Open Access
Add to BibSonomy
Abstract
Here, we demonstrate a polarized high-energy soliton synthesis technique for deep-brain 3-photon microscopy (3PM) excited at the 1700-nm window. Through coherent combining, we generate linearly polarized high-energy solitons whose energy is twice as high than those of each linearly polarized solitons. Due to the nonlinear origin of signals, both measured 3-photon fluorescence signal and third-harmonic signals are thus boosted by ~8 times in a tissue phantom. Using this technique, we further demonstrate 3PM of sulforhodamine 101 labeled vasculature 1600 μm in the mouse brain in vivo, which cannot be achieved by single-polarized soliton excitation.
© 2019 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
Multiphoton microscopy (MPM) has found widespread applications in biological, physiological and medical research [1–3]. As a nonlinear optical technique, MPM can be implemented in different modalities, such as 2/3/4-photon fluorescence, second/third/fourth-harmonic generation, etc [4–9]. MPM is well known for its deep-tissue penetration capability. Among the various modalities, 3-photon microscopy (3PM) currently enables the largest imaging depth, especially for animal models in vivo [2,9,10].
Deep-tissue 3PM necessitates excitation with high-energy femtosecond pulses, at a repetition rate in the range of sub-MHz [2] to MHz [5] from the considerations of both 3-photon signal level and average power [11]. In terms of excitation wavelength, two excitation windows are typically used: the 1300-nm and the 1700-nm window. Theoretical calculation indicate that excitation at both these two windows suffers from less attenuation through the biological tissue especially the brain, compared with the commonly used 800-nm window [5]. Experimental results further justifies theoretical results in mouse brain imaging in vivo [2,5,9,10]. 3PM excited at the 1300-nm window efficiently exploits the mature green fluorescent protein products, and enables functional imaging deep in the mouse hippocampus [2]. In comparison, 3PM excited at the 1700-nm window features even smaller attenuation at the excitation wavelength upon propagation inside the biological tissue, and enables deeper brain structure imaging such as vasculature [5,9,10].
Currently, 3PM depth in the mouse brain in vivo is limited by signal depletion, rather than the fundamental signal-to-background ratio (SBR) limit [5,11–13]. From the laser source perspective, this is determined by the available laser pulse energy at a given repetition rate. For all the in vivo deep-brain 3PM excited at the 1700-nm window demonstrated so far, soliton pulses based on soliton self-frequency shift (SSFS) [11,14] in photonic-crystal (PC) rods, pumped by high-energy 1550-nm fiber lasers have been exclusively used [5,9,10,15]. This technique is easy to implement, robust and stable. PC rods are optical waveguides like optical fibers but cannot be bent. Their exceptionally large effective mode area (40~50 times larger than standard single-mode fibers) make them ideal for generating high-energy solitons at the 1700-nm window, since the soliton energy is proportional to the effective mode area according to the energy scaling law [11]. Commercially available PC rods specify a maximum core diameter of 100 μm, leading to a maximum soliton energy of ~110 nJ [9,16], thus limiting the maximum 3PM depth.
At first glance, resorting to a PC rod with an even larger core size might overcome this energy limit. However, besides its non-availability, this strategy will finally run into the limit set by optical damage to the PC rod [17]. Consequently, it is desirable to develop a technique to further boost the soliton energy, applicable to both currently available PC rods and potentially rods with even larger effective mode areas.
In the aim of further boosting soliton energies at the 1700-nm window for deep-brain MPM, here we demonstrate a technique termed polarized soliton synthesis, which effectively doubles the soliton energy. 3PM in both tissue phantom and mouse brain in vivo proves this technique is effective and promising for deep-tissue 3PM.
2. Principles and methods
2.1. The principle of polarized soliton synthesis
Our idea of polarized soliton synthesis is originated from polarized two-color soliton generation [18,19]. In a polarization-maintained (PM) fiber, when the polarization of the linearly polarized input pump pulse is at an (tunable) angle between the two main axes of the PM fiber, the pump pulse will be split into two orthogonal linearly polarized pump pulses. Each of them will propagate inside the fiber at a different group velocity, generating two orthogonal linearly polarized solitons at the output of the fiber through SSFS with temporal delay between them. The wavelength of each soliton is independently tunable, simply by adjusting the pump pulse energy along each main axis of the PM fiber, which can be implemented by power and polarization adjustment before coupling into the fiber. We can thus expect: when the two orthogonal linearly polarized solitons are of the same wavelength, pulse width, pulse energy, and readjusted to overlap temporally again, they will be synthesized into a single pulse, which is also linearly polarized but with a rotation of polarization angle by 45° regarding the main axis of the PM fiber. The resultant soliton energy will be doubled according to: Es = As2 = A⊥2 + A∥2 = E⊥ + E∥ = 2E⊥ = 2E∥, where Es, E⊥, and E∥ are pulse energies of the synthesized soliton, vertical linearly polarized (short for vertical hereafter) and horizontal linearly polarized (short for horizontal hereafter) solitons, respectively, and As, A⊥, and A∥ are the corresponding electric fields. Constants are omitted in this formula.
Our PC rod is not designed with PM structures, indicating that there is no fast or slow axis in the rod and consequently we cannot use the two-color soliton generation technique simply by adjusting the polarization angle of the pump at the input. However, due to the rigidity and straight layout of the rod, the pulse polarization is virtually maintained: no matter what the polarization angle of the input linearly polarization pulse is, it remains the same at the output. These facts about the PC rod imply the following strategy and procedures for polarized soliton synthesis, shown schematically in Fig. 1:
- (1) Pump pulse splitting. We split the linearly polarized pump pulse at the input into orthogonal linearly polarized pump pulse with optical delay.
- (2) Polarized soliton generation. Each linearly polarized pump pulse will generate a soliton of the same linear polarization, also delayed with each other. Adjust the pump energy in each polarization to make sure the generated solitons are of the same wavelength.
- (3) Soliton synthesis. At the output of the PC rod, we split the orthogonal linearly polarized solitons, and readjust their delay to temporally overlap them, which completes the soliton synthesis.
2.2. High-energy polarized soliton synthesis system
The experimental setup based on the above soliton synthesis principle is shown in Fig. 2. The pump source was a 1550-nm fiber laser (FLCPA-02CSZU, Calmar) with 1-MHz pulse repetition rate, 500-fs pulse width and linear polarization. Prior to coupling into the PC rod, the input polarization of the linearly polarized pump pulses were split into orthogonal linear polarization, using the combination of a half-wave plate (HWP, WPH05M-1550, Thorlabs), a pair of polarizing beamsplitter cubes (PBS, PBS104, Thorlabs), and a pair of silver-coated mirrors. The mirror pair were mounted on a translation stage (NanoMax 300, Thorlabs) for fine delay adjustment to temporally overlap the resultant solitons.
Fig. 2 Experimental setup for polarized soliton synthesis. HWP: half-wave plate, M: silver-coated mirror, PBS: polarizing beamsplitter cube, L1 and L2: focusing and collimating lens, LPF: long-pass filter.
The orthogonal linearly polarized pump pulses were then coupled into a 44-cm PC rod (SC-1500/100-Si-ROD, NKT Photonics) with a 100-μm core diameter. Each linearly polarized pump pulses propagate inside the rod independently, generating solitons of the same linear polarization. A 1575-nm long-pass filter (1575lpf, Omega Optical) at normal incidence was used to remove the residual. At the output of the rod, after collimating and long-pass filtering, a PBS pair similar to that at the input was used to readjust the optical delay between the orthogonal linearly polarized solitons, such that they overlap temporally. The PBS pair together with the two mirrors at the output of the PC rod were mounted in such a way that the plane comprising the beam path were rotated by 90° (Fig. 2). By doing this the delay introduced by the PBS and mirror pair after the PC rod will be opposite to that introduced by the PBS and mirror pair before the PC rod.
The spectra and pulse widths of the solitons generated were measured by a spectrometer (OSA203B, Thorlabs) and a home-built second-order interferometric autocorrelator, respectively.
2.3. MPM system
The generated solitons were then sent into a multiphoton microscope (MOM, Sutter) for 3PM. For deep-brain imaging, dual-channel detection was performed to record 3-photon fluorescence images and THG images simultaneously. A GaAsP PMT (H7422p-40, Hamamatsu) and a GaAs PMT (H7422p-50, Hamamatsu) were used for 3-photon fluorescence signal and THG signal detection, respectively. In front of the GaAsP PMT, a 630/92-nm band-pass filter (FF01-630/92-25, Semrock) and a 593-nm long-pass filter (FF01-593/LP-25, Semrock) were used, while in front of the GaAs PMT, a 535/50 band-pass filter (HQ535/50M-2P, Chroma) was used. A NA = 1.05 water-immersion objective lens with custom coating for the 1700-nm window (XLPLN25XWMP2-SP1700, Olympus) was used to deliver the excitation light onto the sample, and to collect the epi-propagated signal light. To prevent excessive absorption of the excitation light during traversing the immersion medium, D2O immersion was used [20]. Image acquisition and processing were performed using ScanImage (Vidrio Technologies) and ImageJ (NIH), respectively.
2.4. Sample preparation
We prepared a tissue phantom by mixing 1-μm diameter non-fluorescent beads (Polyscience) as scattering elements at a concentration of 5.4 × 109/mL and 500-μM sulforhodamine 101 (Sigma-Aldrich) in low-melting point agarose gel (0.5%, Sigma). The characteristic attenuation length was measured to be 340 μm, mimicking the attenuation property of the cortex.
Animal procedures were reviewed and approved by Shenzhen University. We used male Balb/C mice (26 g, 8 weeks old, Guangdong Medical Laboratory Animal Center) for imaging. Animals were prepared using the methods described in [5]. Craniotomies were performed centred at 2 mm posterior and 2 mm lateral to the Bregma point. A home-made metal piece was glued to the skull using dental cement. Before imaging, 200 μl of sulforhodamine 101 (3.3 mg/ml, dissolved at phosphate buffered saline) was retro-orbitally injected to label the vasculature.
3. Results and discussion
3.1. Characterization of solitons
First we characterized the generated solitons. At maximum output energy of the laser, horizontal and vertical solitons at a peak wavelength of 1613 nm could be generated simultaneously (Fig. 3(a)). The measured soliton spectra for these orthogonal linearly polarized solitons are almost identical. The corresponding measured interferometric autocorrelation traces are shown in Fig. 3(b). After deconvolution assuming a sech2 intensity profile typical of solitons, the pulse widths for the horizontal and vertical solitons are 82.6 fs and 83.3 fs, respectively. The measured optical powers are 75 mW and 76 mW, respectively, with corresponding pulse energies of 75 nJ and 76 nJ for the 1-MHz repetition rate. The similarity in measured spectra, pulse widths and optical powers for the orthogonal linearly polarized solitons paves the way for soliton synthesis.
Fig. 3 Measured soliton spectra (a) and corresponding interferometric autocorrelation traces (b) for the horizontal (green), vertical (blue), and synthesized (red) solitons.
Next we finely adjusted the optical delay between the pump pulses to temporally overlap the horizontal and vertical solitons. This was conveniently diagnosed through monitoring the 2-photon currents generated in the silicon detector in the interferometer. An analyzer was aligned to maximize transmittance of the synthesized soliton. Consequently, the transmitted vertical and horizontal solitons were reduced by half in energy (due to the 45° projection). The generated 2-photon current I2p is proportional to the square of the pulse energy E, leading to I2p,⊥ = (E⊥/2)2 = E⊥2/4, I2p,∥ = (E∥/2)2 = E∥2/4, and I2p,⊥ = I2p,∥, where I2p,⊥ and I2p,∥ are the 2-photon currents generated by the vertical and horizontal solitons, respectively, with constants omitted. When the horizontal and vertical solitons overlap in time, the resultant 2-photon current generated by the synthesized soliton is I2p,s = (2E⊥)2 = (2E∥)2 = 16I2p,⊥ = 16I2p,∥. Equivalently speaking, when the horizontal and vertical solitons overlap in time, the 2-photon current of the synthesized pulse is 16 times higher than that generated by each linearly polarized soliton (after the analyzer).
Using this diagnosis, we temporally overlap the horizontal and vertical solitons. The measured spectrum and interferometric traces of the synthesized pulse are shown in Figs. 3(a) and 3(b), respectively, resembling those for solitons of each linear polarization. The measured optical power for the synthesized pulse is 151 mW, exactly the sum of the horizontal and vertical solitons, as expected. And the deconvolved pulse width is 83.7 fs.
3.2. 3PM test in a tissue phantom
First we tested the applicability of this soliton synthesis technique in the prepared tissue phantom, mimicking the biological tissue [13,21]. We performed 3-photon fluorescence imaging 1200 μm below the surface of the cover glass/tissue phantom interface, with horizontal, vertical, and synthesized solitons, respectively. The corresponding images are shown in Fig. 4. We note that the reason why the non-fluorescent scattering beads can be seen in the fluorescence channel is probably due to the SR101 adsorption onto the beads. SeparateTHG experiment verified that there was no THG signal leaking into the fluorescence channel. In order to quantify the signal difference due to soliton synthesis, we measured mean signal level in the encircled area (excluding scattering beads) in each image. The measured 3-photon fluorescence signals (measured in counts) are 122, 124 and 932 counts for the horizontal, vertical and synthesized solitons, respectively. The corresponding measured optical powers after the objective lens are 29.0 mW, 28.7 mW and 57.7 mW for the horizontal, vertical and synthesized solitons, respectively. The calculated signal increase is 7.6 times, close to the theoretical expectation of 8 times. We also compared THG signals at the cover glass/tissue phantom interface at lowered power after the objective lens. The resultant signal increase is 7.7 times due to soliton synthesis compared with single-polarization soliton excitation. These results prove that our polarized soliton synthesis technique is readily applicable to 3PM in biological tissues.
Fig. 4 3-photon fluorescence imaging of the tissue phantom at a depth of 1200 μm below the cover glass/tissue phantom interface, excited with horizontal (a), vertical (b), and synthesized (c) solitons, respectively. The signal level in the encircled areas were analyzed and compared. Scale bar: 10 μm. Color bar applies to all three images.
3.3. Deep-brain 3PM in mouse in vivo
Next we performed comparative 3PM in the mouse brain in vivo. We have recently demonstrated that SR101 is also appropriate for labeling the vasculature for deep-brain 3-photon fluorescence imaging, which makes it a much cheaper alternative to dextran based fluorescent dyes such as Texas red (submitted). The problem with SR101 for vasculature labeling is that it will be cleared out of the body over time relatively fast [22,23], and as a result 3-photon fluorescence signals will get dimmer over time. So in our experiment, we didn’t take a whole 3D stack excited with horizontal, vertical and synthesized solitons. Instead, we took 2D images at or below the white matter for the purpose of comparison.
Figure 5 shows comparative 3-photon fluorescence and THG imaging results of the mouse brain in vivo, excited by horizontal, vertical and synthesized solitons as indicated in each image. In each image, both 3-photon fluorescence and THG images were acquired simultaneously through the dual-channel detection. The 3-photon fluorescence images reveal a span of SR101-labeled blood vessel, while the THG images reveal the white matter. From these images, we can clearly see that in comparison with single-polarization soliton excitation, synthesized soliton excitation boosts signal level in both 3-photon fluorescence and THG imaging. We further quantified 3-photon signal increase in both channels by calculating the mean signal levels in the same encircled areas in Fig. 5. The calculated results are summarized in Table 1, together with the optical powers on the surface of the brain. Despite the optical powers for both the horizontal and vertical solutions are virtually the same on the sample surface, their generated THG signals are different (by a factor of 29%). This is expected as THG signals excited by linearly polarized light is sensitive to the alignment of structures [24]. Consequently, the calculated THG signals due to soliton synthesis are increased by 6.9 and 9.7 times, in comparison with excitations with horizontal and vertical solitons, respectively. In contrast, the incoherent 3-photon fluorescence signals from SR101 are not sensitively dependent on the angle of linear polarization, so the 3-photon fluorescence signals excited by the horizontal and vertical solitons are differed by a factor of 10%. The calculated 3-photon fluorescence signals due to soliton synthesis are increased by 5.6 and 6.5 times, compared with excitations with horizontal and vertical solitons, respectively. We postulate that for 3-photon fluorescence signals, the deviation from the theoretical signal increase of 8 might be due to the blood flow, which makes the brightness of blood vessels nonuniform as can be seen from the images.
Fig. 5 3-photon fluorescence (red) and THG (green) imaging of the mouse brain in vivo at a depth of 940 μm below the surface of the brain, excited with horizontal, vertical and synthesized solitons as indicated. Fluorescence in the circles and THG in the squares were analyzed. Pixel size: 512x512; scale bar: 10 μm; frame rate: 4s/frame. Color bars apply to all three images.
Table 1. Summary of the soliton and imaging results for Fig. 5
The deep-brain imaging capability of our soliton synthesis technique is further illustrated by imaging the deepest structures. Figure 6 shows comparative 3PF imaging results ~1600 μm below the surface of the brain in the same mouse. No THG signals could be detected below the white matter, so only 3PF images were acquired and compared. Using either horizontal or vertical soliton excitation, no structure could be resolved. In sharp contrast, using synthesized soliton excitation, we can clearly resolve the SR101-labeld blood vessels at the same depth. This all-or-none contrast qualitatively and intuitively justifies the effectiveness of our soliton synthesis technique in deep-brain 3PM.
Fig. 6 3-photon fluorescence imaging of the mouse brain in vivo at a depth of 1596 μm to 1602 μm below the surface of the brain, excited with horizontal, vertical and synthesized solitons as indicated. Pixel size: 512x512; scale bar: 50 μm; frame rate: 8s/frame. Each image is an average z-projection of 4 frames. Color bar applies to all three images.
4. Conclusion and discussion
In the aim of boosting signal level and imaging depth in deep-brain 3PM in vivo, here we demonstrate a polarized soliton synthesis technique. This technique overcomes the energy limit set by the PC rod, leading to twice increase in soliton energy in comparison with single-polarized solitons. 3PF and THG imaging in both the tissue phantom and the mouse brain in vivo show the effectiveness of this technique, in terms of both 3-photon signal level and in vivo brain imaging depth.
In this demonstration our synthesized soliton wavelength is limited to 1613 nm, while the 1700-nm window refers to a broader range (roughly 1600 nm to 1840 nm [5,25]) and it has been demonstrated that longer wavelengths within this window can also be used for 3PM [5,10,26]. This wavelength limit is set by the maximum available energy from the 1550-nm pump laser so far. With a more energetic pump laser, it can be expected our polarized soliton synthesis technique will be applicable to generating energetic pulses covering the entire 1700-nm window.
The concern for using this energetic laser as the excitation source for brain 3PM is photodamage. In our in vivo experiment, the maximum optical power and corresponding pulse energy were 61.6 mW and 61.6 nJ (considering the 1-MHz repetition rate). However, they were only used when imaging the deep structures (in the white matter and the hippocampus, ≥940 μm in depth). We performed repetitive imaging without seeing any structural damage to the white matter and the structures below. According to our experience, if there is photodamage incurred by the high-energy excitation pulse to the brain (e.g., depositing all the optical power/pulse energy to the brain while focusing within and close to the brain surface), the imaging depth will be instantaneously and dramatically degraded. These two criteria (i.e., large imaging depth and no observable structural damage) lead us to conclude there is no photodamage for such high power/pulse energy when focusing 940 μm into the brain. It is also worth mentioning that using a 0.8-MHz energetic laser system [10], Wang et al didn’t observe any damage to the brain with 70-mW optical power (corresponding to 87.5-nJ pulse energy) deposited on the sample surface. Despite they used a different excitation wavelength (1320 nm), their results are in agreement with our results that when imaging deep, ~60 mW power deposited on the surface would not cause photodamage.
Funding
National Natural Science Foundation of China (NSFC) (61775143); Key Project of Department of Education of Guangdong Province (2017KZDXM073); China Postdoctoral Science Foundation (2019M653025).
References
1. W. Denk, J. H. Strickler, and W. W. Webb, “Two-photon laser scanning fluorescence microscopy,” Science 248(4951), 73–76 (1990). [CrossRef] [PubMed]
2. D. G. Ouzounov, T. Wang, M. Wang, D. D. Feng, N. G. Horton, J. C. Cruz-Hernández, Y. T. Cheng, J. Reimer, A. S. Tolias, N. Nishimura, and C. Xu, “In vivo three-photon imaging of activity of GCaMP6-labeled neurons deep in intact mouse brain,” Nat. Methods 14(4), 388–390 (2017). [CrossRef] [PubMed]
3. S. Liebscher and M. Meyer-Luehmann, “A peephole into the brain: neuropathological features of Alzheimer’s disease revealed by in vivo two-photon imaging,” Front. Psychiatry 3, 26 (2012). [CrossRef] [PubMed]
4. W. R. Zipfel, R. M. Williams, and W. W. Webb, “Nonlinear magic: multiphoton microscopy in the biosciences,” Nat. Biotechnol. 21(11), 1369–1377 (2003). [CrossRef] [PubMed]
5. N. G. Horton, K. Wang, D. Kobat, C. G. Clark, F. W. Wise, C. B. Schaffer, and C. Xu, “In vivo three-photon microscopy of subcortical structures within an intact mouse brain,” Nat. Photonics 7(3), 205–209 (2013). [CrossRef] [PubMed]
6. M. J. Farrar, F. W. Wise, J. R. Fetcho, and C. B. Schaffer, “In vivo imaging of myelin in the vertebrate central nervous system using third harmonic generation microscopy,” Biophys. J. 100(5), 1362–1371 (2011). [CrossRef] [PubMed]
7. D. Débarre, W. Supatto, A. M. Pena, A. Fabre, T. Tordjmann, L. Combettes, M. C. Schanne-Klein, and E. Beaurepaire, “Imaging lipid bodies in cells and tissues using third-harmonic generation microscopy,” Nat. Methods 3(1), 47–53 (2006). [CrossRef] [PubMed]
8. L. C. Cheng, N. G. Horton, K. Wang, S. J. Chen, and C. Xu, “Measurements of multiphoton action cross sections for multiphoton microscopy,” Biomed. Opt. Express 5(10), 3427–3433 (2014). [CrossRef] [PubMed]
9. W. Wen, Y. Wang, H. Liu, K. Wang, P. Qiu, and K. Wang, “Comparison of higher-order multiphoton signal generation and collection at the 1700-nm window based on transmittance measurement of objective lenses,” J. Biophotonics 11(1), e201700121 (2018). [CrossRef] [PubMed]
10. T. Wang, D. G. Ouzounov, C. Wu, N. G. Horton, B. Zhang, C. H. Wu, Y. Zhang, M. J. Schnitzer, and C. Xu, “Three-photon imaging of mouse brain structure and function through the intact skull,” Nat. Methods 15(10), 789–792 (2018). [CrossRef] [PubMed]
11. K. Wang, N. Horton, K. Charan, and C. Xu, “Advanced Fiber Soliton Sources for Nonlinear Deep Tissue Imaging in Biophotonics,” IEEE J. Sel. Top. Quant. 20(2), 6800311 (2014). [CrossRef]
12. P. Theer and W. Denk, “On the fundamental imaging-depth limit in two-photon microscopy,” J. Opt. Soc. Am. A 23(12), 3139–3149 (2006). [CrossRef] [PubMed]
13. D. Kobat, M. E. Durst, N. Nishimura, A. W. Wong, C. B. Schaffer, and C. Xu, “Deep tissue multiphoton microscopy using longer wavelength excitation,” Opt. Express 17(16), 13354–13364 (2009). [CrossRef] [PubMed]
14. J. P. Gordon, “Theory of the soliton self-frequency shift,” Opt. Lett. 11(10), 662–664 (1986). [CrossRef] [PubMed]
15. K. Wang, Y. Du, H. Liu, M. Gan, S. Tong, W. Wen, Z. Zhuang, and P. Qiu, “Visualizing the “sandwich” structure of osteocytes in their native environment deep in bone in vivo,” J. Biophotonics 12(4), e201800360 (2019). [CrossRef] [PubMed]
16. D. Sinefeld, H. P. Paudel, D. G. Ouzounov, T. G. Bifano, and C. Xu, “Adaptive optics in multiphoton microscopy: comparison of two, three and four photon fluorescence,” Opt. Express 23(24), 31472–31483 (2015). [CrossRef] [PubMed]
17. C. J. Saraceno, O. H. Heckl, C. R. Baer, T. Südmeyer, and U. Keller, “Pulse compression of a high-power thin disk laser using rod-type fiber amplifiers,” Opt. Express 19(2), 1395–1407 (2011). [CrossRef] [PubMed]
18. N. Nishizawa, R. Okamura, and T. Goto, “Simultaneous generation of wavelength tunable two-colored femtosecond soliton pulses using optical fibers,” IEEE Photonic. Tech. L. 11(4), 421–423 (1999). [CrossRef]
19. Y. Du, Z. Zhuang, J. He, H. Liu, P. Qiu, and K. Wang, “Self-referenced axial chromatic dispersion measurement in multiphoton microscopy through 2-color third-harmonic generation imaging,” J. Biophotonics 11(9), e201800071 (2018). [CrossRef] [PubMed]
20. Y. X. Wang, W. H. Wen, K. Wang, P. Zhai, P. Qiu, and K. Wang, “Measurement of absorption spectrum of deuterium oxide (D2O) and its application to signal enhancement in multiphoton microscopy at the 1700-nm window,” Appl. Phys. Lett. 108(2), 021112 (2016). [CrossRef]
21. P. Theer, M. T. Hasan, and W. Denk, “Two-photon imaging to a depth of 1000 microm in living brains by use of a Ti:Al2O3 regenerative amplifier,” Opt. Lett. 28(12), 1022–1024 (2003). [CrossRef] [PubMed]
22. A. Nimmerjahn, F. Kirchhoff, J. N. D. Kerr, and F. Helmchen, “Sulforhodamine 101 as a specific marker of astroglia in the neocortex in vivo,” Nat. Methods 1(1), 31–37 (2004). [CrossRef] [PubMed]
23. F. Appaix, S. Girod, S. Boisseau, J. Römer, J. C. Vial, M. Albrieux, M. Maurin, A. Depaulis, I. Guillemain, and B. van der Sanden, “Specific in vivo staining of astrocytes in the whole brain after intravenous injection of sulforhodamine dyes,” PLoS One 7(4), e35169 (2012). [CrossRef] [PubMed]
24. D. Oron, D. Yelin, E. Tal, S. Raz, R. Fachima, and Y. Silberberg, “Depth-resolved structural imaging by third-harmonic generation microscopy,” J. Struct. Biol. 147(1), 3–11 (2004). [CrossRef] [PubMed]
25. H. Liu, J. Wang, X. Peng, Z. Zhuang, P. Qiu, and K. Wang, “Ex and in vivo characterization of the wavelength-dependent 3-photon action cross-sections of red fluorescent proteins covering the 1700-nm window,” J. Biophotonics 11(7), e201700351 (2018). [CrossRef] [PubMed]
26. K. Wang, W. Wen, H. Liu, Y. Du, Z. Zhuang, and P. Qiu, “Transmittance Characterization of Objective Lenses Covering all Four Near Infrared Optical Windows and its Application to Three-Photon Microscopy Excited at 1820 nm,” IEEE Photonics J. 10(3), 3900607 (2018). [CrossRef]
