Abstract

Single-frame blood flow maps from laser speckle contrast imaging (LSCI) contain high spatiotemporal variation that obscures high spatial-frequency vascular features, making precise image registration for signal amplification challenging. In this work, novel bivariate standardized moment filters (BSMFs) were used to provide stable measures of vessel edge location, permitting a more robust LSCI registration. Relatedly, BSMFs enabled the stable reconstruction of vessel edges from sparsely distributed blood flow map outliers, which were found to retain most of the temporal dynamics. Consequently, data discarding and BSMF-based reconstruction enable efficient real-time quantitative LSCI data compression. Smaller LSCI-kernels produced log-normal blood flow distributions, enhancing sparse-to-dense inference.

© 2018 Optical Society of America under the terms of the OSA Open Access Publishing Agreement

Full Article  |  PDF Article
OSA Recommended Articles
Flux or speed? Examining speckle contrast imaging of vascular flows

S. M. Shams Kazmi, Ehssan Faraji, Mitchell A. Davis, Yu-Yen Huang, Xiaojing J. Zhang, and Andrew K. Dunn
Biomed. Opt. Express 6(7) 2588-2608 (2015)

Laser speckle contrast imaging with extended depth of field for in-vivo tissue imaging

Iliya Sigal, Raanan Gad, Antonio M. Caravaca-Aguirre, Yaaseen Atchia, Donald B. Conkey, Rafael Piestun, and Ofer Levi
Biomed. Opt. Express 5(1) 123-135 (2014)

Evaluation of laser speckle contrast imaging as an intrinsic method to monitor blood brain barrier integrity

Suzie Dufour, Yaaseen Atchia, Raanan Gad, Dene Ringuette, Iliya Sigal, and Ofer Levi
Biomed. Opt. Express 4(10) 1856-1875 (2013)

References

  • View by:
  • |
  • |
  • |

  1. D. Ringuette, M. A. Jeffrey, S. Dufour, P. L. Carlen, and O. Levi, “Continuous multi-modality brain imaging reveals modified neurovascular seizure response after intervention,” Biomed. Opt. Express 8(2), 873–889 (2017).
    [Crossref] [PubMed]
  2. L. M. Richards, E. L. Towle, D. J. Fox, and A. K. Dunn, “Intraoperative laser speckle contrast imaging with retrospective motion correction for quantitative assessment of cerebral blood flow,” Neurophotonics 1(1), 015006 (2014).
    [Crossref]
  3. S. E. Skipetrov, J. Peuser, R. Cerbino, P. Zakharov, B. Weber, and F. Scheffold, “Noise in laser speckle correlation and imaging techniques,” Opt. Express 18(14), 14519–14534 (2010).
    [Crossref] [PubMed]
  4. I. Sigal, M. M Koletar, D. Ringuette, R. Gad, M. A. Jeffrey, P. L. Carlen, B. Stefanovic, and O. Levi, “Imaging brain activity during seizures in freely behaving rats using a miniature multi-modal imaging system,” Biomed. Opt. Express 7(9), 3596–3609 (2016).
    [Crossref] [PubMed]
  5. R. Farraro, O. Fathi, and B. Choi, “Handheld, point-of-care laser speckle imaging,” J. Biomed. Opt. 21(9), 094001 (2016).
    [Crossref]
  6. P. Miao, A. Rege, N. Li, N. V. Thakor, and S. Tong, “High resolution cerebral blood flow imaging by registered laser speckle contrast analysis,” IEEE Trans. Biomed. Eng. 57(5), 1152–1157 (2010).
    [Crossref] [PubMed]
  7. S. J. Kirkpatrick, D. D. Duncan, and E. M. Wells-Gray, “Detrimental effects of speckle-pixel size matching in laser speckle contrast imaging,” Opt. Lett. 33(24), 2886–2888 (2008).
    [Crossref] [PubMed]
  8. D. Coleman, “TIFF compression options: ZIP vs LZW” (TIFF compression options, 2018). https://havecamerawilltravel.com/photographer/tiff-image-compression/ .
  9. S. S Parikh, D. Ruiz, H. Kalva, G. Fernández-Escribano, and V. Adzic, “High bit-depth medical image compression with hevc,” IEEE J. Biomed. Health Informat. 22(2), 552–560 (2018).
    [Crossref]
  10. K. Amolins, Y. Zhang, and P. Dare, “Wavelet based image fusion techniques-An introduction, review and comparison,” ISPRS J. Photogramm. Remote Sens. 62(4), 249–263 (2007).
    [Crossref]
  11. D. Ringuette, I. Sigal, R. Gad, and O. Levi, “Reducing misfocus-related motion artefacts in laser speckle contrast imaging,” Biomed. Opt. Express 6(1), 266–276 (2015).
    [Crossref] [PubMed]
  12. C.-H. Teh and R. T. Chin, “On image analysis by the methods of moments,” IEEE Trans. Pattern Anal. Mach. Intell. 10(4), 496–513 (1988).
    [Crossref]
  13. D. Shen and H. H. S. Ip, “Discriminative wavelet shape descriptors for recognition of 2-d patterns,” Pattern Recognition 32(2), 151–165 (1999).
    [Crossref]
  14. H. Levy, D. Ringuette, and O. Levi, “Rapid monitoring of cerebral ischemia dynamics using laser-based optical imaging of blood oxygenation and flow,” Biomed. Opt. Express,  3(4), 777–791 (2012).
    [Crossref] [PubMed]
  15. P. Thevenaz, U. E Ruttimann, and M. Unser, “A pyramid approach to subpixel registration based on intensity,” IEEE Trans. Image Process.,  7(1), 27–41 (1998).
    [Crossref]
  16. J. W. Goodman, “Speckle with a finite number of steps,” Appl. Opt.,  47(4), A111–A118 (2008).
    [Crossref] [PubMed]
  17. I. D. Schizas and G. B. Giannakis, “Covariance eigenvector sparsity for compression and denoising,” IEEE Trans. Signal Process. 60(5), 2408–2421 (2012).
    [Crossref]
  18. J. D. Shutler, M. S. Nixon, and C. J. Harris, “Global statistical description of temporal features,” Int. Arch. Photogramm. Remote Sens. 33(B5/2; PART 5), 720–726 (2000).
  19. J. D. Shutler and M. S. Nixon, “Zernike velocity moments for sequence-based description of moving features,” Image Vis. Comput. 24(4), 343–356 (2006).
    [Crossref]
  20. D. Wen, Y. Jiang, H. Hua, R. Yu, Q. Gao, and Y. Zhang, “Laser speckle reduction based on compressive sensing and edge detection,” Proc. SPIE 8905, 890506 (2013).
    [Crossref]
  21. S. M. Shams Kazmi, E. Faraji, M. A. Davis, Y. Huang, X. J. Zhang, and A. K. Dunn, “Flux or speed? examining speckle contrast imaging of vascular flows,” Biomed. Opt. Express 6(7), 2588–2608 (2015).
    [Crossref]
  22. M. Hu, “Visual pattern recognition by moment invariants,” IRE Trans. Inf. Theory 8(2), 179–187 (1962).
    [Crossref]
  23. J. F. Boyce and W. J. Hossack, “Moment invariants for pattern recognition,” Pattern Recognit. Lett. 1(5–6), 451–456 (1983).
    [Crossref]
  24. D. G. Voelz, Computational fourier optics: a MATLAB tutorial (SPIE Press, 2011), Chap. 5.

2018 (1)

S. S Parikh, D. Ruiz, H. Kalva, G. Fernández-Escribano, and V. Adzic, “High bit-depth medical image compression with hevc,” IEEE J. Biomed. Health Informat. 22(2), 552–560 (2018).
[Crossref]

2017 (1)

2016 (2)

2015 (2)

2014 (1)

L. M. Richards, E. L. Towle, D. J. Fox, and A. K. Dunn, “Intraoperative laser speckle contrast imaging with retrospective motion correction for quantitative assessment of cerebral blood flow,” Neurophotonics 1(1), 015006 (2014).
[Crossref]

2013 (1)

D. Wen, Y. Jiang, H. Hua, R. Yu, Q. Gao, and Y. Zhang, “Laser speckle reduction based on compressive sensing and edge detection,” Proc. SPIE 8905, 890506 (2013).
[Crossref]

2012 (2)

H. Levy, D. Ringuette, and O. Levi, “Rapid monitoring of cerebral ischemia dynamics using laser-based optical imaging of blood oxygenation and flow,” Biomed. Opt. Express,  3(4), 777–791 (2012).
[Crossref] [PubMed]

I. D. Schizas and G. B. Giannakis, “Covariance eigenvector sparsity for compression and denoising,” IEEE Trans. Signal Process. 60(5), 2408–2421 (2012).
[Crossref]

2010 (2)

P. Miao, A. Rege, N. Li, N. V. Thakor, and S. Tong, “High resolution cerebral blood flow imaging by registered laser speckle contrast analysis,” IEEE Trans. Biomed. Eng. 57(5), 1152–1157 (2010).
[Crossref] [PubMed]

S. E. Skipetrov, J. Peuser, R. Cerbino, P. Zakharov, B. Weber, and F. Scheffold, “Noise in laser speckle correlation and imaging techniques,” Opt. Express 18(14), 14519–14534 (2010).
[Crossref] [PubMed]

2008 (2)

2007 (1)

K. Amolins, Y. Zhang, and P. Dare, “Wavelet based image fusion techniques-An introduction, review and comparison,” ISPRS J. Photogramm. Remote Sens. 62(4), 249–263 (2007).
[Crossref]

2006 (1)

J. D. Shutler and M. S. Nixon, “Zernike velocity moments for sequence-based description of moving features,” Image Vis. Comput. 24(4), 343–356 (2006).
[Crossref]

2000 (1)

J. D. Shutler, M. S. Nixon, and C. J. Harris, “Global statistical description of temporal features,” Int. Arch. Photogramm. Remote Sens. 33(B5/2; PART 5), 720–726 (2000).

1999 (1)

D. Shen and H. H. S. Ip, “Discriminative wavelet shape descriptors for recognition of 2-d patterns,” Pattern Recognition 32(2), 151–165 (1999).
[Crossref]

1998 (1)

P. Thevenaz, U. E Ruttimann, and M. Unser, “A pyramid approach to subpixel registration based on intensity,” IEEE Trans. Image Process.,  7(1), 27–41 (1998).
[Crossref]

1988 (1)

C.-H. Teh and R. T. Chin, “On image analysis by the methods of moments,” IEEE Trans. Pattern Anal. Mach. Intell. 10(4), 496–513 (1988).
[Crossref]

1983 (1)

J. F. Boyce and W. J. Hossack, “Moment invariants for pattern recognition,” Pattern Recognit. Lett. 1(5–6), 451–456 (1983).
[Crossref]

1962 (1)

M. Hu, “Visual pattern recognition by moment invariants,” IRE Trans. Inf. Theory 8(2), 179–187 (1962).
[Crossref]

Adzic, V.

S. S Parikh, D. Ruiz, H. Kalva, G. Fernández-Escribano, and V. Adzic, “High bit-depth medical image compression with hevc,” IEEE J. Biomed. Health Informat. 22(2), 552–560 (2018).
[Crossref]

Amolins, K.

K. Amolins, Y. Zhang, and P. Dare, “Wavelet based image fusion techniques-An introduction, review and comparison,” ISPRS J. Photogramm. Remote Sens. 62(4), 249–263 (2007).
[Crossref]

Boyce, J. F.

J. F. Boyce and W. J. Hossack, “Moment invariants for pattern recognition,” Pattern Recognit. Lett. 1(5–6), 451–456 (1983).
[Crossref]

Carlen, P. L.

Cerbino, R.

Chin, R. T.

C.-H. Teh and R. T. Chin, “On image analysis by the methods of moments,” IEEE Trans. Pattern Anal. Mach. Intell. 10(4), 496–513 (1988).
[Crossref]

Choi, B.

R. Farraro, O. Fathi, and B. Choi, “Handheld, point-of-care laser speckle imaging,” J. Biomed. Opt. 21(9), 094001 (2016).
[Crossref]

Dare, P.

K. Amolins, Y. Zhang, and P. Dare, “Wavelet based image fusion techniques-An introduction, review and comparison,” ISPRS J. Photogramm. Remote Sens. 62(4), 249–263 (2007).
[Crossref]

Davis, M. A.

Dufour, S.

Duncan, D. D.

Dunn, A. K.

S. M. Shams Kazmi, E. Faraji, M. A. Davis, Y. Huang, X. J. Zhang, and A. K. Dunn, “Flux or speed? examining speckle contrast imaging of vascular flows,” Biomed. Opt. Express 6(7), 2588–2608 (2015).
[Crossref]

L. M. Richards, E. L. Towle, D. J. Fox, and A. K. Dunn, “Intraoperative laser speckle contrast imaging with retrospective motion correction for quantitative assessment of cerebral blood flow,” Neurophotonics 1(1), 015006 (2014).
[Crossref]

Faraji, E.

Farraro, R.

R. Farraro, O. Fathi, and B. Choi, “Handheld, point-of-care laser speckle imaging,” J. Biomed. Opt. 21(9), 094001 (2016).
[Crossref]

Fathi, O.

R. Farraro, O. Fathi, and B. Choi, “Handheld, point-of-care laser speckle imaging,” J. Biomed. Opt. 21(9), 094001 (2016).
[Crossref]

Fernández-Escribano, G.

S. S Parikh, D. Ruiz, H. Kalva, G. Fernández-Escribano, and V. Adzic, “High bit-depth medical image compression with hevc,” IEEE J. Biomed. Health Informat. 22(2), 552–560 (2018).
[Crossref]

Fox, D. J.

L. M. Richards, E. L. Towle, D. J. Fox, and A. K. Dunn, “Intraoperative laser speckle contrast imaging with retrospective motion correction for quantitative assessment of cerebral blood flow,” Neurophotonics 1(1), 015006 (2014).
[Crossref]

Gad, R.

Gao, Q.

D. Wen, Y. Jiang, H. Hua, R. Yu, Q. Gao, and Y. Zhang, “Laser speckle reduction based on compressive sensing and edge detection,” Proc. SPIE 8905, 890506 (2013).
[Crossref]

Giannakis, G. B.

I. D. Schizas and G. B. Giannakis, “Covariance eigenvector sparsity for compression and denoising,” IEEE Trans. Signal Process. 60(5), 2408–2421 (2012).
[Crossref]

Goodman, J. W.

Harris, C. J.

J. D. Shutler, M. S. Nixon, and C. J. Harris, “Global statistical description of temporal features,” Int. Arch. Photogramm. Remote Sens. 33(B5/2; PART 5), 720–726 (2000).

Hossack, W. J.

J. F. Boyce and W. J. Hossack, “Moment invariants for pattern recognition,” Pattern Recognit. Lett. 1(5–6), 451–456 (1983).
[Crossref]

Hu, M.

M. Hu, “Visual pattern recognition by moment invariants,” IRE Trans. Inf. Theory 8(2), 179–187 (1962).
[Crossref]

Hua, H.

D. Wen, Y. Jiang, H. Hua, R. Yu, Q. Gao, and Y. Zhang, “Laser speckle reduction based on compressive sensing and edge detection,” Proc. SPIE 8905, 890506 (2013).
[Crossref]

Huang, Y.

Ip, H. H. S.

D. Shen and H. H. S. Ip, “Discriminative wavelet shape descriptors for recognition of 2-d patterns,” Pattern Recognition 32(2), 151–165 (1999).
[Crossref]

Jeffrey, M. A.

Jiang, Y.

D. Wen, Y. Jiang, H. Hua, R. Yu, Q. Gao, and Y. Zhang, “Laser speckle reduction based on compressive sensing and edge detection,” Proc. SPIE 8905, 890506 (2013).
[Crossref]

Kalva, H.

S. S Parikh, D. Ruiz, H. Kalva, G. Fernández-Escribano, and V. Adzic, “High bit-depth medical image compression with hevc,” IEEE J. Biomed. Health Informat. 22(2), 552–560 (2018).
[Crossref]

Kirkpatrick, S. J.

Koletar, M. M

Levi, O.

Levy, H.

Li, N.

P. Miao, A. Rege, N. Li, N. V. Thakor, and S. Tong, “High resolution cerebral blood flow imaging by registered laser speckle contrast analysis,” IEEE Trans. Biomed. Eng. 57(5), 1152–1157 (2010).
[Crossref] [PubMed]

Miao, P.

P. Miao, A. Rege, N. Li, N. V. Thakor, and S. Tong, “High resolution cerebral blood flow imaging by registered laser speckle contrast analysis,” IEEE Trans. Biomed. Eng. 57(5), 1152–1157 (2010).
[Crossref] [PubMed]

Nixon, M. S.

J. D. Shutler and M. S. Nixon, “Zernike velocity moments for sequence-based description of moving features,” Image Vis. Comput. 24(4), 343–356 (2006).
[Crossref]

J. D. Shutler, M. S. Nixon, and C. J. Harris, “Global statistical description of temporal features,” Int. Arch. Photogramm. Remote Sens. 33(B5/2; PART 5), 720–726 (2000).

Parikh, S. S

S. S Parikh, D. Ruiz, H. Kalva, G. Fernández-Escribano, and V. Adzic, “High bit-depth medical image compression with hevc,” IEEE J. Biomed. Health Informat. 22(2), 552–560 (2018).
[Crossref]

Peuser, J.

Rege, A.

P. Miao, A. Rege, N. Li, N. V. Thakor, and S. Tong, “High resolution cerebral blood flow imaging by registered laser speckle contrast analysis,” IEEE Trans. Biomed. Eng. 57(5), 1152–1157 (2010).
[Crossref] [PubMed]

Richards, L. M.

L. M. Richards, E. L. Towle, D. J. Fox, and A. K. Dunn, “Intraoperative laser speckle contrast imaging with retrospective motion correction for quantitative assessment of cerebral blood flow,” Neurophotonics 1(1), 015006 (2014).
[Crossref]

Ringuette, D.

Ruiz, D.

S. S Parikh, D. Ruiz, H. Kalva, G. Fernández-Escribano, and V. Adzic, “High bit-depth medical image compression with hevc,” IEEE J. Biomed. Health Informat. 22(2), 552–560 (2018).
[Crossref]

Ruttimann, U. E

P. Thevenaz, U. E Ruttimann, and M. Unser, “A pyramid approach to subpixel registration based on intensity,” IEEE Trans. Image Process.,  7(1), 27–41 (1998).
[Crossref]

Scheffold, F.

Schizas, I. D.

I. D. Schizas and G. B. Giannakis, “Covariance eigenvector sparsity for compression and denoising,” IEEE Trans. Signal Process. 60(5), 2408–2421 (2012).
[Crossref]

Shams Kazmi, S. M.

Shen, D.

D. Shen and H. H. S. Ip, “Discriminative wavelet shape descriptors for recognition of 2-d patterns,” Pattern Recognition 32(2), 151–165 (1999).
[Crossref]

Shutler, J. D.

J. D. Shutler and M. S. Nixon, “Zernike velocity moments for sequence-based description of moving features,” Image Vis. Comput. 24(4), 343–356 (2006).
[Crossref]

J. D. Shutler, M. S. Nixon, and C. J. Harris, “Global statistical description of temporal features,” Int. Arch. Photogramm. Remote Sens. 33(B5/2; PART 5), 720–726 (2000).

Sigal, I.

Skipetrov, S. E.

Stefanovic, B.

Teh, C.-H.

C.-H. Teh and R. T. Chin, “On image analysis by the methods of moments,” IEEE Trans. Pattern Anal. Mach. Intell. 10(4), 496–513 (1988).
[Crossref]

Thakor, N. V.

P. Miao, A. Rege, N. Li, N. V. Thakor, and S. Tong, “High resolution cerebral blood flow imaging by registered laser speckle contrast analysis,” IEEE Trans. Biomed. Eng. 57(5), 1152–1157 (2010).
[Crossref] [PubMed]

Thevenaz, P.

P. Thevenaz, U. E Ruttimann, and M. Unser, “A pyramid approach to subpixel registration based on intensity,” IEEE Trans. Image Process.,  7(1), 27–41 (1998).
[Crossref]

Tong, S.

P. Miao, A. Rege, N. Li, N. V. Thakor, and S. Tong, “High resolution cerebral blood flow imaging by registered laser speckle contrast analysis,” IEEE Trans. Biomed. Eng. 57(5), 1152–1157 (2010).
[Crossref] [PubMed]

Towle, E. L.

L. M. Richards, E. L. Towle, D. J. Fox, and A. K. Dunn, “Intraoperative laser speckle contrast imaging with retrospective motion correction for quantitative assessment of cerebral blood flow,” Neurophotonics 1(1), 015006 (2014).
[Crossref]

Unser, M.

P. Thevenaz, U. E Ruttimann, and M. Unser, “A pyramid approach to subpixel registration based on intensity,” IEEE Trans. Image Process.,  7(1), 27–41 (1998).
[Crossref]

Voelz, D. G.

D. G. Voelz, Computational fourier optics: a MATLAB tutorial (SPIE Press, 2011), Chap. 5.

Weber, B.

Wells-Gray, E. M.

Wen, D.

D. Wen, Y. Jiang, H. Hua, R. Yu, Q. Gao, and Y. Zhang, “Laser speckle reduction based on compressive sensing and edge detection,” Proc. SPIE 8905, 890506 (2013).
[Crossref]

Yu, R.

D. Wen, Y. Jiang, H. Hua, R. Yu, Q. Gao, and Y. Zhang, “Laser speckle reduction based on compressive sensing and edge detection,” Proc. SPIE 8905, 890506 (2013).
[Crossref]

Zakharov, P.

Zhang, X. J.

Zhang, Y.

D. Wen, Y. Jiang, H. Hua, R. Yu, Q. Gao, and Y. Zhang, “Laser speckle reduction based on compressive sensing and edge detection,” Proc. SPIE 8905, 890506 (2013).
[Crossref]

K. Amolins, Y. Zhang, and P. Dare, “Wavelet based image fusion techniques-An introduction, review and comparison,” ISPRS J. Photogramm. Remote Sens. 62(4), 249–263 (2007).
[Crossref]

Appl. Opt. (1)

Biomed. Opt. Express (5)

IEEE J. Biomed. Health Informat. (1)

S. S Parikh, D. Ruiz, H. Kalva, G. Fernández-Escribano, and V. Adzic, “High bit-depth medical image compression with hevc,” IEEE J. Biomed. Health Informat. 22(2), 552–560 (2018).
[Crossref]

IEEE Trans. Biomed. Eng. (1)

P. Miao, A. Rege, N. Li, N. V. Thakor, and S. Tong, “High resolution cerebral blood flow imaging by registered laser speckle contrast analysis,” IEEE Trans. Biomed. Eng. 57(5), 1152–1157 (2010).
[Crossref] [PubMed]

IEEE Trans. Image Process. (1)

P. Thevenaz, U. E Ruttimann, and M. Unser, “A pyramid approach to subpixel registration based on intensity,” IEEE Trans. Image Process.,  7(1), 27–41 (1998).
[Crossref]

IEEE Trans. Pattern Anal. Mach. Intell. (1)

C.-H. Teh and R. T. Chin, “On image analysis by the methods of moments,” IEEE Trans. Pattern Anal. Mach. Intell. 10(4), 496–513 (1988).
[Crossref]

IEEE Trans. Signal Process. (1)

I. D. Schizas and G. B. Giannakis, “Covariance eigenvector sparsity for compression and denoising,” IEEE Trans. Signal Process. 60(5), 2408–2421 (2012).
[Crossref]

Image Vis. Comput. (1)

J. D. Shutler and M. S. Nixon, “Zernike velocity moments for sequence-based description of moving features,” Image Vis. Comput. 24(4), 343–356 (2006).
[Crossref]

Int. Arch. Photogramm. Remote Sens. (1)

J. D. Shutler, M. S. Nixon, and C. J. Harris, “Global statistical description of temporal features,” Int. Arch. Photogramm. Remote Sens. 33(B5/2; PART 5), 720–726 (2000).

IRE Trans. Inf. Theory (1)

M. Hu, “Visual pattern recognition by moment invariants,” IRE Trans. Inf. Theory 8(2), 179–187 (1962).
[Crossref]

ISPRS J. Photogramm. Remote Sens. (1)

K. Amolins, Y. Zhang, and P. Dare, “Wavelet based image fusion techniques-An introduction, review and comparison,” ISPRS J. Photogramm. Remote Sens. 62(4), 249–263 (2007).
[Crossref]

J. Biomed. Opt. (1)

R. Farraro, O. Fathi, and B. Choi, “Handheld, point-of-care laser speckle imaging,” J. Biomed. Opt. 21(9), 094001 (2016).
[Crossref]

Neurophotonics (1)

L. M. Richards, E. L. Towle, D. J. Fox, and A. K. Dunn, “Intraoperative laser speckle contrast imaging with retrospective motion correction for quantitative assessment of cerebral blood flow,” Neurophotonics 1(1), 015006 (2014).
[Crossref]

Opt. Express (1)

Opt. Lett. (1)

Pattern Recognit. Lett. (1)

J. F. Boyce and W. J. Hossack, “Moment invariants for pattern recognition,” Pattern Recognit. Lett. 1(5–6), 451–456 (1983).
[Crossref]

Pattern Recognition (1)

D. Shen and H. H. S. Ip, “Discriminative wavelet shape descriptors for recognition of 2-d patterns,” Pattern Recognition 32(2), 151–165 (1999).
[Crossref]

Proc. SPIE (1)

D. Wen, Y. Jiang, H. Hua, R. Yu, Q. Gao, and Y. Zhang, “Laser speckle reduction based on compressive sensing and edge detection,” Proc. SPIE 8905, 890506 (2013).
[Crossref]

Other (2)

D. G. Voelz, Computational fourier optics: a MATLAB tutorial (SPIE Press, 2011), Chap. 5.

D. Coleman, “TIFF compression options: ZIP vs LZW” (TIFF compression options, 2018). https://havecamerawilltravel.com/photographer/tiff-image-compression/ .

Supplementary Material (1)

NameDescription
» Visualization 1       Laser speckle flow index (SFI) maps and low coherence near-infrared (NIR) reflectance images. The SFI maps have distinct vascular features obscured by small-scale rapid variation. Conversely, NIR reflectance images have little tissue/vessel contras

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 (10)

Fig. 1:
Fig. 1: Stable edge detection properties of BSMFs. (a) Single-frame LSCI blood flow map. (b) Filters based on extrema of 2nd-order moments i) σ min 2 and ii) σ max 2 applied to a. (c) The BSMFs i) ξ3 and ii) ξ4 applied to a. (d) ξ4 as in c for i) smaller and ii) larger filter windows. Filter window sizes in b–d indicated by black circle. (e–h) Cross-section profiles from a–d, respectively, for dashed-line indicated in a. The red curve in e corresponds to vertical pixel averaging within white box in a. Higher-order BSMFs ξ5 and ξ6 also shown in g.
Fig. 2:
Fig. 2: Spatial precision assessment of filters via cross-correlation based estimation. (a) Comparison of inferred displacement from ξ3 and ξ5 filters against inference using blood flow maps. Actual displacement image is approximately 15 pixels. (b) Comparison of ξ4 and ξ6 filters for the same series in a. (c) Comparison of all ξγ for simulated imaging of an optical phase varying cross flow shape displaced 10 pixels at frame 20. Computed flow map and associated ξ3 filter shown above graph. (d) Same as c for optical phase varying square.
Fig. 3:
Fig. 3: Improved sub-pixel registration of blood flow maps through BSMF guidance. (a) A high resolution blood flow map produced by temporally averaging maps occurring between breathing related motions (generated from fifteen 0.8 s stable epochs within a 16 s epoch). All graphs b–d, f, & g share a common correspondence with respect to registration strategy: stable reference (blue), ξ3 filter directed (red), ξ4 filter-directed (green), no registration (orange), and self-directed (black). (b) Vessel-edge cross-sections for motion artifact associated blood flow maps. The paired (same color) curves represent the cross-section distributions (μ ± σ) for fourteen blood flow maps each associated with a 0.3 s breath-related motion within the same 16 s epoch. (c) Artery adjacent region blood flow time trace distributions associated with the same motion artifacts in b (each registration was initiated from the indicated time zero). (d) Correlation loss of single-frame blood flow maps with reference map a. Distributions correspond to repeat series registrations each initialed at 1 of 5 sequential motion artifacts. (e) Correction for motion artifact due to a 34 s unintentional lateral stage drift. (f) Cross-sections corresponding to region indicated in e. (g) Correction for 11 s lateral motion artifact due to stroke induction. (h) Cross-sections for central vessel in g.
Fig. 4:
Fig. 4: Effect of filter window shape, orientation searching and non-linear value re-scale. (a) Map of square window ξ4 filter based on only vertical univariate projections. Extension of filter in a to include (b) eight univariate orientations, and (c) a circular filter window. (d & e) Logarithmic value rescaling prior application of ξ4 in b & c, respectively. (e) Same as c with prior logarithmic rescaling. (f) Same as c with prior exponential rescaling.
Fig. 5:
Fig. 5: Distribution of blood flow index values based on kernel size kn. (a) Distribution for first four odd kernels n = 3, 5, 7, &9. (b) Distribution shift associated with ictal event for n = 3&9. (c & d) Distributions after temporal averaging of frames in a & b, respectively.
Fig. 6:
Fig. 6: Spatial reconstruction and fast temporal dynamic preservation in the presence of sparsity enforcement. (a) Static epoch blood flow maps (25 frame average) with reduced pixel retention over five orders of magnitude (factors of 10−1/5). (b) The ξ3 filter applied to single-frame “sparse” blood flow maps from a (25 frame average). All maps in a and b were spatially low-pass filtered at the kernel size, k5, for visual clarity. (c) Temporal flow profile of artery and vein for 25 frame static epoch during four cardiac cycle events (occurring at 5 Hz) for i) 10−1.6, ii) 10−2.6, iii) 10−3.6 and iv) 10−4.6, pixel retention (dashed curves represent no sparsity condition). (d) Correlation, R2, of sparse and dense temporal profiles from c across all retention levels shown in a for both i) the k5 results in a and ii) for k3 (dashed curves are [(R′ ± σR′)]2). (e) Sum of spatial Fourier coefficients (features > 6 pixels) for both k3 and k5 and their corresponding ξ3 filters (all curves are μ ± σ).
Fig. 7:
Fig. 7: Measuring vasodilation associated with seizure from sparse blood flow maps. (a–c) Superimposed pre-ictal (green) and ictal (red) maps of i) blood flow, ii) sparse blood flow, and iii) ξ3 of single-capture sparse blood flow maps. All maps were produced from temporal average within single static epoch of 0.8–0.9 s. (d) Cross-sections for region in a (mean projection along the short axis of the indicated rectangle). (e) Cross-sections across sparsity for regions in a–c. All cross-section d–e regions match the ξ3 kernel width to account for filter-based averaging. (f) Relative vessel diameter (pre-ictal no sparsity normalized to one) based on the estimated edge locations across sparsity from the cross sections in e. The edge features based on ξ3 are contrasted with edge location estimation based on the FWHM. Paired same-color curves correspond to μ ± σ for 5 sequential static epochs. (g) The pre-ictal versus ictal difference in relative vessel diameters for example a including diameter estimation based on profile standard deviation.
Fig. 8:
Fig. 8: Edge detection of features within a larger FOV. (a) Photothrombotic stroke induced through fluorescent excitation of rose bengal. i) Stroke (green) and pre-stroke (red) blood flow maps. ii) Corresponding temporally averaged 1% sparsity-enforced single-capture applied ξ3 filter maps. (b) Vasodilation associated with absence-like seizure imaged with miniature imaging device. Panels i & ii same as a but for pre-ictal (green) and ictal (red). Panel iii is application of smaller ξ3 for sparsity enforcement on region neglecting largest vessels. (c) Global stroke through carotid artery obstruction. i) Blood flow maps of pre-stroke (green), stroke (red), and ischemia-induced spreading depression (blue). ii) Corresponding temporally averaged 1% sparsity-enforced single-capture applied ξ4 filter maps. (d) Cross section for middle vessel from c (yellow box). (e) Miniature device imaging of concurrent vessel dilation and constriction during animal waking; anesthetized (green) and awake (red).
Fig. 9:
Fig. 9: Temporal dynamics associated large blood flow changes are preserved in sparse values. The dependency of full-field blood flow temporal dynamics on sparsity during a (a) seizure-like event and (b) ischemic stroke. In both instances, level of sparsity set by threshold calibrated to baseline.
Fig. 10:
Fig. 10: Rolling sum algorithms and direction searching. (a) Rolling sum algorithm for square sliding window. Sub-columns of window height are produced for all terms Σαβ in the first row. The first window is initialized by adding sub-columns and the window is moved horizontally by adding and removing sub-columns. The sub-columns are moved vertically by adding and subtracting pixels. Every pixel and sub-column term is added once and subtracted, at most, once. (b) Rolling sum algorithm for a circular sliding window. A circular window is first initialized and proceeds though successive arc additions and subtractions to trace the entire image. Every pixel is added and subtracted once times the circle diameter. (c) Exhaustive search across directions using centered expectation values.

Tables (4)

Tables Icon

Table 1: The SNR associated with LSCI blood flow maps and associated BSMFs assessed during static epochs, for the full FOV in Fig. 1 and a 100 × 100 μm arteriole-centered ROI.

Tables Icon

Table 2: The spatial location precision of LSCI blood flow maps and associated BSMFs assessed with cross-correlation during static epochs. Units are pixels (scale 0.8 μm/pixel).

Tables Icon

Table 3: Accuracy of registration approaches. Corresponding to Fig. 3: (b) The variation of edge profiles during movement artifact. (c) The variation of time traces before and after registration of recurrent artifact. (d) Rate of correlation loss with respect to static reference.

Tables Icon

Table 4: The absolute error (normalized RMSE) and dynamical error (1 − R2) of sparse reconstructed temporal flow traces against dense flow trace. Note: the paired RMSE values are baseline and peak seizure/stroke flow increase/decrease in Fig 9, respectively.

Equations (7)

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

U ( x , y ; z ) = FFT 1 { FFT { e i k z ( i λ z ) 1 e i k 2 z ( x 2 + y 2 ) } ( Δ x ) 2 FFT { U ( x , y ; 0 ) } }
( x x α ) = i = 0 n w i ( x i x ) α i = 0 n w i , where x = i = 0 n w i x i i = 0 n w i
Σ α β = i = 0 n w i x i α y i β
w γ = i = 0 γ ( γ i ) a γ i b i u γ i v i
u 2 = Σ 20 Σ 00 1 x 2 , u v = Σ 11 Σ 00 1 x y , v 2 = Σ 02 Σ 00 1 y 2
u 3 = Σ 30 Σ 00 1 3 x u 2 x 3 , u 2 v = c 21 Σ 00 1 x u v y u 2 v 3 = Σ 03 Σ 00 1 3 y v 2 y 3 , v 2 u = c 21 Σ 00 1 y u v x v 2
u 4 = Σ 40 Σ 00 1 4 x u 3 6 x 2 u 2 x 4 , u 3 v = c 31 Σ 00 1 y ( x 3 + u 3 ) v 4 = Σ 04 Σ 00 1 4 y v 3 6 y 2 v 2 y 4 , v 3 u = c 13 Σ 00 1 x ( y 3 + v 3 ) u 2 v 2 = ( Σ 22 2 y c 21 2 x c 12 ) Σ 00 1 + x 2 v 2 + y 2 ( u 2 x 2 )

Metrics