Evaluation of advanced Lukas–Kanade optical flow on thoracic 4D-CT

Abstract

Extensive use of high frequency imaging in medical applications permit the estimation of velocity fields which corresponds to motion of landmarks in the imaging field. The focus of this work is on the development of a robust local optical flow algorithm for velocity field estimation in medical applications. Local polynomial fits to the medical image intensity-maps are used to generate convolution operators to estimate the spatial gradients. A novel polynomial window function with a compact support is used to differentially weight the optical flow gradient constraints in the region of interest. Tikhonov regularization is exploited to synthesize a well posed optimization problem and to penalize large displacements. The proposed algorithm is tested and validated on benchmark datasets for deformable image registration. The ten datasets include large and small deformations, and illustrate that the proposed algorithm outperforms or is competitive with other algorithms tested on this dataset, when using mean and variance of the displacement error as performance metrics.

This is a preview of subscription content, access via your institution.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7

Notes

  1. 1.

    The author would like to acknowledge the help of Richard Castillo, MS who processed the result of our algorithm and provided feedback regarding its performance.

  2. 2.

    Figure created using renderings provided by Richard Castillo, MS, The University of Texas M D Anderson Cancer Center

  3. 3.

    Note that this analysis was carried out using the publicly available dataset of 300 landmarks per case. Thus, the numerical values of “Polynomial Filter + Inverse Flow” differ from the values of “ALK” in Table 2 and Figs. 3 and 4. Those were obtained in the final, external analysis using the (non-public) comprehensive dataset.

References

  1. 1.

    Bouguet JY. Pyramidal implementation of the Lucas Kanade feature tracker: description of the algorithm. Intel Corporation 2001.

  2. 2.

    Castillo E, Castillo R, Martinez J, Shenoy M, Guerrero T. Four-dimensional deformable image registration using trajectory modeling. Phys Med Biol. 2010;55(1):305.

    PubMed  Article  Google Scholar 

  3. 3.

    Castillo E, Castillo R, Zhang Y, Guerrero T. Compressible image registration for thoracic computed tomography images. J Med Biol Eng. 2009;29(5):222–233.

    Google Scholar 

  4. 4.

    Castillo R, Castillo E, Guerra R, Johnson VE, McPhail T, Garg AK, Guerrero T. A framework for evaluation of deformable image registration spatial accuracy using large landmark point sets. Phys Med Biol. 2009;54:1849.

    PubMed  Article  Google Scholar 

  5. 5.

    Chen M, Lu W, Chen Q, Ruchala KJ, Olivera GH. A simple fixed-point approach to invert a deformation field. Med Phys. 2008;35(1):81.

    PubMed  Article  CAS  Google Scholar 

  6. 6.

    Christensen G, Johnson H. Consistent image registration. IEEE Trans Med Imaging. 2001;20(7):568–582.

    PubMed  Article  CAS  Google Scholar 

  7. 7.

    Delaunay BN. Sur la sphère vide. Bull Acad Sci USSR. 1934;6:793–800.

    Google Scholar 

  8. 8.

    Duan Q, Herz S, Ingrassia C, Costa K, Holmes J, Laine A, Angelini E, Gerard O, Homma S. Dynamic cardiac information from optical flow using four dimensional ultrasound. In: 27th annual international conference of the engineering in medicine and biology society. IEEE-EMBS, vol 4. 2005. p. 4465–4468.

  9. 9.

    Gessat M, Frauenfelder T, Altwegg L, Grünenfelder J, Falk V. Transcatheter aortic valve implantation. Role of imaging. Aswan Heart Centre Sci Pract Ser. 2011;1:3.

    Google Scholar 

  10. 10.

    Gu X, Pan H, Liang Y, Castillo R, Yang D, Choi D, Castillo E, Majumdar A, Guerrero T, Jiang SB. Implementation and evaluation of various demons deformable image registration algorithms on a gpu. Phys Med Biol. 2010;55(1):207–219.

    PubMed  Article  Google Scholar 

  11. 11.

    Horn BK, Schunck BG. Determining optical flow. Artif Intell. 1981;17(1–3):185–203.

    Article  Google Scholar 

  12. 12.

    Lucas BD, Kanade T. An iterative image registration technique with an application to stereo vision. In: Proceedings of the 7th international joint conference on artificial intelligence, vol 3. 1981. p. 674–679.

  13. 13.

    Murphy K, van Ginneken B, Reinhardt JM, Kabus S, Ding K, Deng X, Cao K, Du K, Christensen GE, Garcia V, Vercauteren T, Ayache N, Commowick O, Malandain G, Glocker B, Paragios N, Navab N, Gorbunova V, Sporring J, de Bruijne M, Han X, Heinrich MP, Schnabel Ja, Jenkinson M, Lorenz C, Modat M, McClelland JR, Ourselin S, Muenzing SEa, Viergever Ma, De Nigris D, Collins DL, Arbel T, Peroni M, Li R, Sharp GC, Schmidt-Richberg A, Ehrhardt J, Werner R, Smeets D, Loeckx D, Song G, Tustison N, Avants B, Gee JC, Staring M, Klein S, Stoel BC, Urschler M, Werlberger M, Vandemeulebroucke J, Rit S, Sarrut D, Pluim JPW. Evaluation of registration methods on thoracic ct: the empire10 challenge. IEEE Trans Med Imaging. 2011;30(11):1901–1920.

    PubMed  Article  Google Scholar 

  14. 14.

    Singla P, Singh T. Desired order continuous polynomial time window functions for harmonic analysis. IEEE Trans Instrum Meas. 2010;59(9):2475–2481.

    Article  Google Scholar 

  15. 15.

    Veronesi F, Corsi C, Caiani EG, Sarti A, Lamberti C. Tracking of left ventricular long axis from real-time three-dimensional echocardiography using optical flow techniques. IEEE Trans Inf Technol Biomed. 2006;10(1):174–181.

    PubMed  Article  CAS  Google Scholar 

  16. 16.

    Yang D, Lu W, Low DA, Deasy JO, Hope AJ, El Naqa I. 4d-ct motion estimation using deformable image registration and 5d respiratory motion modeling. Med Phys. 2008;35(10):4577–4590.

    PubMed  Article  Google Scholar 

  17. 17.

    Zientara GP, Saiviroonporn P, Morrison PR, Fried MP, Hushek SG, Kikinis R, Jolesz F. MRI monitoring of laser ablation using optical flow. J Magn Reson Imaging. 1998;8(6):1306–1318.

    PubMed  Article  CAS  Google Scholar 

Download references

Acknowledgments

The authors would like to acknowledge the help of Richard Castillo, MS who processed the result of our algorithm and assessed its performance. The authors would like to thank the National Science Foundation, which funded this project under grant CMMI-#0928630. C. Hoog Antink also would like to thank Fulbright and the German National Academic Foundation for partial funding. The authors have no financial relationship with the National Science Foundation which funded the research, the results of which are presented in this paper.

Conflict of interest

The authors declare that they have no conflict of interest.

Author information

Affiliations

Authors

Corresponding author

Correspondence to Tarunraj Singh.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Hoog Antink, C.B., Singh, T., Singla, P. et al. Evaluation of advanced Lukas–Kanade optical flow on thoracic 4D-CT. J Clin Monit Comput 27, 433–441 (2013). https://doi.org/10.1007/s10877-013-9454-5

Download citation

Keywords

  • Optical flow
  • Acute illness
  • Deformable image registration
  • Radiotherapy planning