Abstract

We developed and demonstrated real-time compressive sensing (CS) spectral domain optical coherence tomography (SD-OCT) B-mode imaging at excess of 70 fps. The system was implemented using a conventional desktop computer architecture having three graphics processing units. This result shows speed gain of 459 and 112 times compared to the best CS implementations based on the MATLAB and C++, respectively, and that real-time CS SD-OCT imaging can finally be realized.

© 2013 Optical Society of America

Full Article  |  PDF Article

References

  • View by:
  • |
  • |
  • |

  1. D. L. Donoho, IEEE Trans. Inf. Theory 52, 1289 (2006).
    [CrossRef]
  2. E. J. Candes, J. Romberg, and T. Tao, IEEE Trans. Inf. Theory 52, 48 (2006).
  3. X. Liu and J. U. Kang, Opt. Express 18, 22010 (2010).
    [CrossRef]
  4. N. Zhang, T. Huo, C. Wang, T. Chen, J. Zheng, and P. Xue, Opt. Lett. 37, 3075 (2012).
    [CrossRef]
  5. D. Xu, N. Vaswani, Y. Huang, and J. U. Kang, Opt. Lett. 37, 4209 (2012).
    [CrossRef]
  6. C. Liu, A. Wong, K. Bizheva, P. Fieguth, and H. Bie, Opt. Express 20, 10200 (2012).
    [CrossRef]
  7. G. Liu, J. Zhang, L. Yu, T. Xie, and Z. Chen, Appl. Opt. 48, 6365 (2009).
    [CrossRef]
  8. K. Zhang and J. U. Kang, Opt. Express 18, 11772 (2010).
    [CrossRef]
  9. M. Murphy, M. Alley, J. Demmel, K. Keutzer, S. Vasanawala, and M. Lustig, IEEE Trans. Med. Imaging 31, 1250 (2012).
    [CrossRef]
  10. S. Lee and S. J. Wright, “Implementing algorithms for signal and image reconstruction on graphical processing units,” (University of Wisconsin-Madison, 2008).
  11. D. Yang, G. D. Peterson, and H. Li, Parallel Computing 38, 421 (2012).
    [CrossRef]
  12. S. J. Wright, R. Nowak, and M. Figueiredo, IEEE Trans. Signal Process. 57, 2479 (2009).
    [CrossRef]
  13. NVIDIA CUDA C Programming Guide Version 5.0 (2012).
  14. J. Barzilai and J. M. Borwein, IMA Journal of Numerical Analysis 8, 141 (1988).
    [CrossRef]
  15. M. Harris, “Optimizing parallel reduction in CUDA,” (2007).
  16. M. Frigo and S. G. Johnson, Proc. IEEE 93, 216 (2005).
    [CrossRef]
  17. S. M. Potter, A. Mart, and J. Pine, Proc. SPIE 2869, 243 (1997).
    [CrossRef]
  18. S. P. Monacos, R. K. Lam, A. A. Portillo, and G. G. Ortiz, Proc. SPIE 4975, 115 (2003).
    [CrossRef]

2012 (5)

2010 (2)

2009 (2)

G. Liu, J. Zhang, L. Yu, T. Xie, and Z. Chen, Appl. Opt. 48, 6365 (2009).
[CrossRef]

S. J. Wright, R. Nowak, and M. Figueiredo, IEEE Trans. Signal Process. 57, 2479 (2009).
[CrossRef]

2006 (2)

D. L. Donoho, IEEE Trans. Inf. Theory 52, 1289 (2006).
[CrossRef]

E. J. Candes, J. Romberg, and T. Tao, IEEE Trans. Inf. Theory 52, 48 (2006).

2005 (1)

M. Frigo and S. G. Johnson, Proc. IEEE 93, 216 (2005).
[CrossRef]

2003 (1)

S. P. Monacos, R. K. Lam, A. A. Portillo, and G. G. Ortiz, Proc. SPIE 4975, 115 (2003).
[CrossRef]

1997 (1)

S. M. Potter, A. Mart, and J. Pine, Proc. SPIE 2869, 243 (1997).
[CrossRef]

1988 (1)

J. Barzilai and J. M. Borwein, IMA Journal of Numerical Analysis 8, 141 (1988).
[CrossRef]

Alley, M.

M. Murphy, M. Alley, J. Demmel, K. Keutzer, S. Vasanawala, and M. Lustig, IEEE Trans. Med. Imaging 31, 1250 (2012).
[CrossRef]

Barzilai, J.

J. Barzilai and J. M. Borwein, IMA Journal of Numerical Analysis 8, 141 (1988).
[CrossRef]

Bie, H.

Bizheva, K.

Borwein, J. M.

J. Barzilai and J. M. Borwein, IMA Journal of Numerical Analysis 8, 141 (1988).
[CrossRef]

Candes, E. J.

E. J. Candes, J. Romberg, and T. Tao, IEEE Trans. Inf. Theory 52, 48 (2006).

Chen, T.

Chen, Z.

Demmel, J.

M. Murphy, M. Alley, J. Demmel, K. Keutzer, S. Vasanawala, and M. Lustig, IEEE Trans. Med. Imaging 31, 1250 (2012).
[CrossRef]

Donoho, D. L.

D. L. Donoho, IEEE Trans. Inf. Theory 52, 1289 (2006).
[CrossRef]

Fieguth, P.

Figueiredo, M.

S. J. Wright, R. Nowak, and M. Figueiredo, IEEE Trans. Signal Process. 57, 2479 (2009).
[CrossRef]

Frigo, M.

M. Frigo and S. G. Johnson, Proc. IEEE 93, 216 (2005).
[CrossRef]

Harris, M.

M. Harris, “Optimizing parallel reduction in CUDA,” (2007).

Huang, Y.

Huo, T.

Johnson, S. G.

M. Frigo and S. G. Johnson, Proc. IEEE 93, 216 (2005).
[CrossRef]

Kang, J. U.

Keutzer, K.

M. Murphy, M. Alley, J. Demmel, K. Keutzer, S. Vasanawala, and M. Lustig, IEEE Trans. Med. Imaging 31, 1250 (2012).
[CrossRef]

Lam, R. K.

S. P. Monacos, R. K. Lam, A. A. Portillo, and G. G. Ortiz, Proc. SPIE 4975, 115 (2003).
[CrossRef]

Lee, S.

S. Lee and S. J. Wright, “Implementing algorithms for signal and image reconstruction on graphical processing units,” (University of Wisconsin-Madison, 2008).

Li, H.

D. Yang, G. D. Peterson, and H. Li, Parallel Computing 38, 421 (2012).
[CrossRef]

Liu, C.

Liu, G.

Liu, X.

Lustig, M.

M. Murphy, M. Alley, J. Demmel, K. Keutzer, S. Vasanawala, and M. Lustig, IEEE Trans. Med. Imaging 31, 1250 (2012).
[CrossRef]

Mart, A.

S. M. Potter, A. Mart, and J. Pine, Proc. SPIE 2869, 243 (1997).
[CrossRef]

Monacos, S. P.

S. P. Monacos, R. K. Lam, A. A. Portillo, and G. G. Ortiz, Proc. SPIE 4975, 115 (2003).
[CrossRef]

Murphy, M.

M. Murphy, M. Alley, J. Demmel, K. Keutzer, S. Vasanawala, and M. Lustig, IEEE Trans. Med. Imaging 31, 1250 (2012).
[CrossRef]

Nowak, R.

S. J. Wright, R. Nowak, and M. Figueiredo, IEEE Trans. Signal Process. 57, 2479 (2009).
[CrossRef]

Ortiz, G. G.

S. P. Monacos, R. K. Lam, A. A. Portillo, and G. G. Ortiz, Proc. SPIE 4975, 115 (2003).
[CrossRef]

Peterson, G. D.

D. Yang, G. D. Peterson, and H. Li, Parallel Computing 38, 421 (2012).
[CrossRef]

Pine, J.

S. M. Potter, A. Mart, and J. Pine, Proc. SPIE 2869, 243 (1997).
[CrossRef]

Portillo, A. A.

S. P. Monacos, R. K. Lam, A. A. Portillo, and G. G. Ortiz, Proc. SPIE 4975, 115 (2003).
[CrossRef]

Potter, S. M.

S. M. Potter, A. Mart, and J. Pine, Proc. SPIE 2869, 243 (1997).
[CrossRef]

Romberg, J.

E. J. Candes, J. Romberg, and T. Tao, IEEE Trans. Inf. Theory 52, 48 (2006).

Tao, T.

E. J. Candes, J. Romberg, and T. Tao, IEEE Trans. Inf. Theory 52, 48 (2006).

Vasanawala, S.

M. Murphy, M. Alley, J. Demmel, K. Keutzer, S. Vasanawala, and M. Lustig, IEEE Trans. Med. Imaging 31, 1250 (2012).
[CrossRef]

Vaswani, N.

Wang, C.

Wong, A.

Wright, S. J.

S. J. Wright, R. Nowak, and M. Figueiredo, IEEE Trans. Signal Process. 57, 2479 (2009).
[CrossRef]

S. Lee and S. J. Wright, “Implementing algorithms for signal and image reconstruction on graphical processing units,” (University of Wisconsin-Madison, 2008).

Xie, T.

Xu, D.

Xue, P.

Yang, D.

D. Yang, G. D. Peterson, and H. Li, Parallel Computing 38, 421 (2012).
[CrossRef]

Yu, L.

Zhang, J.

Zhang, K.

Zhang, N.

Zheng, J.

Appl. Opt. (1)

IEEE Trans. Inf. Theory (2)

D. L. Donoho, IEEE Trans. Inf. Theory 52, 1289 (2006).
[CrossRef]

E. J. Candes, J. Romberg, and T. Tao, IEEE Trans. Inf. Theory 52, 48 (2006).

IEEE Trans. Med. Imaging (1)

M. Murphy, M. Alley, J. Demmel, K. Keutzer, S. Vasanawala, and M. Lustig, IEEE Trans. Med. Imaging 31, 1250 (2012).
[CrossRef]

IEEE Trans. Signal Process. (1)

S. J. Wright, R. Nowak, and M. Figueiredo, IEEE Trans. Signal Process. 57, 2479 (2009).
[CrossRef]

IMA Journal of Numerical Analysis (1)

J. Barzilai and J. M. Borwein, IMA Journal of Numerical Analysis 8, 141 (1988).
[CrossRef]

Opt. Express (3)

Opt. Lett. (2)

Parallel Computing (1)

D. Yang, G. D. Peterson, and H. Li, Parallel Computing 38, 421 (2012).
[CrossRef]

Proc. IEEE (1)

M. Frigo and S. G. Johnson, Proc. IEEE 93, 216 (2005).
[CrossRef]

Proc. SPIE (2)

S. M. Potter, A. Mart, and J. Pine, Proc. SPIE 2869, 243 (1997).
[CrossRef]

S. P. Monacos, R. K. Lam, A. A. Portillo, and G. G. Ortiz, Proc. SPIE 4975, 115 (2003).
[CrossRef]

Other (3)

M. Harris, “Optimizing parallel reduction in CUDA,” (2007).

NVIDIA CUDA C Programming Guide Version 5.0 (2012).

S. Lee and S. J. Wright, “Implementing algorithms for signal and image reconstruction on graphical processing units,” (University of Wisconsin-Madison, 2008).

Supplementary Material (3)

» Media 1: MOV (569 KB)     
» Media 2: MOV (480 KB)     
» Media 3: MOV (478 KB)     

Cited By

OSA participates in CrossRef's Cited-By Linking service. Citing articles from OSA journals and other participating publishers are listed here.

Alert me when this article is cited.


Figures (3)

Fig. 1.
Fig. 1.

Reconstruction procedure of an A-scan.

Fig. 2.
Fig. 2.

Triple-GPUs architecture.

Fig. 3.
Fig. 3.

(a) Original OCT image of an orange; (b) CS reconstruction result; (c) line rate versus iteration number; (d) speedup versus iteration number; (e), (f) PSNR versus iteration number and τ, respectively, for B-mode imaging of three samples: (1) orange (Media 1); (2) human skin (Media 2); (3) mouse cornea (Media 3). The scale bars in (a) and (b) represent 100 μm.

Equations (4)

Equations on this page are rendered with MathJax. Learn more.

minimizex12Fuxyu22+τWx1,
xk+1=minz12zuk22+ταkz1,
xik+1=minzi12(ziuik)2+ταk|zi|=soft(uik,ταk),
αk=Fusk22/sk22,

Metrics