Open Access
Add to BibSonomy
Abstract
We present a system for generating a 1.7 µm quasi-supercontinuum (SC) as what we believe to be a new broadband light source for spectral domain (SD) optical coherence tomography (OCT). The quasi-SC source, whose spectral range and spectral shape can be arbitrarily changed, is based on wavelength-tunable soliton pulse generation and high-speed intensity modulation. A Gaussian-shaped quasi-SC was introduced into an SD-OCT system. A maximum axial resolution of 14 µm in air (corresponding to 10 µm in tissue) and an imaging sensitivity of 94 dB was obtained. We also demonstrated high-resolution imaging for tape stack, human nail and pig thyroid gland samples using a quasi-SC. From these results, we confirmed the usefulness of a quasi-SC for SD-OCT imaging.
© 2023 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Optical coherence tomography (OCT) with micrometer resolution and cross-sectional imaging capabilities has been widely applied to clinical diagnosis and medical research [1–4]. Theoretically, the spatial resolution and imaging depth are determined by the light source. The axial resolution is primarily determined by the coherence length of the light source, given by $\frac{{2\ln 2}}{\pi }({{{{\lambda_c}^2} / {\Delta \lambda }}} )$ [5], where ${\lambda _c}$ and $\Delta \lambda $ are the central wavelength and bandwidth of the light source, respectively. A broadband light source has to be used for higher resolution. In recent years, supercontinuum (SC) sources with broad bandwidths have been applied to OCT for ultrahigh resolution deep tissue imaging [6–13]. The coherent SC generated by using normally dispersive highly nonlinear fiber (HNLF) has low spectral intensity noise [14,15]. However, the spectral shape and wavelength range are limited by the fiber device used and the characteristics of the pulse source, and generally it is difficult to control the spectral shape of the generated SC [9,16,17].
In 2008, Sumimura et al. developed a new broadband light source—a quasi-SC source, whose spectrum looks like a SC and is controllable [18]. They continued their work and demonstrated 1.3 µm quasi-SC generation for time domain (TD) OCT imaging in 2010 [19]. The developed quasi-SC source employs a combination of wavelength-tunable Raman soliton pulse generation and fast intensity control [20,21]. When an ultrashort pulse with a high peak power propagates in the nonlinear fiber, a Raman soliton pulse is created at the longer wavelength side of the seed pulse spectrum due to the nonlinear effect. The wavelength of the Raman soliton pulse can be varied by changing the fiber length and fiber input power. A quasi-SC is obtained by changing the wavelengths of the shifted soliton pulses continuously and rapidly through ultrafast optical modulation.
The quasi-SC has an interesting broadband light source with high tunability. The central wavelength and bandwidth are tunable by changing the modulation function. In OCT imaging, the theoretical axial resolution depends on the bandwidth of the light source. Since chromatic dispersion and absorption limit the penetration depth and resolution, the control of the central wavelength and bandwidth are important to optimize the OCT imaging. Besides, the quasi-SC is the first wavelength tunable broadband light source. Using this light source, the effect of speckle reduction is expected in OCT imaging.
In this work, we demonstrated highly sensitive, high-resolution SD-OCT using a quasi-SC source. We generated Gaussian-shaped quasi-SC spectra with tunable bandwidth in the 1.7 µm spectral band, which has a local minimum of water absorption and lower scattering in biological tissue, helping us to observe deeper tissue regions compared with the 1.3 µm wavelength region [22,23]. Based on a 1.55 µm ultrashort pulse fiber laser and a polarization-maintaining (PM) fiber [24,25], a quasi-SC was obtained by using an ultra-fast optical intensity modulator. We demonstrated SD-OCT using a tunable 1.7 µm quasi-SC for the first time. The characteristics of OCT imaging were examined while tuning the bandwidth and modulation frequency of the quasi-SC source. Then, we successfully observed the cross-sectional image of biological samples, and confirmed the availability of quasi-SC in realizing high-resolution deep tissue imaging.
2. Experimental
2.1 Quasi-SC generation in a 1.7 µm spectral band
The experimental setup for 1.7 µm quasi-SC generation is shown in Fig. 1. As the seed pulse source, we utilized an Er-doped ultrashort-pulse fiber laser using a single-wall carbon nanotube (SWNT) polyimide film as the saturable absorber [24]. The ultrashort pulse had an ideal sech2 shape with a spectral width of 5.3 nm full-width at half-maximum (FWHM) and a central wavelength of 1556 nm. The average output power was about 23 mW with a repetition rate of 95.5 MHz. In order to obtain a 1.7 µm spectral band, the 1.55 µm ultrashort pulse was amplified by an Er-doped fiber amplifier (EDFA). The maximum power of the amplified soliton pulse was about 360 mW. The amplified pulse was coupled into 300 m of PM fiber. A Raman soliton pulse is formed gradually during PM fiber transmission as follows: when an ultrashort pulse is coupled into an anomalous-dispersion fiber, the pulse spectrum is initially broadened by the higher-order soliton compression effect [18,19]. Then the longer-wavelength components in the broadened spectrum are enhanced through intra-pulse stimulated Raman scattering (SRS). As a result, pulse breakup occurs, and an ultrashort pulse is generated at the longer wavelength side. The generated pulse experiences the soliton effect, and a sech2-shaped soliton pulse is gradually generated. Intra-pulse SRS also causes the energy of the shorter-wavelength components in the Raman soliton pulse to be transferred to the longer-wavelength components in the same pulse, resulting in the central wavelength of the Raman soliton shifting toward the longer-wavelength side. As the fiber length and fiber-input power are increased, the central wavelength of the Raman soliton is shifted more toward the longer wavelength side. Thanks to the PM fibers, the output power and the central wavelength of Raman soliton were stable, and not sensitive to environmental fluctuations. Here, we inserted an electro-optical (EO) intensity modulator (Thorlabs LN81S-FC) between the seed pulse and EDFA to change the power of the amplified pulse. A long pass filter was used after the PM fiber to remove the seed pulse. The cut-off wavelength was 1.6 µm. Two combinations composed of a half-wave plate and a quarter-wave plate were also introduced to match the polarization direction of the pulse beam with the birefringent axis of the fiber [26]. The first combination after the seed pulse source helped to obtain the highest amplified power, whereas the second one before the PM fiber caused the Raman soliton to be shifted to the longest wavelength.
First, we demonstrated the wavelength-tunable soliton pulse generation. Figures 2(a) and 2(b) show the wavelength shift and output power of the Raman soliton as a function of the fiber-input power for fiber lengths of 100 and 300 m. As the fiber-input power was increased, the wavelength shift and output power increased almost linearly. These experimental results were close to the theoretical results and previously reported experimental results [20,21,24]. The 300 m fiber produced a longer wavelength shift than the 100 m fiber, but a reduced output power compared to the 100 m fiber. When the fiber length was 100 m, the maximum wavelength shift was 1835nm with an output power of 97.5 mW. When the fiber length was 300 m, the maximum wavelength shift was increased up to 1930nm, but it was accompanied by a second soliton pulse generation. For the subsequent quasi-SC generation, we had to adjust the polarization direction of the fiber input pulse to maintain single soliton generation. In this situation, the maximum wavelength shift was 1895nm with an output power of 76.5 mW when the fiber length was 300 m. To achieve the maximum tuning wavelength range, we chose a 300 m PM fiber (SM15-PR-U24A-H) as the wavelength shift device.
Fig. 2. (a) Wavelength shift and (b) output power of Raman soliton in terms of the fiber-input power for 100 m and 300 m fibers.
Figure 3(a) shows the generated Raman solitons in the wavelength range from 1.6 µm to 1.9 µm. The pulse spectra kept the ideal sech2 shapes. When the wavelength tuning frequency was higher than the sampling frequency of the detection system, the wavelength of the Raman soliton shifted continuously within one detection period, and the detected spectra was the superposition of these solitons and looked like a supercontinuum. For instance, when we increased the modulation frequency of the PM fiber input intensity beyond the 20 kHz detection rate of the spectrum analyzer (Yokogawa AQ6375), the shifted Raman solitons seemed to be generated at the same time, resulting in a detected spectrum that appeared to be super-continuous, that is to say, a quasi-SC. In the quasi-SC generation, the modulation function determined the shape of the quasi-SC by changing the wavelength tuning speed of the shifted Raman soliton pulse. In order to obtain a Gaussian-shaped spectrum, we used a programable function in a multifunction generator (NF WF1968), as shown in Fig. 3(b). The modulation amplitude and offset affected the central wavelength and bandwidth. By adjusting the amplitude and offset of the modulation function, a Gaussian-shaped quasi-SC spectrum with a bandwidth of 138 nm was generated in the 1.7 µm spectral band, as shown in Fig. 3(c).
Fig. 3. (a) Spectra of generated wavelength-tunable Raman solitons. (b) Intensity modulation function for Gaussian-shaped quasi-SC generation. (c) Spectra of broadened Gaussian-shaped quasi-SC output (λc = 1743nm, Δλ = 138 nm, output power = 51.2 mW, modulation frequency was about 1.1 MHz.).
Figure 4 shows the RF spectra of the generated quasi-SC, Raman soliton, and the 1.7 µm conventional SC developed by our group. The output power of conventional SC was 52.5 mW, and the bandwidth and central wavelength were 1682 nm and 142 nm, respectively. The RF spectra were observed by using a fast pin photodiode (EOT, ET-5000) and an RF spectrum analyzer (Anritsu MS2830A). We compared the spectra of the quasi-SC (with modulation), the 1750nm Raman soliton (without modulation), and the 1.7 µm conventional SC. In the quasi-SC generation, the fundamental soliton pulse was modulated with a frequency of 1.1 MHz. The noise level did not increase compared to Raman soliton generation and SC source below 100 kHz, but there were some small modulation noise peaks above that. It was considered that these spectral peaks were generated by intensity and phase modulation in the modulator. As shown below, these small peaks did not affect the SNR of the SD-OCT measurement.
2.2 SD-OCT using a 1.7 µm quasi-SC source
We applied the above quasi-SC to SD-OCT, as shown in Fig. 5(a). The whole system including the light source, reference arm, sample arm, and detection part, was connected by a 50/50 fiber coupler [21]. The sample arm consisted of XY-axis galvanometer scanners to obtain a cross-sectional or volumetric image, and a focusing lens with a focal length of 30 mm. In the reference arm, we inserted dispersion-compensating glass plates to match the dispersions of the two arms. Polarization controllers were also introduced to remove polarization mismatches. For OCT signal detection, we used a custom-built spectrometer composed of a 150 lines/mm blazed diffraction grating (Shimadzu 015-200), two focusing achromatic lenses and a 47 k lines/s InGaAs line scan camera (Goodrich SU1024LDH-2.2RT-0250/LC). The camera has an extended detection wavelength range from 1400 nm to 2000nm. The pixel number and digital output resolution were 1024 pixels and 14 bits, respectively.
Fig. 5. (a) Experimental setup for SD-OCT. (b) Interference signal of SD-OCT using a 1.7 µm quasi-SC.
To evaluate the performance of SD-OCT using a 1.7 µm quasi-SC, we examined the system sensitivity and axial resolution using a reflective mirror as a sample, as shown in Fig. 5(b). The incident power of the OCT system was 33.9 mW. The total system sensitivity was 94 dB, including 55 dB signal power and a 39 dB round-trip attenuation. The maximum axial resolution was 14.8 µm in air, corresponding to 10.7 µm in biological tissues, which were close to the theoretical values of 13.4 µm in air and 9.7 µm in tissue, respectively. The lateral resolution was 39.8 µm, and the theoretical value was 32.3 µm. We also examined the characteristics of OCT with a tunable quasi-SC.
Figure 6(a) shows the imaging sensitivity and axial resolution in terms of tunable modulation frequency. To achieve effective OCT imaging, we have to set the modulation frequency to more than 100 kHz because the scanning speed of the detection camera was 47 k lines/s. We varied the modulation frequency from 100 kHz to 10 MHz. The imaging sensitivity and axial resolution remained stable with tunable modulation frequency. This result, indicating that the SD-OCT was not sensitive to the modulation frequency, was similar to TD-OCT for modulation frequencies beyond several hundred kHz [18]. When the modulation frequency was sufficiently higher than the scanning rate of the line camera, the OCT imaging characteristics were stable. Figure 6(b) shows the imaging sensitivity and axial resolution in terms of tunable quasi-SC bandwidth. The imaging sensitivity was almost constant, but the axial resolution was improved as the bandwidth was increased, and this result was in good agreement with the theoretical value.
Fig. 6. (a) Imaging sensitivity and axial resolution in terms of modulation frequency at a bandwidth of 138 nm. (b) Imaging sensitivity and axial resolution in terms of tunable bandwidth of quasi-SC at a modulation frequency of 300 kHz.
2.3 Imaging results of SD-OCT
We performed cross-sectional OCT imaging of samples using the quasi-SC in the 1.7 µm spectral band. The OCT images consisted of 512 A-scans with 1024 pixels per scan. The final cross-sectional images were obtained by averaging three images, where the imaging time for one cross-sectional frame required about 0.05 s. Figure 7 shows the cross-sectional images of a tape stack, pig thyroid, and in vivo human fingernail. About 20 or more layers were observed in the tape stack, as shown in Fig. 7(a). The gland structure of the pig thyroid was clearly visible in Fig. 7(b). The biological tissues in the fingernail, such as the epidermis, dermis, nail plate and nail bed, can be clearly observed in Fig. 7(c).
We compared the imaging results obtained using 1.7 µm conventional coherent SC source [27]. The incident power for OCT was 35.6 mW. Figure 8(a) shows a photograph of the measured human tooth. Figures 8(b) and 8(c) show cross-sectional images of the human tooth obtained at shallow and deep regions. The enamel–dentine junction can be seen in both images. We set the middle position of a B-scan as the surface line and defined the penetration depth as the maximum depth at which we can distinguish a signal and background noise. The above two cross-sectional images were reconstructed by simple connection, as shown in Fig. 8(d). It is clear that the reconstructed image enables us to observe the tooth structure to 1.6 mm. Figure 8(e) shows a similar image by SC source obtained in the same way. The penetration depth is 1.4 mm. The result demonstrates the superiority in penetration depth compared to the reported 0.8 mm depth for 1.3 µm and 1.2 mm for 1.7 um wavelengths in TD-OCT [23].
Fig. 8. (a) Photograph of human tooth sample. Cross-sectional images at (b) shallow and (c)deep regions by quasi-SC. Connected OCT image by (d) quasi-SC and (e) conventional coherent SC.
Table 1 summarizes the detailed information of the 1.7 µm quasi-SC and SC sources. The two sources exhibited similar imaging characteristics in terms of sensitivity, axial resolution, lateral resolution, and imaging depth. However, the compared images and imaging indexes indicate that the quasi-SC source is as capable of high-resolution deep tissue imaging for biological samples as the SC source.
Table 1. Comparison of light source specifications and characteristics for OCT.
3. Conclusion
We demonstrated a 1.7 µm quasi-SC source as a light source for SD-OCT for the first time. The quasi-SC, based on a wavelength-tunable ultrashort soliton pulse, was controlled by fast intensity modulation. The tunable modulation frequency affected the quasi-SC spectrum only slightly, and therefore, the SD-OCT imaging was not sensitive to the modulation frequency. Tunable modulation amplitude and offset changed the central wavelength and bandwidth, which affected the axial resolution but not the imaging sensitivity. We successfully obtained high-resolution deep tissue cross-sectional images of tape stack, pig thyroid, human fingertip, and human teeth samples. We compared the imaging results obtained by our quasi-SC and a conventional SC source. The similar images suggested the usefulness of a 1.7 µm quasi-SC for SD-OCT imaging.
A quasi-SC with tunable central wavelength and bandwidth will make it possible to obtain uncorrelated speckle patterns on the same sample structure [28,29]. We are investigating low-speckle OCT imaging using quasi-SC [30]. It is expected that the 1.7 µm quasi-SC source will not only help us to realize high-resolution imaging in deep tissue regions, but also enable speckle reduction for OCT imaging.
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, and J. G. Fujimoto, “Optical coherence tomography,” Science 254(5035), 1178–1181 (1991). [CrossRef]
2. J. M. Schmitt, “Optical coherence tomography (OCT): a review,” IEEE J. Sel. Top. Quantum Electron. 5(4), 1205–1215 (1999). [CrossRef]
3. M. C. Pierce, J. Strasswimmer, B. H. Park, B. Cense, and J. F. De Boer, “Advances in optical coherence tomography imaging for dermatology,” J. Invest. Dermatol. 123(3), 458–463 (2004). [CrossRef]
4. A. Z. Freitas, D. M. Zezell, N. D. Vieira Jr, A. C. Ribeiro, and A. S. L. Gomes, “Imaging carious human dental tissue with optical coherence tomography,” J. Appl. Phys. 99(2), 024906 (2006). [CrossRef]
5. W. Drexler and J. G. Fujimoto, Optical coherence tomography: technology and applications, Vol. 2. Springer: Berlin, 2015.
6. W. Drexler, U. Morgner, F. X. Kärtner, C. Pitris, S. A. Boppart, X. D. Li, E. P. Ippen, and J. G. Fujimoto, “In vivo ultrahigh-resolution optical coherence tomography,” Opt. Lett. 24(17), 1221–1223 (1999). [CrossRef]
7. I. Hartl, X. D. Li, C. Chudoba, R. K. Ghanta, T. H. Ko, J. G. Fujimoto, and R. S. Windeler, “Ultrahigh-resolution optical coherence tomography using continuum generation in an air–silica microstructure optical fiber,” Opt. Lett. 26(9), 608–610 (2001). [CrossRef]
8. B. Povazay, K. Bizheva, A. Unterhuber, B. Hermann, H. Sattmann, A. F. Fercher, and E. Scherzer, “Submicrometer axial resolution optical coherence tomography,” Opt. Lett. 27(20), 1800–1802 (2002). [CrossRef]
9. Y. Wang, Y. Zhao, J. S. Nelson, Z. Chen, and R. S. Windeler, “Ultrahigh-resolution optical coherence tomography by broadband continuum generation from a photonic crystal fiber,” Opt. Lett. 28(3), 182–184 (2003). [CrossRef]
10. K. Bizheva, B. Považay, B. Hermann, H. Sattmann, W. Drexler, M. Mei, R. Holzwarth, T. Hoelzenbein, V. Wacheck, and H. Pehamberger, “Compact, broad-bandwidth fiber laser for sub-2-µm axial resolution optical coherence tomography in the 1300-nm wavelength region,” Opt. Lett. 28(9), 707–709 (2003). [CrossRef]
11. A. Unterhuber, B. Považay, K. Bizheva, B. Hermann, H. Sattmann, A. Stingl, and W. Drexler, “Advances in broad bandwidth light sources for ultrahigh resolution optical coherence tomography,” Phys. Med. Biol. 49(7), 1235–1246 (2004). [CrossRef]
12. N. Nishizawa, Y. Chen, P. Hsiung, E. P. Ippen, and J. G. Fujimoto, “Real-time, ultrahigh-resolution, optical coherence tomography with an all-fiber, femtosecond fiber laser continuum at 1.5 µm,” Opt. Lett. 29(24), 2846–2848 (2004). [CrossRef]
13. M. Nishiura, T. Kobayashi, M. Adachi, J. Nakanishi, T. Ueno, Y. Ito, and N. Nishizawa, “In vivo ultrahigh-resolution ophthalmic optical coherence tomography using Gaussian-shaped supercontinuum,” J. Appl. Phys. 49(1), 012701 (2010). [CrossRef]
14. T. Hori, J. Takayanagi, N. Nishizawa, and T. Goto, “Flatly broadened, wideband and low noise supercontinuum generation in highly nonlinear hybrid fiber,” Opt. Express 12(2), 317–324 (2004). [CrossRef]
15. N. Nishizawa and J. Takayanagi, “Octave spanning high-quality supercontinuum generation in all-fiber system,” J. Opt. Soc. Am. B 24(8), 1786–1792 (2007). [CrossRef]
16. S. Bourquin, A. D. Aguirre, I. Hartl, P. Hsiung, T. H. Ko, J. G. Fujimoto, and D. Kopf, “Ultrahigh resolution real time OCT imaging using a compact femtosecond Nd: Glass laser and nonlinear fiber,” Opt. Express 11(24), 3290–3297 (2003). [CrossRef]
17. B. E. Bouma, G. J. Tearney, I. P. Bilinsky, B. Golubovic, and J. G. Fujimoto, “Self-phase-modulated Kerr-lens mode-locked Cr: forsterite laser source for optical coherence tomography,” Opt. Lett. 21(22), 1839–1841 (1996). [CrossRef]
18. K. Sumimura, T. Ohta, and N. Nishizawa, “Quasi-super-continuum generation using ultrahigh-speed wavelength-tunable soliton pulses,” Opt. Lett. 33(24), 2892–2894 (2008). [CrossRef]
19. K. Sumimura, T. Ohta, and N. Nishizawa, “Quasi-supercontinuum generation using 1.06 µm ultrashort-pulse laser system for ultrahigh-resolution optical-coherence tomography,” Opt. Lett. 35(21), 3631–3633 (2010). [CrossRef]
20. N. Nishizawa and T. Goto, “Compact system of wavelength-tunable femtosecond soliton pulse generation using optical fibers,” IEEE Photonics Technol. Lett. 11(3), 325–327 (1999). [CrossRef]
21. N. Nishizawa, R. Okamura, and T. Goto, “Analysis of widely wavelength tunable femtosecond soliton pulse generation using optical fibers,” Jpn. J. Appl. Phys. 38(8R), 4768 (1999). [CrossRef]
22. V. M. Kodach, J. Kalkman, D. J. Faber, and T. G. Leeuwen, “Quantitative comparison of the OCT imaging depth at 1300 nm and 1600 nm,” Biomed. Opt. Express 1(1), 176–185 (2010). [CrossRef]
23. H. Kawagoe, S. Ishida, M. Aramaki, Y. Sakakibara, E. Omoda, H. Kataura, and N. Nishizawa, “Development of a high-power supercontinuum source in the 1.7 µm wavelength region for highly penetrative ultrahigh-resolution optical coherence tomography,” Biomed. Opt. Express 5(3), 932–943 (2014). [CrossRef]
24. N. Nishizawa, Y. Seno, K. Sumimura, Y. Sakakibara, E. Itoga, H. Kataura, and K. Itoh, “All-polarization-maintaining Er-doped ultrashort-pulse fiber laser using carbon nanotube saturable absorber,” Opt. Express 16(13), 9429–9435 (2008). [CrossRef]
25. Y. Senoo, N. Nishizawa, Y. Sakakibara, K. Sumimura, E. Itoga, H. Kataura, and K. Itoh, “Polarization maintaining, high-energy, wavelength-tunable, Er-doped ultrashort pulse fiber laser using carbon-nanotube polyimide film,” Opt. Express 17(22), 20233–20241 (2009). [CrossRef]
26. N. Nishizawa, R. Okamura, and T. Goto, “Simultaneous generation of wavelength tunable two-colored femtosecond soliton pulses using optical fibers,” IEEE Photonics Technol. Lett. 11(4), 421–423 (1999). [CrossRef]
27. H. Kawagoe, M. Yamanaka, S. Makita, Y. Yasuno, and N. Nishizawa, “Full-range ultrahigh-resolution spectral-domain optical coherence tomography in 1.7 µm wavelength region for deep-penetration and high-resolution imaging of turbid tissues,” Appl. Phys. Express 9(12), 127002 (2016). [CrossRef]
28. J. M. Schmitt, S. H. Xiang, and K. M. Yung, “Speckle in optical coherence tomography,” J. Biomed. Opt. 4(1), 95–105 (1999). [CrossRef]
29. M. Pircher, E. Götzinger, R. Leitgeb, A. F. Fercher, and C. K. Hitzenberger, “Speckle reduction in optical coherence tomography by frequency compounding,” J. Biomed. Opt. 8(3), 565–569 (2003). [CrossRef]
30. Y. Chen, M. Yamanaka, S. Kitajima, and N. Nishizawa, “Speckle Reduction by Frequency Compounding in 1.7 µm Optical Coherence Tomography using Tunable Quasi-Supercontinuum Laser Source,” in Conference on Lasers and Electro-Optics, Technical Digest Series (Optica Publishing Group, 2022), paper JW3B.152.
