhu, W. Chen, M. Qiu, J. Med. Imag. Health In. 6(4), 948–956 (2016)

    Google Scholar 

  22. 22.

    N. Tajbakhsh, J.Y. Shin, S.R. Gurudu, R.T. Hurst, C.B. Kendall, M.B. Gotway, J. Liang, IEEE Trans. Med. Imaging 35(5), 1299–1312 (2016)

    Article  Google Scholar 

  23. 23.

    H. Greenspan, B. Van Ginneken, R.M. Summers, IEEE Trans. Med. Imaging 35(5), 1153–1159 (2016)

    Article  Google Scholar 

  24. 24.

    O. Ronneberger, P. Fischer, T. Brox, The 18th international conference on medical image computing and computer assisted interventions (MICCAI 2015, Munich, Germany, 2015) 234–241 (2015)

  25. 25.

    F. Liu, Z. Zhou, H. Jang, A. Samsonov, G. Zhao, R. Kijowski, Magn. Reson. Med. 79(4), 2379–2391 (2018)

    Article  Google Scholar 

  26. 26.

    Z. Zhou, G. Zhao, R. Kijowski, F. Liu, Magn. Reson. Med. 80(6), 2759–2770 (2018)

    Article  Google Scholar 

  27. 27.

    B. Norman, V. Pedoia, S. Majumdar, Radiology 288(1), 177–185 (2018)

    Article  Google Scholar 

  28. 28.

    A. Tack, A. Mukhopadhyay, S. Zachow, Osteoarthr. Cartilage 26(5), 680–688 (2018)

    Article  Google Scholar 

  29. 29.

    F. Ambellana, A. Tack, M. Ehlke, S. Zachow, Med. Image Anal. 52, 109–118 (2019)

    Article  Google Scholar 

  30. 30.

    E.A. Rashed, J. Gomez-Tames, A. Hirata, NeuroImage. 202(15), 116132 (2019). (1–16)

    Article  Google Scholar 

  31. 31.

    E.A. Rashed, Y.L. Diao, A. Hirata, Phys. Med. Biol. 65, 6 (2020)

    Article  Google Scholar 

  32. 32.

    E.F. Meliadò, A.J.E. Raaijmakers, A. Srizzi, B.R. Steensma, M. Maspero, M.H.F. Savenije, P.R. Luijten, C.A.T. van den Berg, Magn. Reson. Med. 83(2), 695–711 (2019)

    Article  Google Scholar 

  33. 33.

    H. Homann, P. Börnert, H. Eggers, K. Nehrke, O. Dössel, I. Graesslin, Magn Reson Med. 66(6), 1767–1776 (2011)

    Article  Google Scholar 

  34. 34.

    A. Christ, W. Kainz, E.G. Hahn, K. Honegger, M. Zefferer, E. Neufeld, W. Rascher, R. Janka, W. Bautz, C. Ji, B. Kiefer, P. Schmitt, H.P. Hollenbach, S. Jianxiang, M. Oberle, D. Szczerba, A. Kam, J.W. Guag, N. Kuster, Phys. Med. Biol. 55(2), N23–N38 (2010)

    ADS  Article  Google Scholar 

  35. 35.

    C.M. Collins, M.B. Smith, J Magn Reson Imaging 18(3), 383–388 (2003)

    Article  Google Scholar 

  36. 36.

    Tissue properties in database of IT'IS Foundation. https://itis.swiss/virtual-population/tissue-properties/database/dielectric-properties/. Accessed 10 Jan 2020

  37. 37.

    S. Wolf, D. Diehl, M. Gebhardt, J. Mallow, O. Speck, Magn. Reson. Med. 69(4), 1157–1168 (2013)

    Article  Google Scholar 

  38. 38.

    D.M. Peterson, C.E. Carruthers, B.L. Wolverton, K. Meister, M. Werner, G.R. Duensing, J.R. Fitzsimmons, Magn. Reson. Med. 42(2), 215–221 (1999)

    Article  Google Scholar 

  39. 39.

    S. Orzada, M.E. Ladd, A.K. Bitz, Magn. Reson. Med. 78(2), 805–811 (2017)

    Article  Google Scholar 

  40. 40.

    V. Badrinarayanan, A. Kendall, R. Cipolla, IEEE Trans. Pattern Anal. Mach. Intell. 39(12), 2481–2495 (2017)

    Article  Google Scholar 

Download references

Acknowledgements

This work was supported by the research fund for the top scientific and technological innovation teams from Beijing University of Chemical Technology (No. buctylkjcx06). Many thanks are due to Professor Paul Glover of Sir Peter Mansfield Imaging Centre, the University of Nottingham, for his kind help with coil modeling and electromagnetic simulation.

Funding

This work was supported by the research fund for the top scientific and technological innovation teams from Beijing University of Chemical Technology (No. buctylkjcx06).

Author information

Affiliations

Authors

Corresponding author

Correspondence to Liang Xiao.

Ethics declarations

Conflicts of Interest

No.

Ethics Approval

This research was conducted with the approval of the Institutional Review Board of the Zhangjiakou Charity Hospital (Hebei, China).

Consent to Participate

All participants signed an informed consent form.

Consent to Publication

All authors agree the publication of this research.

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

Verify currency and authenticity via CrossMark

Cite this article

Xiao, L., Zhou, H., Chen, N. et al. Architecture of Multiple Convolutional Neural Networks to Construct a Subject-Specific Knee Model for Estimating Local Specific Absorption Rate. Appl Magn Reson (2020). https://doi.org/10.1007/s00723-020-01301-2

Download citation