Open Access
Add to BibSonomyAbstract
We demonstrate that high-quality images of the deep regions of a thick sample can be obtained from its surface by multi-focal multiphoton microscopy (MMM). The MMM system incorporates a spatial light modulator to separate the excitation beam into a multi-focal excitation beam and modulate the pre-distortion wavefront to correct spherical aberration (SA) caused by a refractive index mismatch between the immersion medium and the biological sample. When fluorescent beads in transparent epoxy resin were observed using four SA-corrected focal beams, the fluorescence signal of the obtained images was ~52 times higher than that obtained without SA correction until a depth of ~1100 μm, similar to the result for single-focal multiphoton microscopy (SMM). The MMM scanning time was four times less than that for SMM, and MMM showed an improved fluorescence intensity and depth resolution for an image of blood vessels in the brain of a mouse stained with a fluorescent dye.
© 2017 Optical Society of America
1. Introduction
Three-dimensional imaging using multiphoton laser scanning fluorescence microscopy (MLSM) is useful for visualizing and understanding biological structures [1, 2]. Aiming for further increase in the quality of observation, three-dimensional imaging of vessels and neurons from the surface of a thick biological sample has been performed [3, 4]. A three-dimensional image is formed from multiple two-dimensional images acquired at different depths. Generally, in MLSM, a two-dimensional image is acquired by a galvo scanning system, which scans the excitation beam in the direction perpendicular to the optical axis (the x-direction and the y-direction). After the excitation beams is scanned at a certain depth, the objective lens is moved in the optical-axis direction (the z direction) and a two-dimensional image at a different depth is acquired. Thousands of laser scans are needed for a thick sample to be observed with a high depth resolution and wide range. Therefore, the measurement sometimes takes several hours.
Some methods have been reported in an effort to shorten the measurement time. One of them involves replacing the galvo scanner with a high-speed scanner, such as a polygon scanner or a resonant scanner mirror. Another method is multi-focal multiphoton microscopy (MMM). In addition, the methods using an ultrasonic lens [5], Bessels beam imaging [6], and passive pulse splitting [7] have been studied. Adopting a faster scanner [8,9] does not require major changes to the optical system compared to MMM. However, the excitation beam must of the system adopting the faster scanner have a high power to obtain sufficient information from the observation point because the beam exposure of the system duration is shorter than that of the system adopting a galvo scanner and the fluorescence is weaker. Thus, the excitation beam is made stronger, causing fluorescence intensity saturation due to the absorption saturation of fluorescence. As a result, photodamage of the specimen may occur. On the other hand, in MMM, multi-point scanning with a multi-focal excitation beam is performed using a device that divides the excitation beam into multi-focal beams. A fluorescent signal is excited by each of the divided beams, and is detected by each element of a detector array such as a multi-anode photo-multiplier tube (mPMT). If four beams in a row are used to excite the specimen, the scanning area of each beam becomes a quarter of the scanning area for single multiphoton microscopy (SMM). Therefore, when the intensity of each focal beam is the same as that in SMM, the scan time for MMM is four times less than that for SMM, without changing the laser exposure duration. Since the measurement time of SMM for acquiring multiple two-dimensional images is several hours, it is effective to reduce the required time by even a quarter of the original. Moreover, if photodamage does not occur, further reduction in the measurement time by combining MMM and the other methods such as a high-speed scanner can be expected.
Several devices are suggested to generate a multi-focal excitation beam: a microlens array [10, 11], a diffractive optical element [12–15], a beam splitter [16], and a spatial light modulator (SLM) [17–22]. The SLM has an advantage in that it can adaptively excite an arbitrary number of multi-focal beams with arbitrary positions and intensities by applying an appropriate computer generated hologram (CGH). Previously, we proposed a method for exciting a uniform fluorescence pattern using MMM with a CMOS image sensor [22].
Moreover, the SLM is also useful as a device for aberration correction [23–28] because the SLM can control the wavefront of the excitation beam by changing a CGH applied to the SLM. Generally, in MLSM, an objective lens using immersion fluid is used for reducing aberration caused by a refractive index (RI) mismatch at the surface of the biological sample. However, the aberration caused by the RI mismatch is small but not negligible in the deep region. Additionally, when living sample like mouse is observed, preparations are necessary for attaching a biological sample to a glass bottom dish in which the immersion fluid should be filled. The weight of the fluid and the dish may induce stress on the sample. If oil is used for immersion, contamination may occur. Observation using a dry objective lens can be performed via non-contact means, which can realize observation without preparations. However, aberration caused by an RI mismatch between air and the average RI of the biological sample is serious. In particular, the RI mismatches result in spherical aberration (SA), and the SA increases as the observation depth increases. Recently, a method for SA correction using numerically calculated pre-distortion wavefront has been developed [26–28].
In this study, in order to observe deep regions of the sample from the surface in a short time and with less preparation, an SLM is incorporated in MLSM, and MMM scanning is performed with a dry objective lens. The SLM generates SA-corrected multi-focal excitation beams. To the best of our knowledge, this is the first study to construct a three-dimensional image with a depth of over 750 μm using MMM. We also verify the efficacy of the MMM method as compared to SMM by performing both SMM and MMM with and without SA correction, by simply changing the CGH. We demonstrate the observation of fluorescent beads in transparent epoxy resin from the surface to a depth of over 1000 μm using a dry objective lens. Furthermore, we observe the structure of a biological sample with a high fluorescence intensity and high depth resolution, even with MMM.
2. Experimental setup
Figure 1 is a simplified schematic showing the setup of our experimental MLSM system using an SLM. As shown in the figure, a Ti:sapphire laser (Chameleon vision II, Coherent Inc.) is used to deliver a horizontally polarized beam to a beam expander. A femtosecond train of an optical pulse (880 nm wavelength, 150 fs pulse duration, 80MHz repetition rate) is projected onto an LCOS-SLM (1272 × 1024 pixels, 12.5 μm pixel pitch, Hamamatsu Photonics K.K.) with a Peltier system [29, 30]. To enhance the reflectivity, the SLM includes a dielectric multilayer mirror with a bandwidth of 700 to 1050 nm. According to the CGH applied to the SLM, the wavefront of the excitation beam is modulated to separate the multi-focal beam, and the wavefront of these beams is also modulated to a pre-distortion wavefront for correcting the SA. The method of designing the CGH is explained in the following section. In addition, the CGH is superimposed with a correction pattern for the distortion of the SLM [30]. The beam is reflected and directed through one telecentric lens system to an x-y galvo scanning system (6220H, Cambridge Technology), which is coupled to a second telecentric lens system. The beam, with its incident angle varied using the scanner, is then directed to a dry objective lens (UCPLFLN 20× magnification, NA = 0.7, 1800 μm working distance, Olympus) by a third telecentric lens system. These telecentric lens systems are used to ensure that the wavefront of the beam is transmitted from the plane of the SLM to the pupil plane of the objective lens in an upright microscope system. By applying the pre-distortion pattern, the wavefront of the multi-focal excitation beam is distorted before it is incident on the interface between air and the sample. The SA due to RI mismatch is canceled inside the sample by pre-distortion. The multi-focal beams are focused on the sample using the dry objective lens in order to excite multiple fluorescence from the focal spots, which is then collected by the objective lens. Since the collected multiple fluorescences are descanned by the x-y galvo system and the focus lens to a GaAsP multi-anode photo-multiplier tube (mPMT: H12311, Hamamatsu Photonics K.K.), the fluorescence is always focused on each anode of the mPMT by the telecentric lens system and the galvo system. In the case of the observation of the deep region of the sample by the scattering caused by the medium on the path, the fluorescence spreads, resulting in an image with a ghost image [13, 14]. To reduce the influence of the spread of the fluorescence, the fluorescence is detected by every two anodes of the mPMT (Fig. 1), i.e., a gap is constructed so that the multiple fluorescences will not overlap. The two-dimensional image (x-y image) is formed from the multiple fluorescence detected by the mPMT. To obtain the three-dimensional image, scans are performed at each depth by moving the objective lens.
Fig. 1 Schematic of the experimental MLSM system using an SLM. The solid line (red) and dashed lines (green) represent the excitation beams and fluorescence, respectively. By changing the CGH applied to the SLM, the system can perform four types of scans: MMM with SA correction, MMM without SA correction, SMM with SA correction, and SMM without SA correction. To reduce the influence of the spread of the fluorescence, the multiple fluorescence are detected by every two anodes of the mPMT. The blue double-head arrow indicates the polarization direction of the excitation beam.
3. Method of designing a CGH for MMM with spherical aberration correction
Here, we introduce the method for generating multi-focal excitation beams with SA correction using the SLM. The CGHs for generating multi-focal excitation beams and correcting SA are designed separately and then superimposed. The CGH ϕ1 for generating multi-focal excitation beams is designed using the modified iterative Fourier transform algorithm (IFTA) [22]. We incorporated an experimental weight into the IFTA to reflect the experimental result such that four excitation beams of uniform intensity were generated. Here, we briefly outline the design procedure for the CGH for generating multi-focal excitation beams:
- Design of the CGH using a weighted IFTA in the manner of an overcompensation (OC) method [31].
- Generation of multi-focal excitation beams by applying the CGH, and generation of multiple fluorescence by irradiating fluorescent acrylic plate with the multi-focal excitation beams.
- Acquisition of a snapshot of the multiple fluorescence and evaluation of the intensity distribution of the fluorescence.
- Calculation of an experimental weight from the intensity distribution of the fluorescence.
- Redesign of the CGH using an IFTA incorporating the experimental weight.
We generated four excitation beams in the slow axis of the raster scan, as shown in Fig. 2(a), and shortened the measurement time by narrowing the movement range of the galvo system. The advantage of our proposed MLSM system is that the system can perform a scan with a single focal excitation beam or with multi-focal excitation beams simply by switching the CGH. When a single focal excitation beam is scanned, the system performs SMM. However, when multi-focal excitation beams are scanned, the system performs MMM. Figures 2(c) and 2(d) show the x-y fluorescent polystyrene beads in transparent epoxy resin. Since the fluorescence intensity in the MMM image should be the same as that in the SMM image near the surface, the intensity distribution of the multi-focal excitation beam for MMM can be improved by comparing the SMM and MMM images. That is, the CGH is redesigned so that the MMM image is equivalent to the SMM image. Here, we investigated the PSF using the fluorescence polystyrene bead (size was 0.2 μm) in transparent epoxy resin. With a 0.7-NA dry objective lens, the average lateral FWHM of the PSF was 1.0 μm and the average axial FWHM of the PSF was 3.9 μm when the MMM scan with four excitation beams was performed. On the other hand, the lateral FWHM of the PSF was 1.0 μm and the axial FWHM of the PSF was 4.0 μm when the SMM scan was performed.
Fig. 2 The scanner performed a raster scan with: (a) four excitation beams and (b) a single excitation beams. The system performed MMM when four excitation beams are scanned, and SMM when a single excitation beam is scanned. The resulting x-y images observed using (c) MMM and (d) SMM. The red arrows indicate the boundaries of the scanning areas. The scale bar indicates 20 μm.
The CGH ϕ2 for correcting SA caused by an RI mismatch between air and the averaged RI of the sample is calculated numerically using the method described in [28]. By applying the CGH, the wavefront of the excitation beam, which we call the pre-distortion wavefront, is distorted before the beam is incident on the interface between air and the sample. The SA due to the RI mismatch is canceled inside the sample and the excitation beam is focused without the influence of the RI mismatch. When a dry objective lens with a numerical aperture ( moves to the inside of the sample (at a distance d from the interface), the pre-distortion wavefront for correcting the RI mismatch is expressed as follows:
where λ is the wavelength of the excitation beam, ρ is the normalized pupil radius, η is the factor for changing the depth of the focal spot, and n1 and n2 are the RIs of air and the sample, respectively. Since a glass and fluid are not used, observation using a dry objective lens is performed taking only the average RI of the sample into account. The pre-distortion wavefront generated by the CGH ϕ2 is the inverse of the wavefront influenced by aberration. After designing the two CGHs ϕ1 and ϕ2, they are superimposed. The combined CGH is then applied to the SLM, simultaneously realizing the division of the excitation beam and the correction of the SA. In this experiment, the CGH ϕ1 was designed only once before the measurement, while the CGH ϕ2 was designed depending on the observation depth during the measurement.4. Experimental results and discussion
4.1. Fluorescent polystyrene beads of 3 μm in diameter in transparent epoxy resin
We observed 3μm-diameter fluorescent polystyrene beads in transparent epoxy resin using a dry objective lens. We constructed medium 2 (transparent epoxy resin, n2=1.59) such that the interface between medium 1 (air, n1=1) and medium 2 was almost flat. As a result, the SA caused by the RI mismatch between air and the transparent epoxy resin was dominant. We carried out scans from the surface of the transparent epoxy resin to an optical depth of 1100 μm with the objective lens positioned at 0.75 μm increments. At each depth, we performed scans in the following order: SMM scan without SA correction, SMM scan with SA correction, MMM scan without SA correction, and MMM scan with SA correction. In the case of the scans with SA correction, the CGH for SA correction was calculated and the wavefront of the excitation beam was modulated every 1.5 μm that the objective lens was moved. In MMM, the excitation beam was separated into four-focal excitation beams by the SLM. Here, the intensity of the excitation beam was changed depending on the observation depth in order to acquire images with high signal-to-noise ratios in the deeper regions, as shown in Fig. 3. Compared with SMM, the total intensity of the excitation beam for MMM was 4.6 times higher due to the loss of excitation beam caused by the CGH.
Fig. 3 Intensity of the excitation beam for MMM and SMM. To clarify the effect of the SA correction in the deeper regions, the intensity of the excitation beam was changed depending on the observation depth. The intensity of the excitation beam was measured under the objective lens.
By scanning, we acquired 1000 x-y images, and constructed a three-dimensional image from these images. Figures 4(a) and 4(b) show Y Z projected images obtained using the maximum fluorescence intensity [3] of the fluorescent beads in transparent epoxy resin from an optical depth of −35 μm to 1100 μm, when MMM with and without SA correction was performed. For comparison, the Y Z projected images for SMM scans with/without SA correction are shown in Figs. 4(c) and 4(d). The brightness of each image was normalized using the maximum fluorescence intensity of the beads from the MMM scan with SA correction. Figures 4(b) and 4(d) clearly show that the fluorescence intensity of the beads observed near the surface was highest. In addition, the fluorescence intensity of the beads in Figs. 4(b) and 4(d) decreased as the observed depth increased; fluorescence was hardly observed at optical depths greater than 200 μm. On the other hand, with SA correction, beads were still observed at a 1100 μm optical depth in Figs. 4(a) and 4(c).
Fig. 4 Results of observations of the fluorescent polystyrene beads in transparent epoxy resin with a dry objective lens (NA = 0.7). (a), (b) Y Z projected images for an optical depth of −35 μm to 1100 μm from MMM scans performed with and without SA correction, respectively. (c), (d) Y Z projected images from SMM scans performed with and without SA correction, respectively. (e)–(p) Magnified Y Z projected images for optical depths of (e)–(h) −35 μm to 100 μm, (i)–(l) 450 μm to 550 μm, and (m)–(p) 950 μm to 1050 μm. The red arrows indicate the boundary of the scanning area. The scale bar indicates 20 μm.
Figures 4(e)–4(h) show magnified Y Z projected images of Figs. 4(a)–4(d), respectively, from −35 μm to 100 μm. Similarly, magnified Y Z projected images from 450 to 550 μm, and from 950 μm to 1050 μm are shown in Figs. 4(i)–4(p). The brightness of each image was normalized by the respective maximum fluorescence intensity. In the case of MMM and SMM without SA correction, the focal spot of the excitation beam is elongated in the optical-axis direction by the SA, and its power density decreases. As a result, the observed beads are elongated in the direction of their optical axes (the z direction), and as the observation depth increased, the fluorescence intensity of the beads decreased. In the cases with SA correction, the elongation of the observed beads was improved in the MMM scans as well as in the SMM scans. Thus, the fluorescent beads in deeper regions were observable by both MMM and SMM with SA correction. In the magnified figures of the MMM scan, the boundaries of the scanning areas are indicated by red arrows. Owing to the generation of a uniform intensity distribution of the multiple excitation beams, an image without block noise and shading was acquired. In MMM, it is known that a ghost image caused by the spread of the fluorescence appears in the deep region [13,14]. In our experiment, multiple fluorescence is detected by every two anodes of the mPMT (Fig. 1), thus decreasing the influence of the spread of the fluorescence. However, dim ghost bead images were observed in Figs. 4(i) and 4(m). It is possible to further reduce the ghost bead images using a method suggested in previous studies [13,14].
Next, we quantitatively examined the effect of SA correction. Figure 5 shows the improvement ratio of the fluorescence intensity with SA correction to that without SA correction. The improvement ratio Ratioexperiment can be expressed as
where Qwith SA correction, Qwithout SA correction are the averages of the fluorescence intensities in a region-of-interest (ROI) (the central coordinates is (α, β, γ)) in a bead when the MMM scan with and without SA correction is performed at the observation depth of γ μm. Here, the ROI was chosen as a circular area (radius is 6 pixel) on the two-dimensional image around the center of gravity of each bead. The size of the ROI corresponds to the size of the typical bead (3 μm), and the size of the observed beads near the surface of the epoxy resin was almost 3 μm in x-y direction. The ratio near the surface of the resin is small because the SA effect was negligible. As the SA effect at greater depths is large, the ratio (or the effect of the SA correction) becomes large. The maximum fluorescence intensity for MMM with SA correction was approximately 52 times higher than that without SA correction at an optical depth of approximately 1100 μm. This result is equivalent to the result for SMM with SA correction (49 times). The simulation result is shown in Fig. 5. When the beam was completely corrected by the SLM, the ratio was calculated as follows: where A denotes the amplitude distribution of the excitation beam, ϕflat is the uniform phase distribution, and ϕaberration is the phase distribution affected by SA. DESLM was used to include the diffraction efficiency of the SLM. Generally, the CGH for the SLM has a periodic structure because it is wrapped in order to express more than one lambda. The period of the wrapping is spatially varied, and the diffraction efficiency varies depending on the period. For example, a high spatial frequency causes the diffraction efficiency to be lower. In other words, the diffraction efficiency of the SLM is spatially varied, resulting in unintended amplitude modulation.Fig. 5 (a) Ratio of the fluorescence intensity with SA correction to that without SA correction. (b) Ratio of the observed beads length (FWHM) with SA correction to that without SA correction.
The largest value of the improvement ratio of the experimental results is almost the same as that of the simulation results up to an optical depth of 1100 μm. However, there are variations in the ratio for optical depths over 500 μm. We consider that other aberrations, such as astigmatism and coma, distorted the wavefront of the excitation beam for optical depths over 500 μm because the proposed correction method was corrected for SA. For example, at the left end of Fig. 4(m), the observed beads were slightly deformed from the elliptical shape to the shape like half-moon. It is known that such aberrations are caused by coma and astigmatism [32], and we consider that the aberrations were caused by the imperfectly flat surface of the epoxy resin.
We defined the length of the beads as the range across which the intensity of the observed beads is greater than the FWHM of the maximum intensity. Figure 5(b) shows the improvement in depth resolution as the ratio of the length of the observed fluorescent beads with SA correction to that without SA correction. In the case of MMM, at a 6 μm depth, the beads with and without SA correction are 4.7 μm and 5.0 μm, respectively. The lengths of the beads with and without SA correction are 5.0 μm and 21.5 μm at an optical depth of approximately 1084 μm, respectively. The length of the observed fluorescent bead with SA correction is approximately 4.3 times less than that without SA correction at an optical depth of approximately 1084 μm. This result is equivalent to the result for SMM with SA correction (4.0 times).
5. Imaging of blood vessels of a mouse brain stained with a fluorescent dye
We observed the blood vessels from the cerebrum of a C57BL6J mouse stained with DiI, as previously reported [33]. All animal experiments were performed under the approval of the Institutional Animal Care and Use Committees of Hamamatsu University, School of Medicine. After the fluorescent dye was injected, the optical transparency of the brain was enhanced by SeeDB (n2=1.48) [34]. The observation area was a relatively flat area of the cerebrum of the brain without a cover slip. We carried out MMM and SMM scans with a dry objective lens positioned at 2.0 μm increments. At each depth, we performed scans in the following order: SMM scan without SA correction, SMM scan with SA correction, MMM scan without SA correction, and MMM scan with SA correction. In the case of the scans with SA correction, the CGH for SA correction was calculated, and the wavefront of the excitation beam was modulated with every movement of the objective lens by 2.0 μm. The measurement time for MMM was 4.0 times shorter than that for SMM.
Figures 6(a) and 6(b) show the Y Z projected images from an optical depth of 0 μm to 754 μm from the MMM scans with and without SA correction, respectively. The Y Z projected images from the SMM scans with/without SA correction are shown in Figs. 6(c) and 6(d). The brightness of each image was normalized using the maximum fluorescence intensity of Fig. 6(a). Figures 6(e)–6(h) show magnified Y Z projected images of Figs. 6(a)–6(d), respectively, from an optical depth 242 μm to 424 μm (areas surrounded by yellow dashed squares in Figs. 6(a)–6(d)). The brightness of each image is normalized by the respective maximum fluorescence intensity. In these figures, except for intensity normalization, other image processing such as gamma and offset correction was not performed. The quality of images from the SMM scan and MMM scan is similar. When the scans with SA correction were performed, the images from the scan with SA correction became brighter overall, and these results indicate that the fluorescence intensity was improved. However, it seems that the observation limit depth did not change even when there was no SA correction, because thick blood vessels were observed. In fact, thin blood vessels were not observed when the scan without SA correction was performed (Figs. 6(f) and 6(h)). In addition, even in a thick blood vessel, the blood stream can be observed, but it was not clear because the resolution was low. However, with SA correction, thick blood vessels as well as thin blood vessels were observed (Figs. 6(e) and 6(g)).
Fig. 6 Y Z projected images of blood vessels from the cerebrum of a mouse stained with DiI. A dry objective lens (NA = 0.7) was used, and the objective lens was moved in increments of 2.0 μm. (a), (b) Y Z projected images for an optical depth of 0 μm to 754 μm for MMM scans performed with and without SA correction, respectively. (c), (d) Y Z projected images for SMM scans performed with and without SA correction, respectively. (e)–(h) Magnified Y Z projected images for an optical depth of 242 μm to 424 μm (areas surrounded by yellow dashed squares in Figs. 6(a) to 6(d)). The scale bar indicates 100 μm.
Of course, the resolution in the x-y plane was improved with SA correction. Figure 7 shows the x-y image at an optical depth of 345 μm (orange dashed line in Fig. 6). The brightness of each image was normalized by the respective maximum fluorescence intensity. The magnified view of the vessel in the orange dashed box is shown on the upper left of Figs. 7 (a)–7(d). From Figs. 7(b) and 7(d), it can be seen that the blood vessels were blurred when the scan without SA correction was performed. However, with SA correction, the shape of the blood vessels could be confirmed (Figs. 7(a) and 7(c)). The fluorescence intensity of the vessel along the green dashed line is shown in Fig. 7(e). After subtracting the background, the maximum fluorescence intensity for MMM with SA correction was approximately 5.0 times higher than that without SA correction. This result is equivalent to the result for SMM with SA correction (5.3 times).
Fig. 7 Observed x-y images of blood vessels from the cerebrum of a mouse stained with DiI. (a), (b) x-y images at an optical depth of 345 μm for MMM scans performed with and without SA correction, respectively. (c), (d) x-y images for SMM scans performed with and without SA correction, respectively. The scale bar indicates 20 μm. The magnified view of the vessel in the orange dashed box is shown on the upper left of Figs. 7 (a) to (d). (e) the fluorescence intensity of the vessel along the green dashed line in the orange dashed box.
In the measurement using MMM, the maximum observation depth is limited by the power of the excitation laser. The power of the laser reached its limit when the MMM scan was performed at an optical depth of over 412 μm, as shown in Fig. 8.
Fig. 8 Intensity of the excitation beam for MMM and SMM. To clarify the effect of the SA correction in the deeper region, the intensity of the excitation beam was changed depending on the observation depth. The intensity of the excitation beam was measured under the objective lens. Compared to SMM, the intensity of the excitation beam for MMM was 4.6 times higher in the shallow depth region. The power of the laser reached its limit when the MMM scan was performed at an optical depth of over 412 μm.
6. Conclusion
To realize simplified measurement using MLSM with a dry objective lens, an SLM was incorporated in MLSM. Using the SLM, it is possible to realize both an increase in the quality of the fluorescence image and a shortened measurement time. Owing to the use of the SLM, the proposed system could switch between SMM and MMM. We confirmed that the effect of the SA correction in MMM was the same as that seen in SMM for fluorescent beads in transparent epoxy resin. Meanwhile, the scanning time for MMM was four times less than that for SMM. We also observed the blood vessels of the cerebrum of a mouse. For scans with SA correction, the resolutions of both the MMM and SMM images were improved. However, the observation limit for MMM is less than that for SMM due to the limit of the laser power. Of course, it is possible that a higher power laser source will increase the depth limit of MMM. It can also be expected to combine MMM and the other method such as passive pulse splitting [7] to observe further deep regions. Alternatively, we can adopt a method such that the system can perform MMM scans in the shallow region and SMM scans in the deeper region by simply changing the CGH when the scattering tissue is observed.
Funding
This work was partially supported by SENTAN, JST.
Acknowledgments
We are grateful to A. Hiruma, H. Toyoda, and T. Hara for their encouragement and to T. Miwa, N. Fukuchi, K. Nakamura, and H. Tanaka for their invaluable assistance.
References and links
1. D. M. Small, W. Y. Sanchez, S. Roy, M. J. Hickey, and G. C. Gobe, “Multiphoton fluorescence microscopy of the live kidney in health and disease,” J. Biomed. Opt. 19(2), 020901 (2014). [CrossRef] [PubMed]
2. C. O. Yanez, A. R. Morales, X. Yue, T. Urakami, M. Komatsu, T. A. H. Jarvinen, and K. D. Belfield, “Deep vascular imaging in wounds by two-photon fluorescence microscopy,” PLos One 8(7), e67559 (2013). [CrossRef] [PubMed]
3. P. Theer, M. T. Hasan, and W. Denk, “Two-photon imaging to a depth of 1000 μ m in living brains by use of a Ti:Al2O3 regenerative amplifier,” Opt. Lett. 28(12), 1022–1024 (2003). [CrossRef] [PubMed]
4. R. Kawakami, K. Sawada, A. Sato, T. Hibi, Y. Kozawa, S. Sato, H. Yokoyama, and T. Nemoto, “Visualizing hippocampal neurons with in vivo two-photon microscopy using a 1030 nm picosecond pulse laser,” Sci. Rep. 3, 1014 (2013). [CrossRef] [PubMed]
5. L. Kong, J. Tang, J. P. Little, Y. Yu, T. Lammermann, C. P. Lin, R. N. Germain, and M. Cui, “Continuous volumetric imaging via an optical phase-locked ultrasound lens,” Nat. Methods 12, 759–762 (2015). [CrossRef] [PubMed]
6. T. A. Planchon, L. Gao, D. E. Milkie, M. W. Davidson, J. A. Galbraith, C. G. Galbraith, and E. Betzig, “Rapid three-dimensional isotropic imaging of living cells using Bessel beam plane illumination,” Nat. Methods 8, 417–423 (2011). [CrossRef] [PubMed]
7. N. Ji, J. C Magee, and E. Betzig, “High-speed, low-photodamage nonlinear imaging using passive pulse splitters,” Nat. Methods 5, 197–202 (2008). [CrossRef] [PubMed]
8. G. Y. Fan, H. Fujisaki, A. Miyawaki, R. -K. Tsay, R. Y. Tsien, and M. H. Ellisman, “Video-Rate Scanning two-photon excitation fluorescence microscopy and ratio imaging with cameleons,” Biophys. J. 76, 2412–2420 (1999). [CrossRef] [PubMed]
9. K. H. Kim, C. Buehler, and P. T. C. So, “High-speed, two-photon scanning microscope,” Appl. Opt. 38, 6004–6009 (1999). [CrossRef]
10. J. Bewersdorf, R. Pick, and S. W. Hell, “Multifocal multiphoton microscopy,” Opt. Lett. 23, 655–657 (1999). [CrossRef]
11. A. Egner, J. Bewersdorf, and S. W. Hell, “Comparison of the axial resolution of practical Nipkow-disk confocal fluorescence microscopy with that of multifocal multiphoton microscopy: theory and experiment,” J. Microscopy 206, 24–32 (2001). [CrossRef]
12. K. Bahlmann, P. T. C. So, M. Kirber, R. Reich, B. Kosicki, W. McGonagle, and K. Bellve, “multi-focal multiphoton microscopy (MMM) at a frame rate beyond 600 Hz,” Opt. Express 15, 10991–10998 (2007). [CrossRef] [PubMed]
13. K. H. Kim, C. Buehler, K. Bahlmann, T. Ragan, W. A. Lee, E. Nedivi, E. L. Heffer, S. Fantini, and P. T. C. So, “Multi-focal multiphoton microscopy based on multianode photomultiplier tubes,” Opt. Express 15, 11658–11678 (2007). [CrossRef] [PubMed]
14. J. W. Cha, V. R. Singh, K. H. Kim, J. Subramanian, Q. Peng, H. Yu, E. Nedivi, and P. T. C. So, “Reassignment of scattering emission photons in multi-focal multiphoton microscopy,” Sci. Rep. 4, 5153 (2014). [CrossRef]
15. L. Sacconi, E. Froner, R. Antolini, M. R. Taghizadeh, A. Choudhury, and F. S. Pavone, “Multiphoton multi-focal microscopy exploiting a diffractive optical element,” Opt. Lett. 28, 1918–1920 (2003). [CrossRef] [PubMed]
16. T. Nielsen, M. Fricke, D. Hellweg, and P. Andersen, “High efficiency beam splitter for multi-focal mutiphoton microscopy,” J. Microscopy 201, 368–376 (2000). [CrossRef]
17. Y. Shao, W. Qin, H. Lin, J. Qu, X. Peng, H. Niu, and B. Z. Gao, “multi-focal multiphoton microscopy based on a spatial light modulator,” Appl. Phys. B 107, 653–657 (2012). [CrossRef]
18. S. Coelho, S. Poland, N. Krstajic, D. Li, J. Moneypenny, R. Walker, D. Tyndall, T. Ng, R. Henderson, and S. Ameer-Beg, “multi-focal multiphoton microscopy with adaptive optical correction,” Proc. SPIE 8588, 858817 (2013). [CrossRef]
19. S. P. Poland, N. Krstajić, J. Monypenny, S. Coelho, D. Tyndall, R. J. Walker, V. Devauges, J. Richardson, N. Dutton, P. Barber, D. D. Li, K. Suhling, T. Ng, R. K. Henderson, and S. M. Ameer-Beg, “A high speed multifocal multiphoton fluorescence lifetime imaging microscope for live-cell FRET imaging,” Bio. Opt. Express 6(2), 277–296 (2015). [CrossRef]
20. X. Duan, H. Li, X. Li, K. R. Oldham, and T. D. Wang, “Axial beam scanning in multiphoton microscopy with MEMS-based actuator,” Opt. Express 25(3), 2195–2205 (2017). [CrossRef]
21. W. Yang, J. Kang, M. L. Carrillo-Reid, E. Pnevmatikakis, L. Paninski, R. Yuste, and D. S. Peterka, “Simultaneous multi-plane imaging of neural circuits,” Neron 89(2) 269–284 (2016). [CrossRef]
22. N. Matsumoto, S. Okazaki, Y. Fukushi, H. Takamoto, T. Inoue, and S. Terakawa, “An adaptive approach for uniform scanning in multi-focal multiphoton microscopy with a spatial light modulator,” Opt. Express 22(1), 633–645 (2014). [CrossRef] [PubMed]
23. E. Theofanidou, L. Wilson, W. J. Hassack, and J. Arlt, “Spherical aberration correction for optical tweezers,” Opt. Commun. 236(1–3), 145–150 (2004). [CrossRef]
24. M. A. A. Neil, R. Juškaitis, M. J. Booth, T. Wilson, T. Tanaka, and S. Kawata, “Adaptive aberration correction in a two-photon microscope,” J. Microscopy 200(2), 105–108 (2000). [CrossRef]
25. M. J. Booth, “Adaptive optics in microscopy,” Phil. Trans R. Soc. A 365, 2829–2843 (2007). [CrossRef] [PubMed]
26. P. Török, P. Varga, and G. R. Booker, “Electromagnetic diffraction of light focused through a planar interface between materials of mismatched refractive indices: Structure of the electromagnetic field. I,” J. Opt. Soc. Am. A 12(10), 2136–2144 (1995). [CrossRef]
27. P. S. Salter, M. Baum, I. Alexeev, M. Schmidt, and M. J. Booth, “Exploring the depth range for three-dimensional laser machining with aberration correction,” Opt. Express 22(15), 17644–17656 (2014). [CrossRef] [PubMed]
28. N. Matsumoto, T. Inoue, A. Matsumoto, and S. Okazaki, “Correction of depth-induced spherical aberration for deep observation using two-photon excitation fluorescence microscopy with spatial light modulator,” Bio. Opt. Express 6(7), 2575–2587 (2015). [CrossRef]
29. N. Matsumoto, H. Itoh, T. Inoue, T. Otsu, and H. Toyoda, “Stable and flexible multiple spot pattern generation using LCOS spatial light modulator,” Opt. Express 22(20), 24722–24733 (2014). [CrossRef] [PubMed]
30. N. Matsumoto, T. Ando, T. Inoue, Y. Ohtake, N. Fukuchi, and T. Hara, “Generation of high-quality higher-order Laguerre Gaussian beams using liquid-crystal-on-silicon spatial light modulators,” J. Opt. Soc. Am. A 25(7), 1642–1650 (2008). [CrossRef]
31. D. Prongue, H. P. Herzig, R. Dandliker, and M. T. Gale, “Optimized kinoform structures for highly efficient fan-out elements,” Appl. Opt. 31, 5706–5711 (1992). [CrossRef] [PubMed]
32. N. Ji, J. C Magee, and E. Betzig, “Adaptive optics via pupil segmentation for high-resolution imaging in biological tissues,” Nat. Methods 7, 141–147 (2010). [CrossRef]
33. Y. Li, Y. Song, L. Zhao, G. Gaidosh, AM Laties, and R. Wen, “Direct labeling and visualization of blood vessels with lipophilic carbocyanine dye DiI”, Nat. Protocols 3, 1703–1708 (2008). [CrossRef]
34. M. T. Ke, S. Fujimoto, and T. Imai, “SeeDB: A simple and morphology-preserving optical clearing agent for neuronal circuit reconstruction,” Nat. Neurosci. 16, 1154–1161 (2013). [CrossRef] [PubMed]
