Open Access
Add to BibSonomy
Abstract
This paper describes a balanced detection spectral-domain optical coherence tomography (BD-SD-OCT) system for suppressing autocorrelation (AC) artifacts and increasing the signal-to-noise ratio (SNR). The system employed three optical fiber couplers to generate two phase-opposed interference spectra that were acquired by a single line-scan camera simultaneously. When compared with conventional unbalanced detection SD-OCT systems, the developed BD-SD-OCT system improved the SNR by 5.4-6 dB and suppressed the AC term by 5-10 dB. The imaging quality of the BD-SD-OCT system was evaluated by in vivo imaging of human nail folds and retinas.
© 2022 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
As a noncontact and noninvasive imaging technique that measures the interference of backscattered light [1], optical coherence tomography (OCT) has become a powerful tool in biomedical imaging, especially in ophthalmology [2], cardiology [3,4] and dermatology [5–7]. When compared with time domain OCT (TD-OCT), Fourier domain OCT (FD-OCT) has a much faster imaging speed and improved sensitivity [8]. FD-OCT can, in principle, be divided into two types: spectral domain OCT (SD-OCT), which uses a high-speed camera and spectrometer, and swept source OCT (SS-OCT), which uses a rapidly tunable laser [9]. Because of its working wavelength, which is suitable for retina imaging, SD-OCT has become dominant in OCT applications in ophthalmology.
The FD-OCT spectral interferogram includes a direct current term (DC), an autocorrelation term (AC), which is the mutual interference from reflectors at different sample depths [10], and a cross-correlation interference term (CC). To suppress the DC and AC terms that yield low frequency noise around the zero-delay line, the reference spectrum is typically subtracted from the interference spectrum as the background [8,11]. Alternatively, all the A-scan signals in each B-scan image can be averaged as the background and then subtracted. In addition, as a more effective method, balanced detection has been widely implemented in TD-OCT and SS-OCT with balanced photodetectors to eliminate the DC and AC terms [12–14]. Recently, several balanced detection SD-OCT (BD-SD-OCT) techniques have been proposed. In 2005, Ai et al. [15] used an area scan camera-based spectrometer to detect two opposed-phase interferometric spectra simultaneously. The AC and DC terms can be reduced by calculating the difference between the two spectral fringes recorded by the two horizontal pixel lines of the camera. However, the imaging speed of the system is limited by the acquisition rate of the area scan camera. Another approach utilized two spectrometers to perform balanced detection [16,17], but this approach requires bulky and complex setups. Furthermore, it is technically difficult to balance two non-identical spectrometers. In 2015, Bo et al. [18] presented a BD-SD-OCT system with a spectrometer with a v-groove fiber array (VGA) and a three-line CCD camera. However, interline crosstalk in the CCD caused noticeable residual AC artifacts in images. Hyeon et al. [19] proposed an optical switch-based BD-SD-OCT system that used a single line-scan camera for balanced detection of interferometric spectra. However, signal loss and dispersion in the long fiber cord that was used as an optical delay line for BD spectra matching in one channel caused an additional imbalance between the two channels.
In this work, we propose a BD-SD-OCT system with a single spectrometer that uses a 4096 pixel line-scan camera to detect opposed-phase interference spectra simultaneously, with 2048 pixels in each channel. This technique allows for a relatively simple structure while taking advantage of the high acquisition rate of the line-scan camera and avoiding crosstalk between the BD spectra.
2. Methods
Fig. 1 shows the schematic of the developed BD-SD-OCT system based on a fiber optic Michelson interferometer. The light source in the BD-SD-OCT system is a low coherence superluminescent diode (SLD, IPSDS0808, Inphenix, Livermore, CA) with a central wavelength of 850 nm and an FWHM bandwidth of 50 nm, which provides a coherence length of 4.65 μm in tissue (n=1.37). Light from the SLD is split into a sample arm and a reference arm by a 2×2 50/50 fiber coupler FC1 (F-CPL-S22855-FCAPC, Newport, CA) and then recombined in another fiber coupler FC3 (F-CPL-S22855-FCAPC, Newport, CA) and coupled out into two detection channels, generating two phase-opposed interference signals. The spectrometer is composed of a diffraction grating (VPH Transmission Grating, 1200 lines/mm@840 nm, Wasatch Photonics, Logan, UT) with a designed angle of incidence (AOI) of 30.2 degrees, a composite lens (L3, f=140 mm) and a 4096 pixel line scan camera (SPL4096-39Km, Basler AG, Germany) with a pixel size of 10 μm × 10 μm, a line rate of 39 kHz and a resolution of 12 bits. The light beams of channels 1 and 2 are incident on the spectrometer at angles of 26 and 34 degrees, respectively. In the reference arm, the distance between the collimating lenses CL2 and CL3 can be adjusted to alter the optical pathlength. A fiber coupler FC2 (F-CPL-S22855-FCAPC, Newport, CA) was used in the reference arm to compensate for the dispersion induced by the coupler FC1. The incident light power on the cornea of the human eye is 680 μW, which is well within the maximum permissible exposure level set by the American National Standards Institute (ANSI).
The channel 1 signal I1 and channel 2 signal I2 can be expressed in k-space by Eq. (1) and Eq. (2), respectively:
Fig. 1. Schematic diagram of the BD-SD-OCT setup. SLD: superluminescent diode, FC1-3: 50/50 fiber coupler, CL1-5: collimating lens, L1-3: lens, Gavo: galvanometer, DG: diffraction grating. PC: polarization controller
The first term in Eq. (1) and Eq. (2) represents a pathlength-independent DC bias, the second represents the CC term, which encodes the depth information of the sample, and the third represents the AC term, which describes the mutual interference between the different sample layers.
To match the two spectra, precalibration was required. First, 3000 reference spectra were averaged as the background and then subtracted from the interference spectra for each channel to remove the fixed-pattern noise and the DC term. Second, spline interpolation was used to resample the interferograms from λ- to k-space, based on the grating equation and geometrical optics [20]. Finally, the balanced interference spectra were obtained by subtracting the spectra of the two channels.
3. Experiments and results
In this study, one channel of the system was used in an unbalanced detection configuration to allow comparisons with the BD-SD-OCT technique. Fig. 2(A) shows the interference spectra recorded by channels 1 and 2 using a mirror as the sample at an optical path difference (OPD) of 0.4 mm. The spectra of the two channels exhibited nearly perfect opposed phases, as shown in the magnified inset of Fig. 2(A). Fig. 2(B) presents the averaged A-scan OCT signals without the background spectrum subtracted; when compared to unbalanced detection, the BD technique shows a DC suppression of approximately 20 dB and an SNR enhancement of 5.7 dB. The average SNR enhancement measured across the imaging range is 5.83 dB.
Fig. 2. (A) Interference spectra of channel 1 and channel 2 after resampling. (B) Fourier transforms of signals detected by the BD-SD-OCT system and unbalanced detection SD-OCT system.
To evaluate the ability of the system to eliminate the AC terms, a multilayer tape was used as the sample. Although a 50/50 fiber coupler was used to achieve BD, there was residual imbalance between the two channels, mainly due to the wavelength dependence of the coupler splitting ratio. Background subtraction was used to suppress residual imaging effects. When compared with Fig. 3(A) and (B), Fig. 3(C) shows less AC noise and increased visibility and depth while maintaining the sharpness of each layer. For a more detailed comparison, A-scan signals from the dotted lines in Fig. 3(A)-(C) were extracted, as shown in Fig. 3(D). In comparison to the unbalanced detection configuration, the BD technique achieves a DC suppression of 10 dB, an AC suppression of 5-10 dB, and an SNR enhancement of 5.5 dB when the background is subtracted.
Fig. 3. OCT images of a multilayer tape recorded by channel 1 (A), channel 2 (B) and BD (C). (D): A-scan signals of the dotted lines in Fig. 3(A), (B) and (C).
Human nail folds were imaged in vivo, as shown in Fig. 4. Similar to multilayer tape imaging, BD-SD-OCT has enhanced visibility and depth compared to the unbalanced detection configuration, with a DC suppression of 10 dB, an AC suppression of 5-10 dB, and an SNR enhancement of 5.4-6 dB. Fig. 5 shows in vivo imaging of the human retina. As shown in Fig. 5(C), the BD-SD-OCT image displays deeper layers of the choroid with AC noise cancelled.
Fig. 4. In vivo OCT images of human nail folds recorded by channel 1 (A), channel 2 (B) and BD (C). (D): A-scan signals of the dotted lines in Fig. 4(A), (B) and (C).
4. Discussions and conclusion
It is essential for SD-OCT to suppress the DC and AC terms and enhance the SNR. In this work, we propose a low-cost, simple-to-build BD-SD-OCT system with a 4096 pixel single line-scan camera. Two phase-opposed interference spectra incident on the spectrometer at different angles were acquired simultaneously. When compared to the unbalanced detection configuration, the SNR of the BD-SD-OCT system is increased by 5.4-6 dB. With an incident angle about 4 degrees away from the AOI of the grating, the diffraction efficiency drops by about 10% in comparison to that at the optimized AOI which affects the SNR by 0.9 dB correspondingly. This problem can be solved by choosing a high efficient grating with less angular dependence in the future.
To evaluate the performance of the BD-SD-OCT technique, multilayer tape and in vivo human nail fold imaging were performed, which demonstrated DC and AC suppressions of 10 dB and 5-10 dB, respectively. However, it is hard to completely remove AC noise due to the residual imbalance between the two channels.
In vivo imaging of the human retina also demonstrated that the BD-SD-OCT technique proposed in this paper is capable of suppressing AC noise in retina imaging and acquiring deeper layers of the choroid in greater detail. It is demonstrated that the BD-SD-OCT system in this paper can be used to detect human retinal tissue with excellent imaging quality.
Funding
National Natural Science Foundation of China (61975246); Special Fund for Distinguished Experts in Guangxi of China.
Disclosures
The authors declare no conflicts of interest.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
References
1. 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]
2. J. J. Kałuzny, A. Szkulmowska, T. Bajraszewski, M. Szkulmowski, B. J. Kałuzny, I. Gorczyńska, P. Targowski, and M. Wojtkowski, “Retinal imaging by spectral optical coherence tomography,” Eur. J. Ophthalmol. 17(2), 238–245 (2007). [CrossRef]
3. H. C. Lee, O. O. Ahsen, K. Liang, Z. Wang, M. Figueiredo, M. G. Giacomelli, B. Potsaid, Q. Huang, H. Mashimo, and J. G. Fujimoto, “Endoscopic optical coherence tomography angiography microvascular features associated with dysplasia in Barrett's esophagus (with video),” Gastrointest. Endosc. 86(3), 476–484 (2017). [CrossRef]
4. W. J. Choi and R. K. Wang, “Optical coherence tomography imaging of cranial meninges post brain injury in vivo,” Chin. Opt. Lett. 15(9), 090005 (2017). [CrossRef]
5. K. H. Cheng, A. Mariampillai, K. K. Lee, B. Vuong, T. W. Luk, J. Ramjist, A. Curtis, H. Jakubovic, P. Kertes, M. Letarte, M. E. Faughnan, and V. X. Yang, “Histogram flow mapping with optical coherence tomography for in vivo skin angiography of hereditary hemorrhagic telangiectasia,” J. Biomed. Opt. 19(8), 086015 (2014). [CrossRef]
6. H. Zhong, Z. Zhang, Z. Guo, H. Wei, H. Yang, Y. He, S. Xie, and S. Liu, “Using OCT image to distinguish human acupoint from non-acupoint tissues after irradiation with laser in vivo:a pilot study,” Chin. Opt. Lett. 8(4), 418–420 (2010). [CrossRef]
7. W. Gao, M. Hu, Q. Chen, X. Yue, Q. Jiang, P. Li, J. Fu, J. Zhu, X. Yang, and L. Bo, “Effects of layer structure of the skin on the measured signal by optical coherence tomography,” Chin. Opt. Lett. 3(S1), S156–S158 (2005).
8. R. Leitgeb, C. K. Hitzenberger, and A. F. Fercher, “Performance of fourier domain vs. time domain optical coherence tomography,” Opt. Express 11(8), 889–894 (2003). [CrossRef]
9. M. Chen, Z. Ding, L. Xu, T. Wu, C. Wang, G. Shi, and Y. Zhang, “All-fiber ring-cavity based frequency swept laser source for frequency domain OCT,” Chin. Opt. Lett. 8(2), 202–205 (2010). [CrossRef]
10. 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]
11. S. H. Yun, G. J. Tearney, J. F. de Boer, N. Iftimia, and B. E. Bouma, “High-speed optical frequency-domain imaging,” Opt. Express 11(22), 2953–2963 (2003). [CrossRef]
12. A. G. Podoleanu, “Unbalanced versus balanced operation in an optical coherence tomography system,” Appl. Optics 39(1), 173–182 (2000). [CrossRef]
13. J. Mo, M. de Groot, and J. F. de Boer, “Focus-extension by depth-encoded synthetic aperture in Optical Coherence Tomography,” Opt. Express 21(8), 10048–10061 (2013). [CrossRef]
14. M. A. Choma, K. Hsu, and J. A. Izatt, “Swept source optical coherence tomography using an all-fiber 1300-nm ring laser source,” J. Biomed. Opt. 10(4), 44009 (2005). [CrossRef]
15. J. Ai and L. V. Wang, “Synchronous self-elimination of autocorrelation interference in Fourier-domain optical coherence tomography,” Opt. Lett. 30(21), 2939–2941 (2005). [CrossRef]
16. A. Bradu and A. G. Podoleanu, “Fourier domain optical coherence tomography system with balance detection,” Opt. Express 20(16), 17522–17538 (2012). [CrossRef]
17. W. C. Kuo, C. M. Lai, Y. S. Huang, C. Y. Chang, and Y. M. Kuo, “Balanced detection for spectral domain optical coherence tomography,” Opt. Express 21(16), 19280–19291 (2013). [CrossRef]
18. E. Bo, X. Liu, S. Chen, X. Yu, X. Wang, and L. Liu, “Spectral-domain optical coherence tomography with dual-balanced detection for auto-correlation artifacts reduction,” Opt. Express 23(21), 28050–28058 (2015). [CrossRef]
19. M. G. Hyeon, H. J. Kim, B. M. Kim, and T. J. Eom, “Spectral domain optical coherence tomography with balanced detection using single line-scan camera and optical delay line,” Opt. Express 23(18), 23079–23091 (2015). [CrossRef]
20. B. Park, M. C. Pierce, B. Cense, S. H. Yun, M. Mujat, G. Tearney, B. Bouma, and J. de Boer, “Real-time fiber-based multi-functional spectral-domain optical coherence tomography at 1.3 microm,” Opt. Express 13(11), 3931–3944 (2005). [CrossRef]
