OSA's Digital Library

Virtual Journal for Biomedical Optics

Virtual Journal for Biomedical Optics

| EXPLORING THE INTERFACE OF LIGHT AND BIOMEDICINE

  • Editors: Andrew Dunn and Anthony Durkin
  • Vol. 8, Iss. 7 — Aug. 1, 2013
« Show journal navigation

Phase registration based on matching of phase distribution characteristics and its application in FDOCT

Guozhong Liu, Zhehai Zhou, and Ping Li  »View Author Affiliations


Optics Express, Vol. 21, Issue 11, pp. 13241-13255 (2013)
http://dx.doi.org/10.1364/OE.21.013241


View Full Text Article

Acrobat PDF (4174 KB)





Browse Journals / Lookup Meetings

Browse by Journal and Year


   


Lookup Conference Papers

Close Browse Journals / Lookup Meetings

Article Tools

Share
Citations

Abstract

Phase fluctuations in a two-transverse-dimensional scanning Fourier domain optical coherence tomography (FDOCT) seriously affect in vivo phase related applications. The phase difference between two A-scans sampled at the same scanning position or adjacent scanning position is acquired by matching of the phase distribution characteristics on the surface of two A-scans. Finger and palm scanning experiments are performed and defocused images of finger and palm are recovered based on Fresnel scalar diffraction algorithm by using phase compensated OCT complex signals. To further prove the performance of the proposed method, human eye scanning experiments are also performed and blood flow images of retina are extracted from the phase registration results. The accurate, fast and simple phase compensation method is critical for in vivo phase related applications.

© 2013 OSA

1. Introduction

Fourier domain optical coherence tomography (FDOCT) [1

1. A. F. Fercher, C. K. Hitzenberger, G. Kamp, and S. Y. El-Zaiat, “Measurement of intraocular distances by backscattering spectral interferometry,” Opt. Commun. 117(1–2), 43–48 (1995). [CrossRef]

,2

2. M. Wojtkowski, R. Leitgeb, A. Kowalczyk, T. Bajraszewski, and A. F. Fercher, “In vivo human retinal imaging by Fourier domain optical coherence tomography,” J. Biomed. Opt. 7(3), 457–463 (2002). [CrossRef] [PubMed]

] has recently become preferred to the original time domain coherence tomography (TDOCT) [3

3. D. Huang, E. A. Swanson, C. P. Lin, J. S. Schuman, W. G. Stinson, W. Chang, M. R. Hee, T. Flotte, K. Gregory, C. A. Puliafito, and J. G. Fujimoto, “Optical coherence tomography,” Science 254(5035), 1178–1181 (1991). [CrossRef] [PubMed]

] due to its sensitivity and high speed advantages and has facilitated the development of real-time in vivo tissue imaging [4

4. N. A. Nassif, B. Cense, B. H. Park, M. C. Pierce, S. H. Yun, B. E. Bouma, G. J. Tearney, T. C. Chen, and J. F. de Boer, “In vivo high-resolution video-rate spectral-domain optical coherence tomography of the human retina and optic nerve,” Opt. Express 12(3), 367–376 (2004). [CrossRef] [PubMed]

] and also three-dimensional volumetric imaging [5

5. R. K. Wang and Z. Ma, “A practical approach to eliminate autocorrelation artefacts for volume-rate spectral domain optical coherence tomography,” Phys. Med. Biol. 51(12), 3231–3239 (2006). [CrossRef] [PubMed]

]. Phase stability is a primary requirement to many OCT studies that depend on the measurement of complex signals. For instance, phase stable measurements are required for Doppler OCT [6

6. Y. Jia, P. O. Bagnaninchi, Y. Yang, A. E. Haj, M. T. Hinds, S. J. Kirkpatrick, and R. K. Wang, “Doppler optical coherence tomography imaging of local fluid flow and shear stress within microporous scaffolds,” J. Biomed. Opt. 14(3), 034014 (2009). [CrossRef] [PubMed]

], phase microscopy [7

7. M. A. Choma, A. K. Ellerbee, C. Yang, T. L. Creazzo, and J. A. Izatt, “Spectral-domain phase microscopy,” Opt. Lett. 30(10), 1162–1164 (2005). [CrossRef] [PubMed]

,8

8. C. Joo, T. Akkin, B. Cense, B. H. Park, and J. F. de Boer, “Spectral-domain optical coherence phase microscopy for quantitative phase-contrast imaging,” Opt. Lett. 30(16), 2131–2133 (2005). [CrossRef] [PubMed]

], polarization sensitive OCT [9

9. J. F. de Boer, T. E. Milner, M. J. C. van Gemert, and J. S. Nelson, “Two-dimensional birefringence imaging in biological tissue by polarization-sensitive optical coherence tomography,” Opt. Lett. 22(12), 934–936 (1997). [CrossRef] [PubMed]

], coherent averaging [10

10. P. H. Tomlins and R. K. Wang, “Digital phase stabilization to improve detection sensitivity for optical coherence tomography,” Meas. Sci. Technol. 18(11), 3365–3372 (2007). [CrossRef]

], spectroscopic OCT [11

11. D. Adler, T. Ko, P. Herz, and J. Fujimoto, “Optical coherence tomography contrast enhancement using spectroscopic analysis with spectral autocorrelation,” Opt. Express 12(22), 5487–5501 (2004). [CrossRef] [PubMed]

,12

12. C. Xu, C. Vinegoni, T. S. Ralston, W. Luo, W. Tan, and S. A. Boppart, “Spectroscopic spectral-domain optical coherence microscopy,” Opt. Lett. 31(8), 1079–1081 (2006). [CrossRef] [PubMed]

], and defocused image recovery [13

13. T. S. Ralston, D. L. Marks, P. Scott Carney, and S. A. Boppart, “Interferometric synthetic aperture microscopy,” Nat. Phys. 3(2), 129–134 (2007). [CrossRef]

17

17. G. Liu, S. Yousefi, Z. Zhi, and R. K. Wang, “Automatic estimation of point-spread-function for deconvoluting out-of-focus optical coherence tomographic images using information entropy-based approach,” Opt. Express 19(19), 18135–18148 (2011). [CrossRef] [PubMed]

]. The inherent phase fluctuations of FDOCT systems are caused by thermal drifts, galvanometer positioning accuracy and system mechanical jitter, which is much lower than the phase fluctuations in TDOCT systems because the reference arm in the FDOCT system is fixed. Another phase error appearance is the bulk motion artifacts, from sample motion and environmental vibrations, which are an order of magnitude larger than those associated with the inherent phase instability of the system and can obscure true motion in the sample if uncorrected [18

18. B. White, M. Pierce, N. Nassif, B. Cense, B. Park, G. Tearney, B. Bouma, T. Chen, and J. F. de Boer, “In vivo dynamic human retinal blood flow imaging using ultra-high-speed spectral domain optical coherence tomography,” Opt. Express 11(25), 3490–3497 (2003). [CrossRef] [PubMed]

]. The effect of environmental vibrations on the phase-resolved techniques can be eliminated by the using of the vibration isolation platform. However, for in vivo imaging applications, it is difficult to maintain phase stability over a whole two-transverse-dimensional scanning because of the phase fluctuations caused by the respiratory motion, cardiac motion and the involuntary movement of the subject. There physiological fluctuations are turned out to be much bigger than other fluctuations [19

19. J. Lee, V. Srinivasan, H. Radhakrishnan, and D. A. Boas, “Motion correction for phase-resolved dynamic optical coherence tomography imaging of rodent cerebral cortex,” Opt. Express 19(22), 21258–21270 (2011). [CrossRef] [PubMed]

]. Therefore, the phase fluctuations caused by the sample motion is the most significant one of them all and contributes more as a noise, and this has been testified by our experiment results that the en face out-of-focus images of static sample (such as fresh onion) can be digitally focused without phase compensation, however the out-of-focus images of in vivo tissues (such as human finger, palm) acquired from the same FDOCT system cannot be digitally focused if phase compensation is not performed.

A number of methods were investigated to overcome the phase stable problem. In general, phase fluctuations can be mitigated by development of ultrahigh speed swept laser sources, higher speed acquisition, vibration isolation, and minimizing fiber lengths to reduce thermal drifts. Besides, free-space or common path design [7

7. M. A. Choma, A. K. Ellerbee, C. Yang, T. L. Creazzo, and J. A. Izatt, “Spectral-domain phase microscopy,” Opt. Lett. 30(10), 1162–1164 (2005). [CrossRef] [PubMed]

,8

8. C. Joo, T. Akkin, B. Cense, B. H. Park, and J. F. de Boer, “Spectral-domain optical coherence phase microscopy for quantitative phase-contrast imaging,” Opt. Lett. 30(16), 2131–2133 (2005). [CrossRef] [PubMed]

,18

18. B. White, M. Pierce, N. Nassif, B. Cense, B. Park, G. Tearney, B. Bouma, T. Chen, and J. F. de Boer, “In vivo dynamic human retinal blood flow imaging using ultra-high-speed spectral domain optical coherence tomography,” Opt. Express 11(25), 3490–3497 (2003). [CrossRef] [PubMed]

] or a feedback control loop in conjunction with a fiber stretcher or piezo-electric modulator [21

21. D. Lin, X. Jiang, F. Xie, W. Zhang, L. Zhang, and I. Bennion, “High stability multiplexed fiber interferometer and its application on absolute displacement measurement and on-line surface metrology,” Opt. Express 12(23), 5729–5734 (2004). [CrossRef] [PubMed]

] are adopted to compensate for phase fluctuations. To eliminate the interferometric noise, Yaqoob et al. [22

22. Z. Yaqoob, W. Choi, S. Oh, N. Lue, Y. Park, C. Fang-Yen, R. R. Dasari, K. Badizadegan, and M. S. Feld, “Improved phase sensitivity in spectral domain phase microscopy using line-field illumination and self phase-referencing,” Opt. Express 17(13), 10681–10687 (2009). [CrossRef] [PubMed]

] used the line illumination beam and taken the phase measured at a part of the beam illuminating outside of the sample as a reference phase to remove the common-mode noise. However, these sophisticated and expensive hardware compensators may only be moved with limited speed and accuracy.

Post-processing techniques were also developed to compensate for phase fluctuations through alignment of the complex A-scans. Tomlins et al. [10

10. P. H. Tomlins and R. K. Wang, “Digital phase stabilization to improve detection sensitivity for optical coherence tomography,” Meas. Sci. Technol. 18(11), 3365–3372 (2007). [CrossRef]

] calculated cross-correlation of the magnitudes of two A-scans and located its maximum peak corresponding to the offset between the two signals by Lorentzian function fitting method to identify the small path-length difference between the two scans. Then, each of the A-scans in a data sets were digitally stabilized by multiplying the spectrogram by a complex exponential, and spectrally averaged in order to enhance detection sensitivity. Because the phase of A-scan is much more sensitive to the change of optical path-length than the magnitude of A-scan, the magnitude cross-correlation method is not suitable for accurately detecting phase fluctuations. Yang et al. [23

23. C. Yang, A. Wax, M. S. Hahn, K. Badizadegan, R. R. Dasari, and M. S. Feld, “Phase-referenced interferometer with subwavelength and subhertz sensitivity applied to the study of cell membrane dynamics,” Opt. Lett. 26(16), 1271–1273 (2001). [CrossRef] [PubMed]

] prepared the sample on a covers lip and referenced the heterodyne phase of the light reflected from the sample to that of the light reflected from the cover slip, as a result, the phase fluctuations between reference and sample arms was mitigated, displacement and velocity sensitivities of 3.6 nm and 1 nm/s were achieved, enabling the observation of cell membrane dynamics. Ralston et al. [24

24. T. S. Ralston, D. L. Marks, P. S. Carney, and S. A. Boppart, “Phase stability technique for inverse scattering in optical coherence tomography,” 3rd IEEE International Symposium on Biomedical Imaging: Nano to Macro, 578–581 (2006). [CrossRef]

] and Yu et al. [14

14. L. Yu, B. Rao, J. Zhang, J. Su, Q. Wang, S. Guo, and Z. Chen, “Improved lateral resolution in optical coherence tomography by digital focusing using two-dimensional numerical diffraction method,” Opt. Express 15(12), 7634–7641 (2007). [CrossRef] [PubMed]

] placed a covers lip above the specimen, located the reference object, and calculated phase and group delay values for each A-scan to compensate for the differential variations in optical path length. They used phase compensated results to ISAM (interferometric synthetic aperture microscopy) and angular spectrum algorithm respectively for recovery of out-of-focus OCT images. However, in most cases, using a phase reference cover slip is impractical or undesirable and need to correct residual phase fluctuations due to motion between the covers lip and the sample. To minimize the motion artifacts in optical micro-angiography (OMAG) flow image caused by the inevitable subject movement, providing volumetric vasculatural images in retina and choroids, An et al. [25

25. L. An and R. K. Wang, “In vivo volumetric imaging of vascular perfusion within human retina and choroids with optical micro-angiography,” Opt. Express 16(15), 11438–11452 (2008). [CrossRef] [PubMed]

] tried to compensate the bulk tissue motion by histogram analysing or averaging all the phase differences along A-scan signals. Although the histogram analysing method is more accurate than the averaging method, it is computationally inefficient. The common disadvantage of these two approaches is that the noise phase in A-scans has great effect on the measurement accuracy of phase differences.

In this paper, the phase registration principle and results based on matching of phase distribution characteristics is introduced firstly. Then, the digitally focused images of human finger and palm by using the phase compensated OCT complex signals are demonstrated, and the blood flow images of retina extracted from the phase registration results are provided. Finally, the influence of parameters on phase compensation method and basic rules to determine parameters are described.

2. Phase registration principle and results

2.1 FDOCT complex signal

2.2 Phase registration principle

Typical 3D volume data of FDOCT is illustrated as Fig. 1
Fig. 1 OCT 3D volume data consisting of sampling points, A-scans and B-scans.
. Each A-scan is digitalized to j sampling points (P1-Pj) in z direction. Each fast scanning in x direction yields a B-scan including n A-scans(A1-An). Slow scanning in y direction yields m B-scans(B1-Bm). Generally speaking, it is reasonable to assume that phases of j sampling points (P1-Pj) in a single A-scan is static because of the high speed of FDOCT. However, it is difficult to maintain phase stability over a whole two transverse dimensional scan because of the phase fluctuations caused by the movement of the sample. It is necessary to detect phase fluctuations between two A-scans sampled at the same or adjacent lateral scanning position, or reference phases of all A-scans in a volume data set to the phase of a single A-scan in the data set for phase information applications.

Adjacent A-scans are considered having the same structure and amplitude distribution and phase distribution along the depth when the horizontal distance between two scanning point is small enough. A small change Δzof path length difference between the sample arm and the reference arm yields translation of sample complex signal as Eq. (6).

F(z)=A(zΔz)ejφ(zΔz)
(6)

Therefore, phase fluctuation (φ(z)φ(zΔz)) between adjacent A-scans can be acquired by comparing phase distribution characteristics of adjacent A-scans.

Figure 2(a)
Fig. 2 Phase comparing of two adjacent A-scans. (a) B-scan image (512*512) of a tomato. (b) Partially magnified image of Fig. (a). (c) Phase curve of A-scan A263from z = 125 to z = 145. (d) Phase curve of A-scan A264from z = 125 to z = 145 and (e) Phase different between A-scan A264and A263.
displays a typical B-scan amplitude image of tomato. Figures 2(c) and 2(d) represents phase distribution curve of A-scanA263and A-scanA264respectively from z = 125 to z = 145, and Fig. 2(e) is the phase difference between A-scanA264 and A-scanA263in Fig. 2(a). It is obvious that the phase difference between these two adjacent A-scans is stable in the area where generate strong scattering signal (we named it as real scattering signal), however, phase difference is irregular in the area where scattering signal is weak or only has noise (we named it as noise signal). So, to distinguish real scattering signal from noise signal is critical for calculating the phase difference between adjacent A-scans accurately.

Phase distribution characteristics is defined as a vector consisting of a sequence of phase differences between adjacent pixels
PDi,M(I)=[PDi,M(I)(1),,PDi,M(I)(j),,PDi,M(I)(M)]=[φi+1(I)φi(I),,φi+j(I)φi+j(I),,φi+M(I)φi+M1(I)],j=1,2,,M
(7)
where I denotes A-scan index, i is a pixel index in A-scan data, j is the element index of vector PDi,M(I) and M is a number of phase difference between successive A-scan pixels used to compute i-th pixel phase distribution.

Sample scattering signals of two adjacent A-scans have similar phase distribution characteristics since they are scattered from similar internal structure. The judging function of the sample real scattering signal in I-th A-scan and (I-1)-th A-scan is defined as
PDerror(I)=j=1M|PDi,M(I)(j)PDi,M(I1)(j)|<ε
(8)
where ε is the threshold value. If PDerror(I)<ε, points i,i+1,,i+M in I-th A-scan and (I-1)-th A-scan are regarded as real strong scattering points. Here, only strong scattering points on the sample surface are searched for calculating the phase difference between I-th and (I-1)-th adjacent A-scans, and the phase difference is acquired by
Δφ=k=1N(φk(I)φk(I1))N
(9)
where N is the number of found consecutive real strong scattering points on the sample surface.

2.3 Phase registration results

Figures 3(a)
Fig. 3 Phase differences between adjacent A-scans of first B-scan image of (a) human finger and (b) human palm.
and 3(b) show the phase differences between adjacent A-scans of first B-scan image of human finger and palm, respectively. From the results we can see that the phase fluctuations within one B-scan are small for both finger and palm because A-scan rate is much faster than the moving speed of the sample.

Figures 4(a)
Fig. 4 Phase differences between adjacent A-scans of adjacent B-scans of (a)-(d) human finger and (e)-(h) human palm.
- 4(d) show the phase differences between adjacent A-scans of adjacent B-scans B20and B21, B21and B22, B22andB23,B23 and B24 of human finger, respectively. Figures 4(e)- 4(h) show the phase differences between adjacent A-scans of adjacent B-scans B20and B21, B21and B22, B22andB23,B23 and B24 of human palm, respectively. By comparing Fig. 3 with Fig. 4, we can find that the phase fluctuations between adjacent B-scans are much larger than the phase fluctuations within one B-scan. By comparing Figs. 4(a) - 4(d) with Figs. 4(e) - 4(h), we can find that the signal phase stability of the palm is poorer than that of the finger because it is difficult to keep the palm still.

Figures 5(a)
Fig. 5 The phase differences of all A-scans to the phase of the first A-scan of the first B-scan. (a) and (b) are phase differences expressed with positive angles (phase lead) of human finger and human palm, respectively. (c) is the enlargement of the selected part A in Fig. (a). (d) is the enlargement of the selected part B in Fig. (b). (e) and (f) are phase differences expressed with cyclic angle i.e. ±π of human finger and human palm, respectively. (g) and (h) are partially enlargement of Figs. (e) and (f), respectively.
and 5(e) show the phase differences of all A-scans to the phase of the first A-scan of the first B-scan of human finger expressed with positive angles (phase lead) and cyclic angle i.e. ±π, respectively, Figs. 5(c) and 5(g) denote partially enlargement of Figs. 5(a) and Fig. 5(e), respectively. Figures 5(b), 5(d), 5(f) and 5(h) show the results of human palm. By comparison, the phase fluctuations of a human palm is much bigger than that of finger. Compensation phase functions in Figs. 5(e) and 5(f) can be expressed as Phasecompensation(x,y)and used to phase compensation of OCT complex signals of finger and palm, respectively for defocused image recovery.

3. Application of phase registration in FDOCT

In order to show the important role of phase registration in FDOCT in vivo applications, two OCT systems were used to scan human finger and palm, and human retina, respectively, and structure image of finger and palm and blood flow image of retina acquired by using phase compensated results are illustrated below.

3.1 Application of phase registration in digital focusing of FDOCT images

When the OCT system scans a tissue, the probe beam is typically focused into the sample by an objective lens. The lateral resolution of the OCT system is inversely proportional to the numerical aperture (NA) of the objective lens, therefore, it can be enhanced by the increase of numerical aperture. However, only the OCT image that falls within the depth of field (DOF) exhibits the desired lateral resolution, whereas the OCT image that falls outside the DOF region is blurred laterally and the increase of the numerical aperture leads to the reduction of DOF. In general, the higher lateral resolution and the longer DOF are two of the most wishful parameters for the most OCT imaging applications; however, they are reciprocally coupled. The relationship between the lateral resolution and DOF is schematically illustrated in Fig. 6(a)
Fig. 6 Schematic illustration on the effect of numerical aperture on the desired lateral resolution and DOF of OCT images, in the case of (a) low NA and (b) high NA.
for low and Fig. 6(b) for high NAs.

To improve the lateral resolution by using digitally focusing method, phase fluctuations of the original FDOCT complex signals need to be compensated using the following equation
UCompensated(x,y)=UOriginal(x,y)ejPhasecompensation(x,y)
(10)
where Phasecompensation(x,y) is the compensation phase function acquired in Subsection 2.3, UOriginal(x,y) is the original en face complex signal, and UCompensated(x,y)is the compensated complex signal.

UFocused(x,y)=12πexp(jkz)FT1{FT[UDefocused(x,y)]×exp[jz2k(kx2+ky2)]}
(11)

Here, UDefocused(x,y) is the defocused en face FDOCT complex signal outside the DOF region, UFocused(x,y) is the focused complex signal, k is the wave number, z is the diffraction distance, kx,ky are spatial frequencies, and FT denotes the Fourier transform.

Figure 7
Fig. 7 (a)The defocused finger image and (b) recovered image using phase compensated complex signal based on scalar diffraction algorithm.
shows a typical en-face image (amplitude image of UDefocused(x,y)) and recovered clear image (amplitude image of UFocused(x,y)) of human finger. Figure 8(a)
Fig. 8 (a)The defocused y-z plane image of palm and (b) recovered image without using phase compensated complex signal and (c) recovered image using phase compensated complex signal based on scalar diffraction algorithm.
is a typical cross-section (y-z plane) image of an original 3D volume data of human palm. Figure 8(b) is the recovered image without using phase compensation, which is acquired by digitally focusing en face images in different depths of 3D volume data according to Eq. (11) using original complex signals, and extracting the y-z plane image with the same x coordinate as Fig. 8(a). Figure 8(c) is the digitally focused image using phase compensated complex signals, which is acquired also according to Eq. (11), but the original complex signals must be phase compensated using Eq. (10) before digitally focusing. It is obvious that the sweat gland in Fig. 8(b) is blurred compared with the sweat gland in Fig. 8(a) because of the influence of phase fluctuations, however, the sweat gland in Fig. 8(c) is focused and clear since the phase fluctuations are compensated. In order to further demonstrate the effectiveness of the phase compensation, we plotted the gray value of the line marked z = zi in Figs. 8(a)-8(c) as displayed in Figs. 9(a)
Fig. 9 Gray value along the line Z = Zi in (a) Fig. 8(a) (b) Fig. 8(b) and (c) Fig. 8(c), respectively.
, 9(b) and 9(c), respectively. It is easy to find that the width of the intensity profile of the sweat gland in Fig. 9(c) is the smallest among three intensity profiles, therefore, the lateral resolution of the FDOCT system is improved by digitally focusing using the phase compensated complex signals.

3.2 Application of phase registration in retina blood flow imaging

Each B-scan complex signal can be written as
F(x,z)=A(x,z)exp[iφ(x,z)]
(12)
where A(x,z)and φ(x,z)are the amplitude and the phase ofF(x,z), respectively. x is the lateral position of the probe beam, and z represents the coordinates along the imaging depth. Before generating B-scan cross-sectional blood flow image using two B frames F(x,z)t2and F(x,z)t1captured in the same B-scan cross section plane at different sampling time t2 and t1, phase fluctuations between adjacent A-scans of F(x,z)t2and F(x,z)t1 should be detected using the proposed method in Subsection 2.2 and expressed as Phasecompensation(x).Thus, phase registration of F(x,z)t2and F(x,z)t1is achieved by

FCompensated(x,z)t2=F(x,z)t2ejPhasecompensation(x)
(13)

The B-scan cross-sectional blood flow image is acquired by using Eq. (14) when the effects of sample motion are eliminated by applying the phase registration method.

Fblow(x,z)=ΔF(x,z)=FCompensated(x,z)t2F(x,z)t1
(14)

From Fig. 10
Fig. 10 Application of phase registration in retina blood flow imaging. (a) and (b) are two original B-scan cross section image of retina. (c) and (e) is extracted blood flow image before and after phase registration of A-scans, respectively. (d) is phase differences between A-scans of two original B-scans. (f) is the three dimensional display of the extracted blood flow image of retina.
, we can observe that the blood flow image can't be extracted effectively from original OCT signals [Figs. 10(a) and 10(b)] without phase registration. Only the phase difference [Fig. 10(d)] between A-scans of two original B-Scans is accurately detected and compensated by the proposed method, the high quality two dimension blood flow image [Fig. 10(e)] can be acquired. The three-dimension blood flow image of retina [Fig. 10(f)], consisted of 200 B-scan blood flow images, shows the structure of blood vessels clearly.

4. Discussion on influence of parameters M and ε

Parameters M and ε in Eq. (7) and Eq. (8), corresponding to a number of elements in phase distribution vector and the threshold value, respectively, have direct influence on the quality of phase registration method. We conducted a large number of experiment researches on the changes of phase difference measurement results, digitally focused image quality and retina blood flow image quality with parameters M and ε.

Figure 11
Fig. 11 Search results of strongly scattering points on the surface of two typical adjacent A-scans of human finger and calculated results of phase difference for different M and ε.
shows a typical B-scan image of human finger and the change of searching results of strongly scattering points and phase difference measurement results of two adjacent A-scans with parameters M and ε. From the B-scan image and the phase difference curves in Fig. 11, we can see that pixels from z = 30 to z = 36 are real strongly scattering points on the surface of two A-scans. The red horizontal lines denote calculated phase differences, and “ο” are used to label the found strongly scattering points. The phase difference between two adjacent A-scans cannot be detected accurately when M is very small (M = 2, ε = 0.5, 1, 1.5, 2; M = 3, ε = 1, 2). Conversely, strongly scattering points cannot be found and phase registration procedure fail when M is very big (M = 6, ε = 0.5; M = 7, ε = 1; M = 8, ε = 1.5, 2). Similarly, when ε is small, the phase difference measurement results is accurate, however, it is easy to cause phase registration fail. Conversely, when ε is bigger, the phase difference measurement results is inaccurate, however, phase registration fail can be avoided.

Figure 12
Fig. 12 Digitally focused results of a y-z plane image of human palm for different M and ε.
shows a typical y-z plane image of human palm and the change of digitally focused results with parameters M and ε. The lateral resolution of the image was improved and clear sweat gland images were acquired when parameters M and ε were set properly, such as M = 3, ε = 0.5, 1, 1.5, 2; M = 4, ε = 1, 1.5, 2; M = 5, ε = 1.5, 2; M = 6, ε = 1.5, 2.

Figure 13
Fig. 13 Results of retina blood flow imaging before phase registration (on the top right corner) and after phase registration for different M and ε.
shows two B-scan frames of retina captured in the same B-scan cross section plane at different sampling time, extracted blood flow image without phase compensation (on the top right corner of Fig. 13), and extracted blood flow images obtained using Eq. (14) and Eq. (13) for different M and ε. The retina blood flow images were effectively extracted after phase registration process was performed with proper M and ε, such as M = 3, ε = 0.5; M = 4, ε = 1, 1.5, 2; M = 5, ε = 1.5, 2; M = 6, ε = 2.

Figure 14
Fig. 14 Better choices of M and ε for three different applications.
shows the better choices of parameters M and ε for three different applications. “o” denotes preferred value for phase difference measurement of human finger, “Δ” denotes preferred value for image digitally focusing of human palm, and “﹡” denotes preferred value for retina blood flow imaging.

Parameters M and ε for phase registration of a specific sample can be selected according to experiment results as displayed in Fig. 14 or refer to the following basic rules:

Second step is to fix the value of ε. Parameter ε is determined based on the selected value of M and the sample variation between two A-scans. If M is big, ε cannot be selected too small, since it is easy to cause phase registration procedure fail. If M is small, ε cannot be selected too big, since phase difference cannot be detected accurately.

5. Conclusion

Phase difference between two FDOCT A-scans sampled at the same or adjacent scanning position is detected by searching similar phase distribution characteristics on the surface of two A-scans. The performance of the proposed phase registration method was tested by the recovery of defocused images of human finger and palm and by the blood flow imaging of retina by using phase compensated complex signals. As displayed in Eq. (7) and Eq. (8), phase distribution characteristics is defined as a vector consisting of M elements and an inequality with a threshold value ε is used to find strongly scattered points on the sample surface for phase registration. Parameters M and ε have direct influence on the quality of phase registration method, improper value of M and ε may cause inaccurate phase registration or fail of phase registration procedure. Parameters M and ε for a specific sample can be determined according to systematic experiment results, the internal structure of the sample and the sample variation between two A-scans caused by sample surface roughness, system parameters, scanning parameters, system mechanical precision and especially the sample movement. Our future work will focus on parameters optimization methods and the joint registration of amplitude and phase. The proposed phase registration method is of great advantage due to the fact that it is accurate because the effect of noise is eliminated. Secondly it is fast since only the sample surface signal is matched. Moreover it is convenient because no hardware is needed. These afore advantages make it possible to achieve 3D phase stability in FDOCT for in vive applications.

Acknowledgments

This work was supported by the National Natural Science Foundation of China (Grant Nos. 61240057 and 61108047)

References and links

1.

A. F. Fercher, C. K. Hitzenberger, G. Kamp, and S. Y. El-Zaiat, “Measurement of intraocular distances by backscattering spectral interferometry,” Opt. Commun. 117(1–2), 43–48 (1995). [CrossRef]

2.

M. Wojtkowski, R. Leitgeb, A. Kowalczyk, T. Bajraszewski, and A. F. Fercher, “In vivo human retinal imaging by Fourier domain optical coherence tomography,” J. Biomed. Opt. 7(3), 457–463 (2002). [CrossRef] [PubMed]

3.

D. Huang, E. A. Swanson, C. P. Lin, J. S. Schuman, W. G. Stinson, W. Chang, M. R. Hee, T. Flotte, K. Gregory, C. A. Puliafito, and J. G. Fujimoto, “Optical coherence tomography,” Science 254(5035), 1178–1181 (1991). [CrossRef] [PubMed]

4.

N. A. Nassif, B. Cense, B. H. Park, M. C. Pierce, S. H. Yun, B. E. Bouma, G. J. Tearney, T. C. Chen, and J. F. de Boer, “In vivo high-resolution video-rate spectral-domain optical coherence tomography of the human retina and optic nerve,” Opt. Express 12(3), 367–376 (2004). [CrossRef] [PubMed]

5.

R. K. Wang and Z. Ma, “A practical approach to eliminate autocorrelation artefacts for volume-rate spectral domain optical coherence tomography,” Phys. Med. Biol. 51(12), 3231–3239 (2006). [CrossRef] [PubMed]

6.

Y. Jia, P. O. Bagnaninchi, Y. Yang, A. E. Haj, M. T. Hinds, S. J. Kirkpatrick, and R. K. Wang, “Doppler optical coherence tomography imaging of local fluid flow and shear stress within microporous scaffolds,” J. Biomed. Opt. 14(3), 034014 (2009). [CrossRef] [PubMed]

7.

M. A. Choma, A. K. Ellerbee, C. Yang, T. L. Creazzo, and J. A. Izatt, “Spectral-domain phase microscopy,” Opt. Lett. 30(10), 1162–1164 (2005). [CrossRef] [PubMed]

8.

C. Joo, T. Akkin, B. Cense, B. H. Park, and J. F. de Boer, “Spectral-domain optical coherence phase microscopy for quantitative phase-contrast imaging,” Opt. Lett. 30(16), 2131–2133 (2005). [CrossRef] [PubMed]

9.

J. F. de Boer, T. E. Milner, M. J. C. van Gemert, and J. S. Nelson, “Two-dimensional birefringence imaging in biological tissue by polarization-sensitive optical coherence tomography,” Opt. Lett. 22(12), 934–936 (1997). [CrossRef] [PubMed]

10.

P. H. Tomlins and R. K. Wang, “Digital phase stabilization to improve detection sensitivity for optical coherence tomography,” Meas. Sci. Technol. 18(11), 3365–3372 (2007). [CrossRef]

11.

D. Adler, T. Ko, P. Herz, and J. Fujimoto, “Optical coherence tomography contrast enhancement using spectroscopic analysis with spectral autocorrelation,” Opt. Express 12(22), 5487–5501 (2004). [CrossRef] [PubMed]

12.

C. Xu, C. Vinegoni, T. S. Ralston, W. Luo, W. Tan, and S. A. Boppart, “Spectroscopic spectral-domain optical coherence microscopy,” Opt. Lett. 31(8), 1079–1081 (2006). [CrossRef] [PubMed]

13.

T. S. Ralston, D. L. Marks, P. Scott Carney, and S. A. Boppart, “Interferometric synthetic aperture microscopy,” Nat. Phys. 3(2), 129–134 (2007). [CrossRef]

14.

L. Yu, B. Rao, J. Zhang, J. Su, Q. Wang, S. Guo, and Z. Chen, “Improved lateral resolution in optical coherence tomography by digital focusing using two-dimensional numerical diffraction method,” Opt. Express 15(12), 7634–7641 (2007). [CrossRef] [PubMed]

15.

T. S. Ralston, D. L. Marks, F. Kamalabadi, and S. A. Boppart, “Deconvolution methods for mitigation of transverse blurring in optical coherence tomography,” IEEE Trans. Image Process. 14(9), 1254–1264 (2005). [CrossRef] [PubMed]

16.

Y. Liu, Y. Liang, G. Mu, and X. Zhu, “Deconvolution methods for image deblurring in optical coherence tomography,” J. Opt. Soc. Am. A 26(1), 72–77 (2009). [CrossRef] [PubMed]

17.

G. Liu, S. Yousefi, Z. Zhi, and R. K. Wang, “Automatic estimation of point-spread-function for deconvoluting out-of-focus optical coherence tomographic images using information entropy-based approach,” Opt. Express 19(19), 18135–18148 (2011). [CrossRef] [PubMed]

18.

B. White, M. Pierce, N. Nassif, B. Cense, B. Park, G. Tearney, B. Bouma, T. Chen, and J. F. de Boer, “In vivo dynamic human retinal blood flow imaging using ultra-high-speed spectral domain optical coherence tomography,” Opt. Express 11(25), 3490–3497 (2003). [CrossRef] [PubMed]

19.

J. Lee, V. Srinivasan, H. Radhakrishnan, and D. A. Boas, “Motion correction for phase-resolved dynamic optical coherence tomography imaging of rodent cerebral cortex,” Opt. Express 19(22), 21258–21270 (2011). [CrossRef] [PubMed]

20.

A. B. Vakhtin, D. J. Kane, W. R. Wood, and K. A. Peterson, “Common-path interferometer for frequency-domain optical coherence tomography,” Appl. Opt. 42(34), 6953–6958 (2003). [CrossRef] [PubMed]

21.

D. Lin, X. Jiang, F. Xie, W. Zhang, L. Zhang, and I. Bennion, “High stability multiplexed fiber interferometer and its application on absolute displacement measurement and on-line surface metrology,” Opt. Express 12(23), 5729–5734 (2004). [CrossRef] [PubMed]

22.

Z. Yaqoob, W. Choi, S. Oh, N. Lue, Y. Park, C. Fang-Yen, R. R. Dasari, K. Badizadegan, and M. S. Feld, “Improved phase sensitivity in spectral domain phase microscopy using line-field illumination and self phase-referencing,” Opt. Express 17(13), 10681–10687 (2009). [CrossRef] [PubMed]

23.

C. Yang, A. Wax, M. S. Hahn, K. Badizadegan, R. R. Dasari, and M. S. Feld, “Phase-referenced interferometer with subwavelength and subhertz sensitivity applied to the study of cell membrane dynamics,” Opt. Lett. 26(16), 1271–1273 (2001). [CrossRef] [PubMed]

24.

T. S. Ralston, D. L. Marks, P. S. Carney, and S. A. Boppart, “Phase stability technique for inverse scattering in optical coherence tomography,” 3rd IEEE International Symposium on Biomedical Imaging: Nano to Macro, 578–581 (2006). [CrossRef]

25.

L. An and R. K. Wang, “In vivo volumetric imaging of vascular perfusion within human retina and choroids with optical micro-angiography,” Opt. Express 16(15), 11438–11452 (2008). [CrossRef] [PubMed]

26.

R. K. Wang and Z. H. Ma, “A practical approach to eliminate autocorrelation artefacts for volume-rate spectral domain optical coherence tomography,” Phys. Med. Biol. 51(12), 3231–3239 (2006). [CrossRef] [PubMed]

27.

G. Liu, Z. Zhi, and R. K. Wang, “Digital focusing of OCT images based on scalar diffraction theory and information entropy,” Biomed. Opt. Express 3(11), 2774–2783 (2012). [CrossRef] [PubMed]

28.

G. Liu and R. K. Wang, “Stripe motion artifact suppression in phase-resolved OCT blood flow images of the human eye based on the frequency rejection filter,” Chin. Opt. Lett. 11(3), 031701–031705 (2013). [CrossRef]

OCIS Codes
(100.5070) Image processing : Phase retrieval
(170.3010) Medical optics and biotechnology : Image reconstruction techniques
(170.3880) Medical optics and biotechnology : Medical and biological imaging
(170.4500) Medical optics and biotechnology : Optical coherence tomography
(170.5755) Medical optics and biotechnology : Retina scanning

ToC Category:
Medical Optics and Biotechnology

History
Original Manuscript: January 31, 2013
Revised Manuscript: April 20, 2013
Manuscript Accepted: May 17, 2013
Published: May 23, 2013

Virtual Issues
Vol. 8, Iss. 7 Virtual Journal for Biomedical Optics

Citation
Guozhong Liu, Zhehai Zhou, and Ping Li, "Phase registration based on matching of phase distribution characteristics and its application in FDOCT," Opt. Express 21, 13241-13255 (2013)
http://www.opticsinfobase.org/vjbo/abstract.cfm?URI=oe-21-11-13241


Sort:  Author  |  Year  |  Journal  |  Reset  

References

  1. A. F. Fercher, C. K. Hitzenberger, G. Kamp, and S. Y. El-Zaiat, “Measurement of intraocular distances by backscattering spectral interferometry,” Opt. Commun.117(1–2), 43–48 (1995). [CrossRef]
  2. M. Wojtkowski, R. Leitgeb, A. Kowalczyk, T. Bajraszewski, and A. F. Fercher, “In vivo human retinal imaging by Fourier domain optical coherence tomography,” J. Biomed. Opt.7(3), 457–463 (2002). [CrossRef] [PubMed]
  3. D. Huang, E. A. Swanson, C. P. Lin, J. S. Schuman, W. G. Stinson, W. Chang, M. R. Hee, T. Flotte, K. Gregory, C. A. Puliafito, and J. G. Fujimoto, “Optical coherence tomography,” Science254(5035), 1178–1181 (1991). [CrossRef] [PubMed]
  4. N. A. Nassif, B. Cense, B. H. Park, M. C. Pierce, S. H. Yun, B. E. Bouma, G. J. Tearney, T. C. Chen, and J. F. de Boer, “In vivo high-resolution video-rate spectral-domain optical coherence tomography of the human retina and optic nerve,” Opt. Express12(3), 367–376 (2004). [CrossRef] [PubMed]
  5. R. K. Wang and Z. Ma, “A practical approach to eliminate autocorrelation artefacts for volume-rate spectral domain optical coherence tomography,” Phys. Med. Biol.51(12), 3231–3239 (2006). [CrossRef] [PubMed]
  6. Y. Jia, P. O. Bagnaninchi, Y. Yang, A. E. Haj, M. T. Hinds, S. J. Kirkpatrick, and R. K. Wang, “Doppler optical coherence tomography imaging of local fluid flow and shear stress within microporous scaffolds,” J. Biomed. Opt.14(3), 034014 (2009). [CrossRef] [PubMed]
  7. M. A. Choma, A. K. Ellerbee, C. Yang, T. L. Creazzo, and J. A. Izatt, “Spectral-domain phase microscopy,” Opt. Lett.30(10), 1162–1164 (2005). [CrossRef] [PubMed]
  8. C. Joo, T. Akkin, B. Cense, B. H. Park, and J. F. de Boer, “Spectral-domain optical coherence phase microscopy for quantitative phase-contrast imaging,” Opt. Lett.30(16), 2131–2133 (2005). [CrossRef] [PubMed]
  9. J. F. de Boer, T. E. Milner, M. J. C. van Gemert, and J. S. Nelson, “Two-dimensional birefringence imaging in biological tissue by polarization-sensitive optical coherence tomography,” Opt. Lett.22(12), 934–936 (1997). [CrossRef] [PubMed]
  10. P. H. Tomlins and R. K. Wang, “Digital phase stabilization to improve detection sensitivity for optical coherence tomography,” Meas. Sci. Technol.18(11), 3365–3372 (2007). [CrossRef]
  11. D. Adler, T. Ko, P. Herz, and J. Fujimoto, “Optical coherence tomography contrast enhancement using spectroscopic analysis with spectral autocorrelation,” Opt. Express12(22), 5487–5501 (2004). [CrossRef] [PubMed]
  12. C. Xu, C. Vinegoni, T. S. Ralston, W. Luo, W. Tan, and S. A. Boppart, “Spectroscopic spectral-domain optical coherence microscopy,” Opt. Lett.31(8), 1079–1081 (2006). [CrossRef] [PubMed]
  13. T. S. Ralston, D. L. Marks, P. Scott Carney, and S. A. Boppart, “Interferometric synthetic aperture microscopy,” Nat. Phys.3(2), 129–134 (2007). [CrossRef]
  14. L. Yu, B. Rao, J. Zhang, J. Su, Q. Wang, S. Guo, and Z. Chen, “Improved lateral resolution in optical coherence tomography by digital focusing using two-dimensional numerical diffraction method,” Opt. Express15(12), 7634–7641 (2007). [CrossRef] [PubMed]
  15. T. S. Ralston, D. L. Marks, F. Kamalabadi, and S. A. Boppart, “Deconvolution methods for mitigation of transverse blurring in optical coherence tomography,” IEEE Trans. Image Process.14(9), 1254–1264 (2005). [CrossRef] [PubMed]
  16. Y. Liu, Y. Liang, G. Mu, and X. Zhu, “Deconvolution methods for image deblurring in optical coherence tomography,” J. Opt. Soc. Am. A26(1), 72–77 (2009). [CrossRef] [PubMed]
  17. G. Liu, S. Yousefi, Z. Zhi, and R. K. Wang, “Automatic estimation of point-spread-function for deconvoluting out-of-focus optical coherence tomographic images using information entropy-based approach,” Opt. Express19(19), 18135–18148 (2011). [CrossRef] [PubMed]
  18. B. White, M. Pierce, N. Nassif, B. Cense, B. Park, G. Tearney, B. Bouma, T. Chen, and J. F. de Boer, “In vivo dynamic human retinal blood flow imaging using ultra-high-speed spectral domain optical coherence tomography,” Opt. Express11(25), 3490–3497 (2003). [CrossRef] [PubMed]
  19. J. Lee, V. Srinivasan, H. Radhakrishnan, and D. A. Boas, “Motion correction for phase-resolved dynamic optical coherence tomography imaging of rodent cerebral cortex,” Opt. Express19(22), 21258–21270 (2011). [CrossRef] [PubMed]
  20. A. B. Vakhtin, D. J. Kane, W. R. Wood, and K. A. Peterson, “Common-path interferometer for frequency-domain optical coherence tomography,” Appl. Opt.42(34), 6953–6958 (2003). [CrossRef] [PubMed]
  21. D. Lin, X. Jiang, F. Xie, W. Zhang, L. Zhang, and I. Bennion, “High stability multiplexed fiber interferometer and its application on absolute displacement measurement and on-line surface metrology,” Opt. Express12(23), 5729–5734 (2004). [CrossRef] [PubMed]
  22. Z. Yaqoob, W. Choi, S. Oh, N. Lue, Y. Park, C. Fang-Yen, R. R. Dasari, K. Badizadegan, and M. S. Feld, “Improved phase sensitivity in spectral domain phase microscopy using line-field illumination and self phase-referencing,” Opt. Express17(13), 10681–10687 (2009). [CrossRef] [PubMed]
  23. C. Yang, A. Wax, M. S. Hahn, K. Badizadegan, R. R. Dasari, and M. S. Feld, “Phase-referenced interferometer with subwavelength and subhertz sensitivity applied to the study of cell membrane dynamics,” Opt. Lett.26(16), 1271–1273 (2001). [CrossRef] [PubMed]
  24. T. S. Ralston, D. L. Marks, P. S. Carney, and S. A. Boppart, “Phase stability technique for inverse scattering in optical coherence tomography,” 3rd IEEE International Symposium on Biomedical Imaging: Nano to Macro, 578–581 (2006). [CrossRef]
  25. L. An and R. K. Wang, “In vivo volumetric imaging of vascular perfusion within human retina and choroids with optical micro-angiography,” Opt. Express16(15), 11438–11452 (2008). [CrossRef] [PubMed]
  26. R. K. Wang and Z. H. Ma, “A practical approach to eliminate autocorrelation artefacts for volume-rate spectral domain optical coherence tomography,” Phys. Med. Biol.51(12), 3231–3239 (2006). [CrossRef] [PubMed]
  27. G. Liu, Z. Zhi, and R. K. Wang, “Digital focusing of OCT images based on scalar diffraction theory and information entropy,” Biomed. Opt. Express3(11), 2774–2783 (2012). [CrossRef] [PubMed]
  28. G. Liu and R. K. Wang, “Stripe motion artifact suppression in phase-resolved OCT blood flow images of the human eye based on the frequency rejection filter,” Chin. Opt. Lett.11(3), 031701–031705 (2013). [CrossRef]

Cited By

Alert me when this paper is cited

OSA is able to provide readers links to articles that cite this paper by participating in CrossRef's Cited-By Linking service. CrossRef includes content from more than 3000 publishers and societies. In addition to listing OSA journal articles that cite this paper, citing articles from other participating publishers will also be listed.


« Previous Article  |  Next Article »

OSA is a member of CrossRef.

CrossCheck Deposited