Method for high-resolution multiphoton fluorescence microscopy in turbid tissues

Abstract

In one aspect, the disclosure relates to a method for correcting distortions or aberrations in a multi-photon microscopy image, the method including at least: (a) illuminating the sample; (b) imaging the illuminated sample with an imaging device having one or more detection elements that receive a signal from the sample; and (c) processing the image using a scattering correction and a phase correction. The phase correction can be a low-order deformable mirror adaptive optics correction with a hardware component and a software component. In some aspects, a spatial light modulator can be employed as a low-order phase correcting device. The scattering correction can be intensity scattering correction, phase scattering correction, or both, and can be performed with a high-order digital micromirror device with binary intensity modulation or a phase or intensity high-order spatial light modulator device. Also disclosed are methods of imaging complex tissue in living subjects.

Description

ATTORNEY DOCKET NO. 222105-2420METHOD FOR HIGH-RESOLUTION MULTIPHOTON FLUORESCENCE MICROSCOPY IN TURBID TISSUESCROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims the benefit of U.S. Provisional Application No. 63 / 723,618 filed on November 22, 2024, which is incorporated herein by reference in its entirety.STATEMENT REGARDING FEDERALLY SPONSORED RESEARCH OR DEVELOPMENT

[0002] This invention was made with government support under grant numbers 1706916 and 2040118 awarded by the National Science Foundation. The government has certain rights in the invention.BACKGROUND

[0003] Over the last 30 years, multi-photon microscopy has been widely applied in the study of biological samples, providing intrinsic optical sectioning and sensitivity that allows visualization of cells and tissues hundreds of microns within biological tissues. Multiphoton microscopy uses high energy sources with excitation wavelengths longer than the emission wavelength, where lower tissue scattering extends the mean free path and enables deeper penetration than typical single photon fluorescence imaging. Although these wavelengths offer deeper imaging, they remain subject to tissue scattering beyond the ballistic range, which distorts the focus of the excitation beam and increases the instantaneous excitation power necessary for adequate signal-to-noise for imaging.

[0004] Mitochondria are intracellular organelles with 0.5 to 10 pm diameter that drive energy production processes through the respiratory chain by oxidative phosphorylation. They play a fundamental role in numerous physiological processes of critical importance in tissue homeostasis and repair, such as cell differentiation, apoptosis, signal transduction, reactive oxygen species generation, and maintenance of healthy organ function. Their function is highly dynamic and reflected in mitochondrial network structure, and imaging technologies are therefore essential to understand physiological mitochondrial processes in health and disease. High energy requirement tissues such as the brain and bone marrow are especially dependent on carefully orchestrated mitochondrial maintenance and activity. In addition to mitochondria, collecting information about other organelles, cellular, and sub-cellular structures can provide useful tools for understanding processes of health and disease in a variety of tissues in addition to bone, brain, and bone marrow.

[0005] Over the last decades, the evaluation of live tissue dynamics at cellular resolution using intravital imaging has transformed the biological understanding of organ function at a singleATTORNEY DOCKET NO. 222105-2420 cell level. In the bone, this has significantly advanced the scientific understanding of vascular dynamics, stem cell biology, and bone homeostasis and regeneration. In the brain, intravital imaging has generated unique insight into brain circuitry and processing, brain cancer, brain trauma, and degenerative diseases. However, one of the most serious obstacles to imaging is the poor penetration depth of intravital optical microscopy. Single photon imaging with confocal detection is a common approach that uses visible light for fluorescence excitation (400-650 nm), where light penetration is attenuated by absorption and scattering of skull bone and tissues. To extend imaging depth, high energy pulses of near-IR excitation light (760- 1 ,080 nm) can be tightly focused to create a non-linear two-photon absorption process using standard fluorophores. Two-photon imaging extends the attainable imaging depth to -500 pm of brain tissue and up to ~150 pm in highly scattering bone. However, non-homogeneous wave propagation through these irregular and highly distorting turbid media induces high magnitude phase and intensity deviations in the wavefront that dramatically reduce image resolution even at moderate depths. Phase aberrations have a greater effect on the total wavefront distortion in comparison to intensity aberrations, and so noninvasive correction of tissue aberrations that disrupt imaging is most frequently performed by modulating the phase of the incoming light. Therefore, cranial windows and skull-thinning methods are commonly adopted to improve optical access. However, the surgery increases the risk of tissue inflammation and may cause stress that could alter biological function in the target tissues. Furthermore, current technology typically allows corrections for a field of view of only about 1 to 2 pm2, thereby limiting the usefulness of such methods.

[0006] One way to overcome this challenge is with adaptive optics (AO). An AO system typically consists of a deformable mirror conjugated to the back pupil plane of a microscope, and either wavefront sensor or image-based sensorless wavefront estimation methods to correct aberrations and improve the resolution, which can improve in vivo imaging in animal models. Wavefront sensor approaches include Shack-Hartmann wavefront sensors with auto- fluorescent or near-IR guide stars, coherence gated wavefront sensing, and image-based methods that use information from acquired images to remove wavefront distortions. Wavefront sensorless approaches usually estimate an initial error and through an iterative scheme converge to an optimized solution based on intensity metrics. Recently, non-linear guide stars with Shack-Hartmann measurements of wavefront aberrations have yielded an accurate measurement of low-order tissue aberrations that proves to be useful to extend imaging depth in biological samples but require long integration times and has a small effective field of view. In an alternative wavefront sensorless approach using rapid anisotropic aberration correction, the adaptive correction element was conjugated to the turbid layer instead of the focus, with the goal of increasing the size of the isoplanatic patch. Typical boneATTORNEY DOCKET NO. 222105-2420 marrow imaging occurs within an extended scattering layer, so such an approach is not appropriate.

[0007] The effects of low order tissue aberrations, such as spherical aberrations, have been well studied and can be effectively modeled and corrected, but sample-induced aberrations caused by the complex structure of tissues can vary dramatically even within a tissue and require correction. Adaptive optics techniques aimed at correcting lower-order aberrations by means of active elements such as deformable mirrors (DMs) have shown success in correcting aberrations in homogenous tissues, such as Drosophila brain, zebrafish, and mouse brain, however, in highly scattering media such as bone, the number of scattered modes is much larger than the number of degrees of freedom of the DM and the correction becomes less effective.

[0008] As the imaging depth increases, rapidly varying aberrations due to tissue scattering impact the beam progressively more, requiring alternate approaches using high degree of freedom devices like liquid crystal spatial light modulators (SLMs). In these regimes, iterative feedback-based optimization of the wavefront modulator is commonly used since low spatial coherence of the focused light does not support direct wavefront sensing. The convergence speed of iterative wavefront optimization is largely dependent on the update rate of the correction device, and so although liquid crystal SLMs yield reasonable convergence times, their update rate (<800Hz) and the small phase isoplanatic patch size in scattering media has limited practical application for epi-detection imaging in live animals. Due to these challenges with phase correction, intensity aberrations are an attractive alternative target, which can be modulated with faster (~ 32kHz) digital micromirror device (DMD) wavefront modulators that often have >15,000 actuators. DM Ds are binary, and so cannot explicitly adjust the pixel phase without implementation of complex off-axis Lee holography setups. Lee holography with a binary modulation has low efficiency in terms of laser throughput and phase range compared to a traditional phase only SLM, and so most DMD correction approaches have performed binary intensity modulation in the Fourier plane to address intensity distortions.

[0009] In strongly scattering samples the correction enhancement for intensity aberrations scales with a slope of 1 / (2TT), which is less than the phase correction (~TT / 4), but transmission mode intensity correction has still found enhancements of 50-500x. This type of intensity signal enhancement can be thought of as selectively modulating portions of the focused beam to remove destructive interference and enhance constructive interference. In transmission mode, a broad variety of binary intensity algorithms have been demonstrated, including particle swarm, genetic algorithm variants., and projection of Hadamard modes, and simulated annealing, However, emitted signal in multiphoton microscopy is collected by epi-detection, with photon fluxes that are orders of magnitude lower than in transmitted mode intensityATTORNEY DOCKET NO. 222105-2420 scattering correction. Low initial photon fluxes reduce the efficacy of correction algorithms, and only one recent work has explored 2-photon binary intensity modulation, using a stochastic optimization algorithm to achieve intensity enhancement in the brain of ~30-40x at -500 / zm depth, or 2.5 - 3 scattering lengths^ This suggests the promise of binary modulation as an epi-detection scattering correction technique, and the potential for further opportunities to enhance intensity correction performance. While intensity and phase modulation are each capable of providing an increase in the imaging intensity in scattering media, complexities in the implementation have heretofore prevented their successful combination in a single microscope system.

[0010] It would be desirable to be able to non-invasively observe and quantify osteocyte characteristics inside bone in order to elucidate the mechanisms of osteogenesis, particularly in the case of osteopathies such as osteoporosis, osteomalacia, Paget’s disease, hypophosphatasia (HPP), aging-related bone changes, X-linked hypophosphatemia, bone repair after injury, or osteogenesis imperfecta (Ol). Doing so through whole skull bone would remove the impact of immune responses to bone clearing methods and would provide a platform from which more representative data can be captured within the skull, and, with application of more power, brain imaging through whole skull. It would further be desirable if methods useful for observe and quantify cells and organelles in bone could be extended for use in other complex tissues including, but not limited to, the brain, skeletal and smooth muscle, and others, and at a greater depth in the tissue with a larger field of view compared to current technologies. These needs and other needs are satisfied by the present disclosure.SUMMARY

[0011] In accordance with the purpose(s) of the present disclosure, as embodied and broadly described herein, the disclosure, in one aspect, relates to a method for correcting distortions or aberrations in a multi-photon microscopy image of a sample, the method including at least the steps of: (a) illuminating the sample; (b) imaging the illuminated sample with an imaging device having one or more detection elements that receive a signal from the sample; and (c) processing the image of the sample using a phase correction and a scattering correction. In an aspect, the phase correction can be a low-order deformable mirror adaptive optics correction and can have a hardware component and a software component. In another aspect, a spatial light modulator (SLM) can be employed as a low-order phase correcting device instead of a deformable mirror. In another aspect, the scattering correction can be intensity scattering correction, phase scattering correction, or both. In another aspect, the scattering correction can be performed with a high-order digital micromirror device with binary intensity modulation. In another aspect, the scattering correction can be performed with a phase or intensity high-order spatial light modulator device. Also disclosed are methods of imagingATTORNEY DOCKET NO. 222105-2420 complex tissue such as, for example, bone, to a depth of up to 250 pm, including in living subjects and without inducing inflammation or another immune response in the living subjects. In one aspect, in the disclosed method, an improvement of at least 10* can be achieved in signal intensity at the same time an enhancement of at least 2* in imaging depth is achieved relative to current imaging technologies.

[0012] Other systems, methods, features, and advantages of the present disclosure will be or become apparent to one with skill in the art upon examination of the following drawings and detailed description. It is intended that all such additional systems, methods, features, and advantages be included within this description, be within the scope of the present disclosure, and be protected by the accompanying claims. In addition, all optional and preferred features and modifications of the described embodiments are usable in all aspects of the disclosure taught herein. Furthermore, the individual features of the dependent claims, as well as all optional and preferred features and modifications of the described embodiments are combinable and interchangeable with one another.BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Many aspects of the present disclosure can be better understood with reference to the following drawings. The components in the drawings are not necessarily to scale, emphasis instead being placed upon clearly illustrating the principles of the present disclosure. Moreover, in the drawings, like reference numerals designate corresponding parts throughout the several views.

[0014] FIG. 1 shows a schematic diagram of a home-built microscope setup with excitation from a Ti:Sapphire laser ranging from 680 nm to 1080 nm. A digital Micromirror device (DMD) conjugated to the back-pupil plane enables the scatter correction via binary wavefront modulation. A deformable mirror (DM) conjugated to the back-pupil plane enables lower-order phase aberration correction. Galvanometer and resonant scanners combine to scan the excitation beam across samples and the resulting emission light captured by photon-multiplier tubes (PMT). To evaluate both low and high order aberrations, the CMOS camera is used to image the point spread function of the second harmonic generation (SHG).

[0015] FIGs. 2A-2E represent a demonstration of DM low-order phase correction in a tissue phantom. System and mean-intensity-based sample phase AO are shown on 0.2 pm beads (FIG. 2B) at 50 pm depth in an agarose gel (FIG. 2A). Zernike mode decomposition of the wavefront (FIG. 2C). Wavefront after full sample phase AO (pm) (FIG. 2D). Spot size (FIG. 2E). The red and green boxes represent the full width at half maximum (FWHM) of each intensity profile. The FOV is 67.5 x 67.5 pm. Data are represented as mean ± standard deviation for each measurement. Scale bar = 10 pm.ATTORNEY DOCKET NO. 222105-2420

[0016] FIGs. 3A-3G show a demonstration of DM low-order phase correction in live brain tissue. Images are of the hippocampus of a young mouse brain with system AO (FIG. 3A) and with mean-intensity-based full AO (FIG. 3B). Corresponding FFTs of the brain image in log scale and line profiles (FIG. 3C). Corresponding signal profiles along the white lines, the y-axis is intensity (FIG. 3D). Wavefront after full AO (pm) (FIG. 3E). Singular value decomposition of the Zernike modes (FIG. 3F) for sample correction. The hippocampus of young mouse was imaged at the red asterisk (FIG. 3G). The field of view (FOV) is 59.32 x 59.32 pm. Scale bar = 10 pm.

[0017] FIGs. 4A-4B show loss of resolution at the depth of 0, 50, 85 pm with system AO and with full sample phase AO in intensity (FIG. 4A) and FWHM (FIG. 4B). The error bar shows the standard deviation of each measurement.

[0018] FIGs. 5A-5D show dynamics of cells. Shown is a time-coded pseudo color max projection of time lapse imaging of GFP-mitochondria mouse bone marrow at the depth of 40 pm with mean-intensity-based full AO. Importantly, the corrections applied remain valid over the length of imaging time. (FIGs. 5A-5B) show an overview of mitochondria traveling in the bone marrow. (FIG. 5C) shows different time points. (FIG. 5D) shows mitochondria fluctuations. The y-axis is intensity. The FOV is 67.5 x 67.5 pm. Scale bar = 10 pm.

[0019] FIG. 6A shows an illustration of the steps followed by the Genetic Algorithm in optimizing the DMD mask. First, randomized masks are generated, projected onto the DMD, and evaluated. Top-ranked masks are propagated to the next generation and used to generate 50% of a new generation by crossover (gene swapping). According to R (the mutation rate), elements of the new generation are then mutated, in this case with a maximum of 52.5% on segments. The resulting population is then projected and evaluated. This process is repeated until a fixed number of iterations have completed. FIGs. 6B-6E show simulation results show that when no decay of the mutation rate is implemented, a rapid convergence (FIG. 6B) is achieved, but the resulting accuracy is low (FIG. 6D). When a decay of the mutation rate is introduced, convergence is delayed (FIG. 6C), but the resulting accuracy is increased (FIG. 6E). Error bars are the standard error.

[0020] FIGs. 7A-7C show experimental enhancement, q, for the three phantom samples, thin fluorescent, extended fluorescent, and extended potassium titanyl phosphate (KTP) crystal under various diffusers, where 373 (FIG. 7A), 1436 (FIG. 7B), and (FIG. 7C) 5785 segments were used during optimization.

[0021] FIGs. 8A-8L show validation in a scattering phantom. Optimized patterns according to segment counts for each of the various phantom samples are evaluated. Optimized patterns for the thin fluorescent sample, with 1 layer of translucent tape as diffuser are shown in row 1 ,ATTORNEY DOCKET NO. 222105-2420 where 373 (FIG. 8A), 1436 (FIG. 8B), and 5785 (FIG. 80) segments are used. Optimized patterns for the thin fluorescent sample, with 3 layers of translucent tape are shown in row 2, where 373 (FIG. 8D), 1436 (FIG. 8E), and 5785 (FIG. 8F) segments were used. Optimized patterns for the autofluorescent plastic slide (extended sample) with 3 layers of translucent tape are shown in row 3, where 373 (FIG. 8G), 1436 (FIG. 8H), and 5785 (FIG. 8I) segments were used. Optimized patterns for the mounted KTP crystal with 1 layer of translucent tape, where 373 (FIG. 8J), 1436 (FIG. 8K), and 5785 (FIG. 8L) segments were used.

[0022] FIGs. 9A-9F show scatter correction implemented at a depth of ± 50 pm with calculated enhancement in (FIG. 9A). (FIG. 9D) shows the average of the static images of the first generation, created using the randomized DMD pattern in (FIG. 9B). (FIG. 9E) shows the average of the static images created using the final generation of the corrected DMD pattern in (FIG. 9C), population size = 100. The improvement in the corrected versus initial spot intensity is shown in (FIG. 9F).

[0023] FIGs. 10A-10I show that, at a depth of ±140 pm, the bone marrow is imaged, and clear punctate mitochondria are visible after correction (FIG. 10A). The corrected DMD pattern is inset in (FIG. 10A), 52.21% segments ”on”. Application of the uncorrected DMD pattern yields low signal in both the GFP (FIG. 10B) and SHG (FIG. 10C) channels. After correction, the GFP channel (FIG. 10D) shows mitochondria that were not visible using the uncorrected DMD pattern, the green line cross section comparison quantified in (FIG. 10F). The improvement in second harmonic generation (SHG) when applying the corrected DMD pattern is clear in (FIG. 10E) and the blue line cross section comparison quantified in (FIG. 10G). Fourier Ring Correlation shows an increase in spatial resolution of 905 nm in GFP (FIG. 10H), and 688 nm in SHG (FIG. 101). Scale bar is 10 pm.

[0024] FIGs. 11A-11B show raw and averaged intensities during the scatter correction within murine skull samples using SHG as the guidestar and a photomultiplier tube (PMT) as the detector. Long integration times of 100 ms per individual result in smooth measurements (FIG. 11 A), while shorter integration times of 10ms result in more stochastic intensity measurements (FIG. 11B).

[0025] FIGs. 12A-12I show that, at a depth of ± 125 pm, GFP present within the bone is visible (FIG. 12A). The corrected DMD pattern is inset in (FIG. 12A), 52.28% segments ”on”. Application of the uncorrected DMD pattern yields low signal in both the GFP (FIG. 12B) and SHG (FIG. 12C) channels. After correction, the GFP channel (FIG. 12D) shows cells that were not visible using the uncorrected DMD pattern, the green line cross section comparison quantified in (FIG. 12F). The improvement in SHG when applying the corrected DMD pattern is clear in (FIG. 12E) and the blue line cross section comparison quantified in (FIG. 12G).ATTORNEY DOCKET NO. 222105-2420Fourier Ring Correlation shows an increase in spatial resolution of 741 nm in GFP (FIG. 12H), and 515 nm in SHG (FIG. 121). Scale bar is 10 pm.

[0026] FIG. 13 shows imaging at a depth of 15 pm within whole skull bone, a minimal improvement is measured when using the optimized DMD pattern. Deeper imaging results in more significant improvements of spatial resolution.

[0027] FIGs. 14A-14F show a comparison of low order aberration corrections executed at two depths. At 65 pm deep the aberration correction fails when an uncorrected pattern is projected onto the DMD (FIG. 14A), but when the corrected pattern was used the correction succeeded (FIG. 14B). A cross section across the center of (FIG. 14A) and (FIG. 14B) shows a 1.7-fold increase in the mean intensity of the SHG point spread function (PSF) (FIG. 14C). At a depth of 50 pm the uncorrected DMD pattern resulted in a resulted in a successful aberration correction (FIG. 14D) and the aberration correction using the corrected DMD pattern resulted in a smaller, brighter SHG PSF (FIG. 14E). A cross section of (FIGs. 14D-14E) shows that the point improved by a factor of 1.7 and narrowed (FIG. 14F).

[0028] FIGs. 15A-15F show that, without scatter or aberration correction, the scanned image of an excised murine skull at a depth of —110 pm presents dim GFP and SHG (FIG. 15A). Conducting a sample aberration correction step improves the intensity slightly with new features visible within the cell of interest (FIG. 15B). After applying a scatter correction mask onto the DMD, but without aberration correction, significant improvement is seen in the intensity of both GFP and SHG (FIG. 15C). With both scatter and aberration correction applied, maximized intensity and finer features are seen within the scanned image (FIG. 15D). A cross section of the GFP in each of the images confirms that the intensity is maximized when both corrections are applied (FIG. 15E) and the same is reflected in the SHG channel (FIG. 15F). Scale bar is 10 pm.

[0029] FIGs. 16A-16F show a comparison of low order aberration corrections at two depths. At 65pm deep the aberration correction fails when an uncorrected pattern is projected onto the DMD (FIG. 16A), but when the corrected pattern was used, the correction succeeded (FIG. 16B). A cross section across the center pf (FIG. 16A) and (FIG. 16B) shows a 1.7-fold increase in the mean intensity of the SHG PSF (FIG. 16C). At a depth of 50 pm the uncorrected DMD pattern resulted in a successful aberration correction (FIG. 16D) and the aberration correction using the corrected DMD pattern resulted in a smaller, brighter SHG PSF (FIG. 16E). A cross section of (FIG. 16D) and (FIG. 16E) shows that the point improved by a factor of 1.7 and narrowed (FIG. 16F).

[0030] FIGs. 17A-17E show that without scatter or aberration correction the scanned image of an excised murine skull at a depth of 130 pm presents dim GFP (FIG. 17A). Conducting aATTORNEY DOCKET NO. 222105-2420 phase sample aberration correction step does not alter the intensity (FIG. 17B). Applying a scattering correction mask to the DMD, but without aberration correction, yields significant improvement in GFP intensity (FIG. 17C). With both aberration and scatter correction applied, maximized intensity and finer features are seen within the scanned image (FIG. 17D). A cross section of the GFP in each of the images confirms that the intensity is maximized when both corrections are applied (FIG. 17E), suggesting a synergistic effect of low order phase and scattering correction. Scale bar is 50 pm.

[0031] Additional advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or can be learned by practice of the invention. The advantages of the invention will be realized and attained by means of the elements and combinations particularly pointed out in the appended claims. It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention, as claimed.DETAILED DESCRIPTION

[0032] Disclosed herein is a method of aberration correction for deep, noninvasive, multiphoton microscopy imaging of bone and other tissues. In one aspect, in the disclosed method, an improvement of at least 10* can be achieved in signal intensity at the same time an enhancement of at least 2* in imaging depth is achieved relative to current imaging technologies. In one aspect, the method involves manipulating the intensity of the excitation wavefront in the Fourier plane utilizing a DMD and optimizing the applied pattern using a genetic algorithm, based on the epi-detected second harmonic generation (SHG) signal generated in bone (specifically by type I collagen) as the guidestar. In some aspects, other optimization algorithms can be used instead of or in addition to the genetic algorithm including, but not limited to, particle swarm, projection of Hadamard modes, simulated annealing, another stochastic optimization algorithm, or a variant of the genetic algorithm. In another aspect, other biological tissue components can also be used as guidestars with appropriate adjustment and processing of signals. In a further aspect, in the disclosed method, at a depth of 140 pm, both the intensity (by a factor of ±8-10) and spatial information (increased by 905 nm in GFP, and 688 nm in SHG) in the captured images, using optimized patterns projected by the DMD were improved. In some aspects, an SLM can be used in place of a DMD or DM.

[0033] Also disclosed herein is a combination of rapid binary intensity modulation with phase enhancement, including an effective demonstration of the same in tissues. With both aberration and scatter correction applied, maximized intensity and finer features are seen within the scanned images. In the present systems and methods, intensity is maximized whenATTORNEY DOCKET NO. 222105-2420 both of these corrections are applied, suggesting a synergistic effect of low order phase and scattering correction (see FIGs. 16A-17E).

[0034] Also disclosed herein are methods for calculating and correcting sample aberrations caused by biologically complex tissues including, but not limited to, mouse cranial bone and brain tissue to improve imaging for dynamic mitochondria localization. In an aspect, it is demonstrated that low-order aberration correction provides a significant improvement when imaging through the bone into the bone marrow. In one aspect, in the disclosed method, the aberrations of mouse cranial bone are compensated for by using a Zernike-mode-based sensorless AO algorithm to optimize and create an improved wavefront profile. In one aspect, a two-photon fluorescence microscope is used with adaptive optics to image mitochondria in mouse cranial bone marrow and brain. In an alternative aspect, three-photon microscopy can be used. In a further aspect, the threshold and performance for intensity-based sample correction are also determined. In one aspect, herein it is shown and disclosed that after AO correction and / or scattering correction, the fluorescence intensity of the point spread function (PSF) is improved, and the resolution of images is significantly improved when imaging through intact mouse cranial bone into the bone marrow, allowing characterization of mitochondrial health and dynamics of functioning cells deep in tissue. In another aspect, the combination of binary intensity modulation and phase modulation provides synergistic enhancement to further increase image brightness and quality.Method for Correcting Distortions or Aberrations in a Multi-Photon Microscopy Image

[0035] In one aspect, disclosed herein is a method for correcting distortions or aberrations in a multi-photon microscopy image of a sample, the method including at least the following steps:(a) illuminating the sample;(b) producing an image of the sample with an imaging device having one or more detection elements that receive a signal from the sample; and(c) processing the image of the sample using a scattering correction and a phase correction.

[0036] In some aspects, the multi-photon microscopy image is a two-photon microscopy image. In other aspects, the multi-photon microscopy image is a three-photon microscopy image. In another aspect, the phase correction can be a low order deformable mirror (DM) adaptive optics (AO) correction. In another aspect, the scattering correction can be a high- order digital micromirror device (DMD) correction. In one aspect, the scattering correction includes a binary scattering correction, a phase scattering correction, or both. In someATTORNEY DOCKET NO. 222105-2420 aspects, an optional spatial light modulator (SLM) correction can also be used. However, in some aspects, the SLM correction is not used. In one aspect, an intensity correction is performed first, followed by a low order phase correction. Further in this aspect, performing the corrections can be faster than other methods, enabling visualization of real-time changes. In some aspects, the order of performing these corrections is not varied. In one aspect, an initial phase correction followed by a limited manipulation of intensity may provide a modest boost, but reversing the order can provide a greater imaging enhancement in terms of scattering, surface area, imaging depth, or the like.Light Source for Performing the Correction

[0037] In one aspect, the microscopy can be two-photon microscopy and the light source can be a Thsapphire laser and can induce fluorescence using a wavelength of from about 680 nm to about 1080 nm, or of from about 700 nm to about 1000 nm, or of from about 800 nm to about 900 nm, or at about 680, 700, 720, 740, 760, 780, 800, 820, 840, 860, 880, 900, 920, 940, 960, 980, 1000, 1020, 1040, 1060, or 1080 nm, or a combination of any of the foregoing values or a range encompassing any of the foregoing values. In another aspect, when the microscopy is three-photon microscopy, a different suitable laser can be used, and can induce fluorescence using a wavelength of from about 1200 nm to about 1900 nm, or about 1200, 1250, 1300, 1350, 1400, 1450, 1500, 1550, 1600, 1650, 1700, 1750, 1800, 1850, or about 1900 nm, or a combination of any of the foregoing values, or a range encompassing any of the foregoing values. In an aspect, the light source can produce a plurality of pulses of energy of from about 75 fs to about 400 fs, or from about 75 fs to about 200 fs, or of about 75, 100, 125, 150, 175, 200, 225, 250, 275, 300, 325, 350, 375, or about 400 fs, or a combination of any of the foregoing values, or a range encompassing any of the foregoing values. In some aspects, the light source produces a plurality of 137 fs pulses of energy. In a further aspect, power of the light source is modulated using a half-waveplate, a polarizing beam splitter, or both. In yet another aspect, after splitting, the beam can be recollimated and passed through a Pockels cell for rapid intensity modulation during imaging. Further in this aspect, after the beam exits the Pockels cell, the beam passes through a pinhole, wherein passing through the pinhole improves the Gaussian profile of the beam. In an aspect, the Pockels cell is further used to block the beam during flyback and to correct beam intensity for a resonant scanner.Adaptive Optics Correction using a Deformable Mirror

[0038] In one aspect, AO correction includes an instrumental component and a software component. In another aspect, in the disclosed method, the AO correction corrects for system aberrations. In some aspects, the AO correction is conducted prior to correcting for sample aberrations. In one aspect, the instrumental component of the AO correction comprises aATTORNEY DOCKET NO. 222105-2420 deformable mirror (DM). Further in this aspect, the DM is placed conjugate to the back pupil plane and wherein the DM is used for low-order AO wavefront correction. In any of these aspects, the DM can have a continuous face sheet or a segmented face sheet. In another aspect, the DM can have a large number of actuators such as, for example, from about 50 to about 500 actuators. However, other numbers of actuators are also contemplated and should be considered disclosed. In one aspect, the DM is inserted into a path of the beam with a first optical relay system and a second optical relay system.Digital Micromirror Device Correction

[0039] In one aspect, the method further includes the step of using a digital micromirror device (DMD) to expand the beam to utilize from about 10% to about 50%, or about 30% of a DMD active array surface. In another aspect, the DMD device comprises a plurality of micromirrors, with each member of the plurality having a diameter of about 13.68 pm, or from about 10 pm to about 20 pm. In another aspect, each of the plurality of micromirrors tilts at ± from about 5° to about 20°, or, in one example, at about +12° or -12°. In still another aspect, a diffraction grating (DG) compensates for dispersion generated by a surface of the DMD. In some aspects, the method further includes the step of using a Genetic Algorithm as described herein to create an optimized pattern set for application to the DMD.Additional Components of the Microscopy System

[0040] In one aspect, in the disclosed method, beam scanning is performed by a resonant scanning box housing a fast resonant galvo scanner for horizontal sweeping and a slow galvo scanner for vertical sweeping, thereby producing a scanned beam. In a further aspect, the fast resonant galvo scanner and the slow galvo scanner are placed close together to reduce astigmatism. In another aspect, the scanned beam is relayed using a first achromatic doublet serving as a scanning lens and a second achromatic doublet serving as a tube lens. Further in this aspect, the scanned beam can be further relayed to a water immersion objective lens.

[0041] In one aspect, back- propagated emission light from the sample is separated from excitation light using a first dichroic mirror and sent to a signal capture means such as, for example, including, but not limited to, an sCMOS camera. In a further aspect, the sCMOS camera is used to look at a point spread function (PSF) shape or separated with a second dichroic mirror and sent to one or more photomultiplier tubes. In a still further aspect, light emitted by the sample is captured by the sCMOS camera through a filter for a second harmonic generation (SHG) guidestar. In an alternative aspect, another detector (e.g. a photomultiplier tube) for SHG or fluorescence can be used. In one aspect, a third dichroic mirror and a fourth dichroic mirror and one or more filters are used to separate spectral channels to capture signals from the illuminated sample. In any of these aspects, the one orATTORNEY DOCKET NO. 222105-2420 more filters can be a 571 / 72 nm filter, a 509 / 22 nm filter, a 390 / 18 nm filter, or any combination thereof. In some aspects, another method can be used to capture the PSF and / or to collect signal, including, but not limited to, another camera or sensor type (charge-coupled device or CCD; complementary metal-oxide-semiconductor or CMOS sensors, electron multiplying CCD), visualization through a lens with or without computer image capture, comparison to a theoretical PSF using a software model, comparison of sample signals and images to experimental PSF for a given set of conditions and equipment where fluorescent beads or quantum dots were used as a baseline, and the like, including but not limited to a PMT or hybrid PMT or single-photon avalanche photodiode. In one aspect, this list is exemplary only and is not intended to be exhaustive and any method for capturing PSF and / or collecting signal known in the art is contemplated and should be considered disclosed.

[0042] In an optional aspect, the beam or components thereof can pass through a Shack- Hartmann wavefront (SHWF) sensor comprising a lenslet array to measure total system aberration prior to the beam encountering an objective lens. In another aspect, the lenslet array creates spots on the image, wherein displacement of the spots is compared to an internal reference guidestar that allows calculation of wavefront distortions. In still another aspect, one correction is performed at each focal plane of the DM. In one aspect, and without wishing to be bound by theory, the SHWF sensor is primarily useful in making system corrections. Thus, in some aspects, the SHWF sensor is not necessary and / or is not used. In an aspect, the method further includes decomposing wavefront distortions into Zernike modes. In a further aspect, the method further includes using the DM to correct for Zernike modes 5 to 37. In one aspect, Zernike decomposition of the wavefront identifies contributions to distortion from tip, tilt, and defocus, and wherein Shack-Hartmann wavefront correction is applied to correct the distortions caused by tip, tilt, and defocus.Method for Imaging Complex Tissue

[0043] In one aspect, provided herein is a method for imaging complex tissue. In a further aspect, the method reduces distortions in the tissue. Further in this aspect, the distortions can be aberrations generated by non-homogeneous wave propagation through the complex tissue. In another aspect, the distortions are non-uniform in the complex tissue.

[0044] In any of these aspects, the method can be completed in one minute or less, or in 45 seconds, 30 seconds, or 15 seconds or less, or a combination of any of the foregoing values, or a range encompassing any of the foregoing values.

[0045] In some aspects, the method can be performed in a histological tissue section. In an alternative aspect, the method can performed in a living subject such as, for example, a mammal. In one aspect, the mammal can be a human, non-human primate, cat, dog, horse,ATTORNEY DOCKET NO. 222105-2420 cattle, swine, sheep, goat, mouse, rat, rabbit, guinea pig, or hamster. In any of these aspects, performing the method does not induce inflammation or another immune response in the subject.

[0046] In one aspect, the complex tissue can be bone, bone marrow, brain, another complex tissue, or any combination thereof. In a further aspect, the method enables visualization of osteocytes, mitochondria, or both osteocytes and mitochondria. In one aspect, the method is capable of internally visualizing the complex tissue at a depth of up to 250 pm.Method for Detecting a Disease or Monitoring the Progress of a Disease

[0047] In one aspect, disclosed herein is a method for detecting a disease or monitoring the progress of a disease in a subject, the method including performing multi-photon imaging on an area of interest in the subject, wherein the area of interest includes complex tissue, and performing the disclosed method as described above to correct distortions. In one aspect, the method can be completed in one minute or less or in 45 seconds, 30 seconds, or 15 seconds or less, or a combination of any of the foregoing values, or a range encompassing any of the foregoing values. In one aspect, in the disclosed method, the field of view is at least one order of magnitude of larger than in previous methods, and can be up to 60 to 120 pm2, with corrections applied consistently across the whole field of view.

[0048] In some aspects, the method can be performed in a histological tissue section. In other aspects, the method can be performed in a living subject such as, for example, a mammal. In one aspect, the mammal can be a human, non-human primate, cat, dog, horse, cattle, swine, sheep, goat, mouse, rat, rabbit, guinea pig, or hamster. In any of these aspects, performing the method does not induce inflammation or another immune response in the subject.

[0049] In one aspect, the complex tissue can include bone, bone marrow, or a combination thereof. In another aspect, the method enables visualization of osteocytes, mitochondria, or both osteocytes and mitochondria. In some aspects, the method is capable of internally visualizing the complex tissue at a depth of up to 250 pm.

[0050] In any of these aspects, the disease can be osteoporosis, osteomalacia, Paget’s disease, hypophosphatasia, aging-related bone changes, X-linked hypophosphatemia, bone repair after injury, or osteogenesis imperfecta. In one aspect, the area of interest includes a fluorophore such as, for example, green fluorescent protein (GFP). In one aspect, when the fluorophore is GFP, the method includes using a 457 / 50 nm bandpass filter to capture images of GFP in the area of interest. In some aspects, other bandpass filters can be used, as can combinations or ranges of different bandpass filters, depending on the fluorophore chosen, the tissue being imaged, and similar considerations.ATTORNEY DOCKET NO. 222105-2420

[0051] In another aspect, the fluorophore can be a small molecule, a protein, a quantum dot, or another fluorophore. In a further aspect, when the fluorophore is a protein, it can be expressed endogenously by the subject or can be introduced to the subject through gene editing or another means. In one aspect, the fluorophore can be introduced to the subject by administering the fluorophore to the subject either systemically, locally, topically, or the like. In an alternative aspect, when a histological specimen is imaged, the histological specimen can be contacted with the fluorophore prior to imaging.

[0052] Many modifications and other embodiments disclosed herein will come to mind to one skilled in the art to which the disclosed compositions and methods pertain having the benefit of the teachings presented in the foregoing descriptions and the associated drawings. Therefore, it is to be understood that the disclosures are not to be limited to the specific embodiments disclosed and that modifications and other embodiments are intended to be included within the scope of the appended claims. The skilled artisan will recognize many variants and adaptations of the aspects described herein. These variants and adaptations are intended to be included in the teachings of this disclosure and to be encompassed by the claims herein.

[0053] Although specific terms are employed herein, they are used in a generic and descriptive sense only and not for purposes of limitation.

[0054] As will be apparent to those of skill in the art upon reading this disclosure, each of the individual embodiments described and illustrated herein has discrete components and features which may be readily separated from or combined with the features of any of the other several embodiments without departing from the scope or spirit of the present disclosure.

[0055] Any recited method can be carried out in the order of events recited or in any other order that is logically possible. That is, unless otherwise expressly stated, it is in no way intended that any method or aspect set forth herein be construed as requiring that its steps be performed in a specific order. Accordingly, where a method claim does not specifically state in the claims or descriptions that the steps are to be limited to a specific order, it is no way intended that an order be inferred, in any respect. This holds for any possible non-express basis for interpretation, including matters of logic with respect to arrangement of steps or operational flow, plain meaning derived from grammatical organization or punctuation, or the number or type of aspects described in the specification.

[0056] All publications mentioned herein are incorporated herein by reference to disclose and describe the methods and / or materials in connection with which the publications are cited. The publications discussed herein are provided solely for their disclosure prior to the filing date of the present application. Nothing herein is to be construed as an admission that the presentATTORNEY DOCKET NO. 222105-2420 invention is not entitled to antedate such publication by virtue of prior invention. Further, the dates of publication provided herein can be different from the actual publication dates, which can require independent confirmation.

[0057] While aspects of the present disclosure can be described and claimed in a particular statutory class, such as the system statutory class, this is for convenience only and one of skill in the art will understand that each aspect of the present disclosure can be described and claimed in any statutory class.

[0058] It is also to be understood that the terminology used herein is for the purpose of describing particular aspects only and is not intended to be limiting. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the disclosed compositions and methods belong. It will be further understood that terms, such as those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the specification and relevant art and should not be interpreted in an idealized or overly formal sense unless expressly defined herein.

[0059] Prior to describing the various aspects of the present disclosure, the following definitions are provided and should be used unless otherwise indicated. Additional terms may be defined elsewhere in the present disclosure.Definitions

[0060] As used herein, “comprising” is to be interpreted as specifying the presence of the stated features, integers, steps, or components as referred to, but does not preclude the presence or addition of one or more features, integers, steps, or components, or groups thereof. Moreover, each of the terms “by”, “comprising,” “comprises”, “comprised of,” “including,” “includes,” “included,” “involving,” “involves,” “involved,” and “such as” are used in their open, non-limiting sense and may be used interchangeably. Further, the term “comprising” is intended to include examples and aspects encompassed by the terms “consisting essentially of” and “consisting of.” Similarly, the term “consisting essentially of” is intended to include examples encompassed by the term “consisting of.

[0061] As used in the specification and the appended claims, the singular forms “a,” “an” and “the” include plural referents unless the context clearly dictates otherwise. Thus, for example, reference to “a histological tissue section,” “a fluorophore,” or “an osteocyte,” include, but are not limited to, mixtures, combinations, or series of two or more such histological tissue sections, fluorophores, or osteocytes, and the like.

[0062] It should be noted that ratios, concentrations, amounts, and other numerical data canATTORNEY DOCKET NO. 222105-2420 be expressed herein in a range format. It will be further understood that the endpoints of each of the ranges are significant both in relation to the other endpoint, and independently of the other endpoint. It is also understood that there are a number of values disclosed herein, and that each value is also herein disclosed as “about” that particular value in addition to the value itself. For example, if the value “10” is disclosed, then “about 10” is also disclosed. Ranges can be expressed herein as from “about” one particular value, and / or to “about” another particular value. Similarly, when values are expressed as approximations, by use of the antecedent “about,” it will be understood that the particular value forms a further aspect. For example, if the value “about 10” is disclosed, then “10” is also disclosed.

[0063] When a range is expressed, a further aspect includes from the one particular value and / or to the other particular value. For example, where the stated range includes one or both of the limits, ranges excluding either or both of those included limits are also included in the disclosure, e.g. the phrase “x to y” includes the range from ‘x’ to ‘y’ as well as the range greater than ‘x’ and less than ‘y’. The range can also be expressed as an upper limit, e.g. ‘about x, y, z, or less’ and should be interpreted to include the specific ranges of ‘about x’, ‘about y’, and ‘about z’ as well as the ranges of ‘less than x’, less than y’, and ‘less than z’. Likewise, the phrase ‘about x, y, z, or greater’ should be interpreted to include the specific ranges of ‘about x’, ‘about y’, and ‘about z’ as well as the ranges of ‘greater than x’, greater than y’, and ‘greater than z’. In addition, the phrase “about ‘x’ to ‘y’”, where ‘x’ and ‘y’ are numerical values, includes “about ‘x’ to about ‘y’”.

[0064] It is to be understood that such a range format is used for convenience and brevity, and thus, should be interpreted in a flexible manner to include not only the numerical values explicitly recited as the limits of the range, but also to include all the individual numerical values or sub-ranges encompassed within that range as if each numerical value and sub-range is explicitly recited. To illustrate, a numerical range of “about 0.1% to 5%” should be interpreted to include not only the explicitly recited values of about 0.1% to about 5%, but also include individual values (e.g., about 1%, about 2%, about 3%, and about 4%) and the sub-ranges (e.g., about 0.5% to about 1.1%; about 5% to about 2.4%; about 0.5% to about 3.2%, and about 0.5% to about 4.4%, and other possible sub-ranges) within the indicated range.

[0065] As used herein, the terms “about,” “approximate,” “at or about,” and “substantially” mean that the amount or value in question can be the exact value or a value that provides equivalent results or effects as recited in the claims or taught herein. That is, it is understood that amounts, sizes, formulations, parameters, and other quantities and characteristics are not and need not be exact, but may be approximate and / or larger or smaller, as desired, reflecting tolerances, conversion factors, rounding off, measurement error and the like, and other factors known to those of skill in the art such that equivalent results or effects are obtained. In someATTORNEY DOCKET NO. 222105-2420 circumstances, the value that provides equivalent results or effects cannot be reasonably determined. In such cases, it is generally understood, as used herein, that “about” and “at or about” mean the nominal value indicated ±10% variation unless otherwise indicated or inferred. In general, an amount, size, formulation, parameter or other quantity or characteristic is “about,” “approximate,” or “at or about” whether or not expressly stated to be such. It is understood that where “about,” “approximate,” or “at or about” is used before a quantitative value, the parameter also includes the specific quantitative value itself, unless specifically stated otherwise.

[0066] As used herein, the term “effective amount” refers to an amount that is sufficient to achieve the desired modification of a physical property of the composition or material. For example, an “effective amount” of a fluorophore refers to an amount that is sufficient to achieve the desired improvement in the property modulated by the formulation component, e.g. achieving the desired level of visibility of the tissue structure, cell, or organelle to be visualized. The specific level in terms of wt% in a composition required as an effective amount will depend upon a variety of factors including the depth of the tissue to be visualized, the body weight of the subject, whether the tissue sample is a histological section or part of a living organism, and the like.

[0067] As used herein, the terms “optional” or “optionally” means that the subsequently described event or circumstance can or cannot occur, and that the description includes instances where said event or circumstance occurs and instances where it does not.

[0068] As used herein, “Genetic Algorithm” refers to the process described as follows: An initial population of parent masks (represented by a one-dimensional binary array) is introduced. This pattern consists of S segments, each segment represented by pj which is set to 1 (on) or 0 (off). For each mask being projected onto the DMD surface, the sample emission is recorded, and the fitness of each mask, quantified as the maximum intensity (gray level pixel value) within a specified region of interest, is recorded. Each mask is evaluated and ranked according to the fitness metric, where higher intensity returns a higher rank. An initial population of parent masks (represented by a one-dimensional binary array) is introduced. This pattern consists of S segments, each segment represented by Pj which is set to 1 (on) or 0 (off). For each mask being projected onto the DMD surface, the sample emission is recorded, and the fitness of each mask, quantified as the maximum intensity (gray level pixel value) within a specified region of interest, is recorded. Each mask is evaluated and ranked according to the fitness metric, where higher intensity returns a higher rank. Subsequent generations are created by selecting the top N / 2 ranked masks which are propagated to the new generation and used to produce G new offspring (G = N / 2) by crossover and mutation. The crossed-over genes are then mutated according to the mutation rate R, defined as R =ATTORNEY DOCKET NO. 222105-2420(Ro-Rena) • + Rend’ where Ro is the initial mutation rate, Rend, the final mutation rate, n is the generation index, and A is the decay factor. Where the search identifies that simply switching more segments on will increase the intensity, the total number of on pixels within a mask is limited so that .n=i Pj does not far exceed 50%. This is implemented by identifying each on pixel and randomly switching enough of them off to return the total number of on pixels to ~50%. This added level of mutation has the potential to disrupt well-established masks and could delay convergence, but with well selected mutation parameters, this impact is negligible. The newly generated masks are applied to the DMD sequentially and the process repeated for a fixed number of generations. Simulation of the search was conducted by generating a modeled correction pattern (binarized vector) as a target solution and having the genetic algorithm search from a random pattern to the target using Euclidean distance as the selection metric. Using this simulation, the impact of the various genetic algorithm parameters on the optimization speed (generations) and accuracy was investigated, defined as the Euclidean distance from the final algorithm output to the target solution. A comparison between random and fixed crossover points showed that random crossover points reduced time to convergence and yielded higher accuracies with or without a decay rate of A. Regardless of crossover, the most critical parameter was a balance between the initial mutation rate Ro and the decay rate A. Without any decay A, convergence is rapid, yet inaccurate, but when an appropriate decay rate A is implemented, the convergence is delayed and accuracies exceeding 80% can be achieved. Therefore, these factors must be balanced in experimental practice when using the algorithm to optimize the DMD pattern. In practice, evaluation of the performance of optimized genetic algorithm is measured by the intensity enhancement metric q, (which is defined by rj =,0Ptimizedgiven in full as:^referencewhere the intensity of the mthpixel is given by lm, thepixel is the PSF intensity, and a and b refer to the image after (a) and before (b) correction. In the case of before, the initial pattern is set to 50% segments on, and in the after state, the number of on segments is limited to ~50%. For experimental implementation of the GA in both transmission and epi-detection setups, each pattern is comprised of a number of DMD pixels arranged in a circle that is aligned with the incident elliptical excitation beam. The diameter of this active area remains constant (across 512 DMD micromirrors) while the degrees of freedom vary from low, set to a total of 373 segments each comprised of 24x24 DMD pixels, medium (1436; 12x12 super pixels), and high (5785; 6x6 super pixels). The strict limit on the number of segments was setATTORNEY DOCKET NO. 222105-2420 to 196 of 373 segments to limit the optimization from significantly exceeding 50% on segments.

[0069] Unless otherwise specified, temperatures referred to herein are based on atmospheric pressure (i.e. one atmosphere).

[0070] Now having described the aspects of the present disclosure, in general, the following Examples describe some additional aspects of the present disclosure. While aspects of the present disclosure are described in connection with the following examples and the corresponding text and figures, there is no intent to limit aspects of the present disclosure to this description. On the contrary, the intent is to cover all alternatives, modifications, and equivalents included within the spirit and scope of the present disclosure.EXAMPLES

[0071] The following examples are put forth so as to provide those of ordinary skill in the art with a complete disclosure and description of how the compounds, compositions, articles, devices and / or methods claimed herein are made and evaluated, and are intended to be purely exemplary of the disclosure and are not intended to limit the scope of what the inventors regard as their disclosure. Efforts have been made to ensure accuracy with respect to numbers (e.g., amounts, temperature, etc.), but some errors and deviations should be accounted for. Unless indicated otherwise, parts are parts by weight, temperature is in °C or is at ambient temperature, and pressure is at or near atmospheric.Example 1 : MethodsPreparation of fluorescent beads stack in gel

[0072] 200 nm yellow-green, fluorescent beads (ThermoFisher Scientific F8811) were diluted to a ratio of 1 :200 in 2.0% agarose. Then, 0.2 grams of agarose powder (Bio-Rad, Certified Molecular Biology Agarose 1613101) were added to 10 milliliters of DI (Deionized) water to make the mass concentration 2.0%. Next, the agarose solution was heated in a microwave for intervals of 45 s. This process was continued until the agarose became a gel mixture. After 1-2 min passed for the agarose gel to cool down to a safe handling temperature, the agarose gel was ready to hold the beads in place for imaging analysis. In total, 2.0 mL of beads were added to 400 mL of agarose to achieve the 1 :200 ratio. After drying completely, the Petri dish was then placed on the stage of the TPFM-AO system and DI water was added for imaging with the water dipping objective.

[0073] During evaluation of fluorescent and Potassium Titanyl Phosphate (KTP, with formula KTiOPC t) samples, translucent tape (70005119527, 3M) was used to diffuse the excitation and emission light. The tape scattering length was measured to be £ss= 0.052 m-1, and aATTORNEY DOCKET NO. 222105-2420 thickness of approximately 30 pm. Various layers were applied to the phantom samples between the objective lens and the sample.

[0074] The epi-detection microscope setup was evaluated using three phantom samples. A thin layer of maleimide was mounted onto a microscope slide and imaged through a coverslip. This thin fluorescent sample was evaluated under conditions of one and three layers of translucent tape applied as diffuser. The extended fluorescent phantom sample, a 1.2 mm thick autofluorescent plastic slide (92001 , Chroma Technology), which provided bright 2- photon emission in the range of 585 nm, when excited by 780 nm, was used to evaluate the enhancement with 3 layers of translucent tape applied. To test the SHG compatibility with the setup, a KTP crystal (KTP-403H, EKSMA Optics) replaced the fluorescent samples, and was tested with one layer of tape acting as the diffuser.Mouse imaging

[0075] For mouse intravital imaging, a transgenic mouse model ubiguitously expressing mitochondrial-targeted Dendra2 green monomeric fluorescent protein (Jackson Laboratory, #018385) was used as previously described. The mouse was initially anesthetized using 4% isoflurane (100 mL / min oxygen flow) and restrained using a 3D printed stereotaxic holder. The holder is similar to those used in previously published works by several groups when studying cell dynamics in the brain and skull and serves to secure and stabilize the mouse skull while reducing mechanical coupling with the trunk of the body so that breathing movement artifacts are reduced. Five minutes before making an incision, 50 pL of 0.25% bupivacaine was locally applied as analgesia. An incision was made on the scalp from between the eyes toward both ears to make a flap. The periosteum layer was removed, and the area of imaging was cleaned using a cotton swab; immediately sterile phosphate buffered saline (PBS) was applied to the incision site. The animal was placed under the microscope objective and sterile PBS was added to fill the gap between the skull and the objective lens. The rate of isoflurane was then reduced to 1.4% during imaging. For vasculature imaging, a 20 pL dose of 70 kDa rhodamine- B dextran (Nanocs) was administered through retro-orbital injection before making an incision. After the intravital imaging session, mice were euthanized using CO2 and cervical dislocation. Immediately after sacrifice, the brain was extracted and mounted in a Petri dish using 2.0% agarose for mitochondrial imaging. All animal procedures and experiments were approved by the UGA Institutional Animal Care and Use Committee (IACUC).

[0076] For biological examination of the scatter correction, a freshly excised, fifteen-week-old PhAM floxed (photo-activatable mitochondria) skull sample (#018385, The Jackson Laboratory) was fixed in 4% paraformaldehyde (PFA) and 96% phosphate-buffered saline (PBS) for 48 hours and then mounted in a Petri dish using a PDMS elastomer (Sylgard 184,ATTORNEY DOCKET NO. 222105-2420Dow Corning). IACUC was strictly adhered to in treatment of animal samples. A manual 3- axis stage was used to prevent spurious movement during the correction and point-scan imaging.Genetic Algorithm

[0077] Parameters important to the genetic algorithm are discussed fully in Example 3. Details regarding the segmentation of the DMD during optimization are laid out here. Each segment as set before running the search is comprised of individual DMD micromirrors arranged in a square. These squares are arranged in a circle that is manually set to align with the incident excitation beam. The size of these squares defines the number of segments within the larger circular area on the DMD. For example, setting the segment size to 24x24 pixels, results in an active area on the DMD with 373 controllable segments. This is limited by the diameter of the active region, measured in individual micromirrors, set in all cases presented in this work to 512. As the segment size is reduced, the number of controllable segments increases, while the circular area (diameter of 512 micromirrors) affecting the wavefront remains constant. Dividing the active area in to segments of size 24x24 yields 373 total segments, 12x12 segment size results in 1436 segments, and 6x6 segment size divides the active area into 5785 segments.Optical setup

[0078] A schematic of the adaptive optics two-photon fluorescence microscopy (AO-TPFM) system is shown in FIG. 1 and includes a DMD device as shown. The optical setup consists of a Chameleon Ti:Sapphire laser producing 680 to 1 ,080 nm 137 fs pulses of energy, with power at the source modulated using a half-waveplate and a polarizing beam splitter. The laser beam is recollimated and passed through a Pockels cell (Conoptics) for rapid intensity modulation during imaging. The beam is then expanded using a telescope with a pinhole in the focus to create a more uniformly Gaussian beam profile. The Pockels cell is used to block the beam during flyback, as well as correct beam intensity for the resonant scanner. The deformable mirror (DM, Alpao DM97-15) placed conjugate to the back pupil plane has a continuous face sheet and 97 actuators and is used for low-order AO wavefront correction. The DM is inserted into the beam path with two optical relay systems. Beam scanning is done by a Sutter instrument MDR-R box that houses a fast resonant galvo and a slow galvo scanner for horizontal and vertical sweeping, respectively. Both scanners are from Cambridge technology and are placed very close to each other to reduce astigmatism, with heatsinks. The scanned beam is relayed using two achromatic doublets — serving as the scanning and tube lenses, respectively — to the 60 x 1.00 NA water immersion objective lens (Nikon, MRD07620). The back-propagated emission light from the sample is separated from theATTORNEY DOCKET NO. 222105-2420 excitation light using a dichroic mirror DiM1 (Semrock FF705-Di01) and sent to an sCMOS camera (AN DOR, Zyla Scientific CMOS) which is used to look at the PSF shape or separated with a dichroic mirror DiM2 (Semrock FF705-Di01) and sent to the photon multiplier tubes (PMT) from Hamamatsu (H10770-40). DiM3-4 (Semrock FF552-DiO2, and FF409-Di03-25*36) and filters F2-4 (Semrock 571 / 72 nm, 509 / 22 nm, and 390 / 18 nm) were used to separate each spectral channel to capture signals from two photon fluorescence (TPF), green fluorescent protein (GFP), and second harmonic generation (SHG) of collagen, respectively. A home-built Shack-Hartmann wavefront sensor (SHWFS) was used, with a LabVIEW-based control and measurement software to measure the total system aberrations (red arrow shows beam direction) just before the objective lens. Another LabVIEW-based control and measurement software was used for full AO correction. The MATLAB -based open-source software, Scanimage was employed to control the microscope after the correction.

[0079] The custom microscope engages a Ti:Sapphire laser (Chameleon Ultra II, Coherent) for excitation in the wavelength range 680nm - 1080nm, and is illustrated in FIG. 1. Output from the source laser is modulated using a Pockels cell (350-105, Conoptics) and immediately thereafter, a pinhole is used to improve the Gaussian properties of the beam. The beam is expanded to utilize approximately 30% of the DMD (ALP-4.3 (V4395), ViALUX) active array surface, within the 10.5mm height limit of the active array. A resonant-galvanometer scanner (MDR-R, Sutter Instrument) allows for scanned imaging, and finally, a 60x water immersion objective lens (LUMFLN60XW, Olympus) focuses the light onto the sample. To measure and optimize the DMD pattern, the sample-emitted light is captured by the sCMOS camera (Zyla 4.2, Andor, Oxford Instruments) through an appropriate filter for the SHG guidestar (FF01- 390-18, Semrock) when excited at 780nm, by moving the dichroic mirror DiM1 to its appropriate position. Capturing an image of the sample over a wider field of view, DiM1 is moved to reflect emitted light to be collected by the photon multiplier tubes (PMTs) (H 10770- 40, Hamamatsu Photonics K.K.), using trans-impedance amplifiers (59-179, Edmund Optics), and scanimage controls the resonant scanner and FPGA (PXIe-7975R and NI5734, National Instruments). To capture images of GFP in the murine samples, an appropriate optical bandpass filter was used (FF01-457 / 50, Semrock). The sCMOS camera and DMD are controlled by a custom LabVIEW (National Instruments) program. All of the epi- detection experiments conducted included lower-order adaptive optics, with the system correction turned on, as per previous adaptive optics work within the group.Sensor-based system aberrations

[0080] Before measuring and correcting the aberrations of the biological samples, system aberrations were first compensated for using a sensor-based AO algorithm. The lenslet arrayATTORNEY DOCKET NO. 222105-2420 creates spots in the image whose displacement versus an internal reference guidestar allows calculation of wavefront distortions.

[0081] Eq. (1) is used to decompose the calculated wavefront into the Zernike modes:where Z is the Zernike mode of order i and ci is the coefficient of the mode Zi. Equation (1) yields a complete Zernike coefficient set that could be applied to a DM for correction. Modes 5 to 37 [using Noll’s ordering of the Zernike modes, up to order 4] are taken into consideration because these modes can be corrected by the DM. To find the corrected wavefront shape, a summation is done such that constructed phase equals

[0082] The root-mean-square (RMS) wavefront error that is corrected by the Zernike modes4 to 37 (piston, tip, and tilt are not included) is calculated by '

[0083] One DM correction is performed at each focal plane by scanning for a fluorescent signal and then optimizing the wavefront. The wavefront of the excitation beam shows <2 waves of distortion, with Zernike decomposition of the wavefront identifying the strongest contributions from tip (Z2), tilt (Z3), and defocus (Z4). After applying the Shack- Hartmann wavefront correction, the PSF in the sample plane has a near diffraction-limited Gaussian shape (FWHM = 350 nm).Sensorless sample aberration correction based on sum or max intensity

[0084] Wavefront aberrations are the difference in phase or optical path length from the ideal (e.g., spherical, or planar), which can be caused by light propagation through an inhomogeneous medium like biological tissue. According to the Zernike mode equation, different combinations of the Zernike coefficients in a phase distribution at the back-pupil plane can alter the point spread function at the focal plane. If used to reconstruct the wavefront phase distribution with proper Zernike modes and coefficients by a DM placed conjugate to the back-pupil plane, this principle can compensate for the aberrations induced by tissue.

[0085] Sensorless AO uses the signal obtained with the microscope as an input for an algorithm that estimates the optical aberrations present in the system. A series of PSFs is acquired with different Zernike aberration modes applied to a DM conjugate with the back pupilATTORNEY DOCKET NO. 222105-2420 plane. To determine the optimal value for each Zernike mode, different values of aberration are applied to the DM. 10 or 15 orders of Zernike modes (tip, tilt, and defocus excluded) were evaluated, which in preliminary experiments using 10 or 15 modes for correction provided 90%of the enhancement found when including higher order modes. An image quality metric (e.g., max intensity) is then selected and evaluated for each image. Then a parabolic function is fitted to the measured points and the mode coefficient corresponding to the estimated peak is applied as the correction. Subsequent modes are corrected in a similar manner to achieve convergence of the wavefront.Sample correction parameter selection

[0086] During preliminary aberration correction experiments, it was observed that the degree of enhancement achieved in sensorless sample correction has a dependence on the initial intensity conditions in low signal environments. Since a low starting signal is common when attempting to use AO, the sensitivity of this correction strategy to different background noise and signal levels was explored by modulating the laser power before the objective lens and the camera exposure time. These measures both strongly influence the signal to background ratio of the PSF images used in the sample correction for a GFP mouse skull. Two power levels were used at the sample, a commonly used average power on the sample for imaging 50-100 pm deep in tissue, and the maximum average power that was experimentally observed not to cause visible damage to the sample within the image acquisition time; and a range of integration times for each Zernike mode measure from 10 ms to 1 s. At each condition, several measures of PSF quality were evaluated, including the mean intensity, max intensity, second moment, and Strehl ratio. To calculate the second moment, the k-th central moment of a data sample is defined as:where n is the number ofsamples and x is the mean. The Strehl ratio is defined as ~where o is the root mean square deviation of the wavefront, A is the wavelength. In the present situation, the experimental Strehl ratio is defined as the ratio of the peak aberrated image intensity from a point source compared to the maximum attainable intensity using an ideal optical system limited only by diffraction over the system’s aperture.

[0087] Although improvement based on mean intensity was not achieved, improvement based on max value, second moment, and Strehl ratio of PSF was achieved when the exposure time is larger than 0.5 s, with the percent changes between the uncorrected and corrected value reaching a constant soon after. Therefore, it was found that by performing the TPFM-AO approach at two distinct signal levels using mean or max intensity of PSF when theATTORNEY DOCKET NO. 222105-2420 improvement percentage of two of the measurements are almost the same, the sample correction has achieved its maximum improvement performance.Example 2: Results and DiscussionIn vitro tissue mimic with AO correction for submicron bead imaging

[0088] The experimental resolution improvement in a tissue phantom of the TPFM-AO microscope was evaluated by measuring the full-width at half-maximum (FWHM) of the intensity profile of 0.2 pm beads embedded in a gel using a 960-nm laser excitation wavelength (FIGs. 2A-2E). Beads were embedded in a 5 mm thick 2% agarose gel and imaged at 50 pm depth (FIG. 2A) to mimic a distorting tissue environment. The sample was imaged with system correction on, and then used a sensorless approach to compensate for the aberrations induced by the gel (FIG. 2B). The full AO correction approach shows relatively high values of astigmatism (Z6) and coma (Z7) and yields the wavefront shown in FIGs. 2C- 2D. AO improved the full width at half maximum of the detected bead fluorescence with the average FWHM Gaussian fit of 10 measured 0.2 pm radius 2-photon excited fluorescent beads improving from 0.538 ± 0.03 pm to 0.408 ± 0.03 pm after sample correction (FIG. 2E).AO correction enables high-resolution imaging of mitochondria organelle morphology in the mouse brain

[0089] Aberrations were then measured and corrected for Dendra- 2 mouse brain mitochondria using a 780 nm laser excitation wavelength. Immediately after sacrifice, the mouse brain was cut into 2 mm thick slices, then embedded in 3% agarose to prevent movement. For imaging, the brain was immersed in phosphate buffered saline (PBS). Two photon fluorescence images were acguired in the hippocampal region, as indicated by a red asterisk in the brain diagram in FIG. 3G. Tissue wavefront distortions (FIG. 3E) and aberrations were found due to the shape and high refractive index of the brain; mostly astigmatism (Z5, Z11), trefoil (Z9), and spherical (Z11) (FIG. 3F). After AO, the images had improvements in mitochondrial intensity and sharpness laterally and, especially, axially by correcting the aberrations (FIGs. 3A-3B). In the spatial freguency space (FIG. 3C), the resolution improvement gained through aberration correction led to a substantial increase in the magnitude of high spatial freguency components, which indicates a sharper image with more fine detail. Signal profiles in the axial plane along the white lines show improved intensity (FIG. 3D).AO correction enables mitochondria imaging deep in mouse bone marrow in vivo

[0090] Mitochondrial imaging was next evaluated through the outside layer of cranial bone into the bone marrow using the Dendra-2 mouse with 780 nm laser excitation wavelength. 15ATTORNEY DOCKET NO. 222105-2420 orders of Zernike modes were evaluated to correct aberrations of the bone marrow in vivo because in preliminary experiments this achieved the best improvement of the PSF intensity and shape. Improvements in image intensity and resolution for in vivo TPF imaging of mitochondria in bone marrow were found at the depth of 0, 50, and 85 pm. Signal profiles along the red line show ~1.55x, ~3.58x, and ~1.77x intensity increases using sample AO correction at the depths of 0, 50, and 85 pm separately. The FWHM with full AO was enhanced by ~0.83x, ~0.74x and ~0.9x at the depth of 0, 50, and 85 pm separately. At the surface of the bone, aberrations due to the shape and high refractive index of the bone were observed; mostly astigmatism (Z6), trefoil (Z9), and secondary astigmatism (Z13). When going 50 pm deep in the bone marrow, the aberrations of primary astigmatism and trefoil are lower magnitude, but secondary astigmatism (Z13) remains. When reaching the opposite side of the bone marrow at a depth of 85 pm, aberrations are mainly coma (Z7) and secondary astigmatism (Z13) resulting from the curved interior surface of the bone.

[0091] When averaging across a whole image, the signal intensity showed ~2.73x, ~3.13x, and ~2.18x intensity increases over TPFM with system AO correction at 0, 50, and 85 pm depth (FIG. 4A). The average detected FWHM improvement was ~0.79X(1.0 pm before and 0.79 pm after), ~0.78x(0.8 pm before and 0.62 pm after), and -0.81 x(0.45 pm before and 0.36 pm after) at the depth of 0, 50, and 85 pm respectively (FIG. 4B). In order to evaluate the effect of AO on other emission wavelengths and confirm that the 85 pm depth was through the bone marrow and into the bone on the other side, second harmonic generation images were captured with a bandpass filter of 390 / 18, finding that the FFT spectrum contains more high frequency information after full AO.

[0092] Another multichannel dynamic in vivo bone marrow sample was also evaluated using the Dendra2 mitochondria mouse and co-labeling the blood vasculature using a rhodamine-B dextran conjugate. This stain required correction of the excitation beam using 840 nm excitation wavelength and emission with a 585 / 40 nm filter. Improved imaging was achieved with full AO correction at three different depths of 30, 50, and 70 pm, demonstrating the utility of the system for AO in dynamic samples at multiple excitation and emission wavelengths.AO correction for dynamic mitochondria evaluation in mouse bone marrow

[0093] To evaluate the potential of AO correction to longitudinally monitor mitochondrial organelle dynamics in the bone, tissue aberration was corrected at a single plane 40 pm deep in the bone marrow and then time lapse imaging performed for a total of 20 min. The temporal change in the cell mitochondria position was quantified (FIGs. 5A-5D) and differences in mitochondrial movement rates and trajectories were observed (insets in FIGs. 5A-5B). Minimal intensity reduction was found over the imaging session, suggesting that there wasATTORNEY DOCKET NO. 222105-2420 little photobleaching or change in the tissue aberrations over 20 min (FIGs. 5A-5D). A variety of cell and mitochondria trajectories are clearly present within relatively small regions of the bone marrow, which is not surprising given the high cellular density and diversity of cell types.Conclusion

[0094] In summary, both the system aberrations and the sample aberrations caused by highly scattering brain tissue and bone were calculated and corrected with both a sensor-based approach using a Shack-Hartmann wavefront sensor and a sensorless AO approach using the PSF intensity as a metric. Herein it was demonstrated that low-order aberration correction provides a significant improvement when imaging through the bone into the living bone marrow. It was found that the TPFM-AO system increases the fluorescence intensity of the PSF and achieves fast imaging of subcellular organelles with -400 nm resolution. Close to a 2 fold increase in intensity and a reduction in PSF width using AO were also achieved in living mouse bone marrow, allowing better characterization of mitochondrial health and the survival of functioning cells. The criteria for stopping iterative sensorless AO correction were also determined, finding that once the improvement percentage of two measurements is near constant, sample correction has reached its best performance. This helps to more rapidly and efficiently correct the PSF. This AO approach could be used for the study of the dynamics of other organelles and is applicable to a wide range of biological tissues. In the future, alternate correction metrics will further be explored (e.g., second moment) for in vivo situations and use this system to study the transient functional responses in a cell population deep in the bone marrow previously unreachable by optical microscopy. This could transform MSC therapeutic approaches and enable new fundamental biological understanding in the musculoskeletal and neural fields.Example 3: Genetic Algorithm

[0095] To correct the wavefront by means of electro-optical components, the optimal pattern must be determined using a search heuristic, such as sequential, parallel, genetic algorithm, or machine learning. Genetic algorithms are inspired by nature, incorporating the principles of genetic evolution to explore a given search space for an optimal solution. Genetic algorithms can be applied to problems where the solution can be constructed by perceived genes. In the case of binary wavefront modulation using a DMD, each pixel, considered a gene, forms part of the binary segments and constructs an optimal pattern to manipulate the wavefront. Further requirements for implementation of genetic algorithms are: the ability to generate random populations of the individuals, the ability to breed the individuals by cross-over (random selection) of genes from each parent, the ability to mutate genes, and selection criteria byATTORNEY DOCKET NO. 222105-2420 means of a metric mechanism. For the present case, the selection metric is the intensity of the point spread function (PSF).

[0096] An in-depth discussion regarding Genetic Algorithm (GA) search heuristics can be found in the literature, while the foundational GA implemented in this solution is illustrated in FIGs. 6A-6E. In summary, an initial population of parent masks (represented by a 1- dimensional binary array) is introduced. This pattern consists of S segments, each segment represented by Pj which is set to 1 (on) or 0 (off). For each mask being projected onto the DMD surface, the sample emission is recorded, and the fitness of each mask, quantified as the maximum intensity (gray level pixel value) within a specified region of interest, is recorded. Each mask is evaluated and ranked according to the fitness metric, where higher intensity returns a higher rank.

[0097] Subsequent generations are created by selecting the top N / 2 ranked masks which are propagated to the new generation, and used to generate G new offspring (G = N / 2) by crossover and mutation. The crossed-over genes are then mutated according to the mutation rate R, defined as R = (Ro - Rend)xe('n / A)+ Rend, where Ro is the initial mutation rate, Rend, the final mutation rate, n is the generation index, and A is the decay factor. Should the total number of on pixels within a masksn=i Pj exceed 52.5%, another stage of mutation is implemented, whereby all on pixels are identified and randomly switched off so that the total number of on pixels returns to 52.5% or below. This added level of mutation has the potential to disrupt well- established masks and could delay convergence. The newly generated masks are applied to the DMD sequentially and the process is repeated for a fixed number of generations.

[0098] In simulations, a comparison between random and fixed crossover points showed that random crossover points reduced time to convergence and higher accuracies whether or not a decay rate of A was implemented (FIGs. 6B-6E). Regardless of crossover, the most critical parameter was a balance between the initial mutation rate Roand the decay rate A. Without any decay A, convergence is rapid, but not accurate (FIGs. 6B, 6D), but when an appropriate decay rate A is implemented, the convergence is delayed, but exceeds accuracies of 80% (FIGs. 6C, 6E). Balancing these factors must be taken into account when using the algorithm to optimize the DMD pattern.

[0099] The performance of the optimized output generated by the genetic algorithm is _ ‘optanszea = evaluated by the intensity enhancement metric q, (which is defined by reference given in full as:ATTORNEY DOCKET NO. 222105-2420where the intensity of the mthpixel is given by lm, the kthpixel is the spot intensity, and a and b refer to the image after (a) and before (b) correction. In the case of before, the initial pattern is set to 50% segments on, and in the after state, the number of on segments is limited to 52.5%.Transmission Geometry

[0100] An initial evaluation of the Genetic Algorithm (GA), applied using binary wavefront modulation and a DMD, was performed in a transmission geometry setup, which has been widely used in scatter correction implementations. Each parameter within the configuration of the GA has a unique influence on the results of the search, to select mutation rates and crossover; many were examined until the parameter selection resulted in an optimization comparable to previous investigations that would be useful for the present investigations. Executing a fixed number of generations of the GA as described above, parameters set to Ro = 0.01 , Rend = 0.0025, and a decay rate of A = 50 were found to return enhancement that matched those found in the literature. It is important to note that calculations of the mutation ratio and decay depend on the number of generations, and should the number of generations be adjusted, the value of A should be selected to match the slope given by these values across 350 generations.

[0101] Comparing three pattern sizes, each with a different number of controlled segments, shows results that match current literature, with the measured enhancement being proportional to S (number of segments). This enhancement is guided by the theoretical maximum achievable enhancement in binary wavefront modulation of q = N / 2TT. Table 1 summarizes the imaging conditions applicable to each case.

[0102] The initial evaluation of the optimization parameters and optical layout employed a standard transmission geometry setup. 780nm light reflected by the DMD (ALP4.3 (V3495), ViaLux) was filtered in the Fourier plane by an iris, and then passed through a 10* air objective lens (Plan Fluor 10x / 0.30, Nikon) and through a coarse grit, ground glass diffuser (DG10-120-ATTORNEY DOCKET NO. 222105-2420MD, Thorlabs). The scattered output was collected by a 20* air objective lens (Plan Fluor 20x / 0.45, Nikon) and captured on an sCMOS camera (Zyla 4.2, Andor, Oxford Instruments). The lenses, diffusor, and camera were all mounted on a cage rod setup to secure the alignment of the critical components.Example 4: Results and Discussion

[0103] Implementation of biological imaging in transmission geometry is highly restrictive, not allowing for imaging of thicker tissue volumes or living samples. As such, the microscope setup used to confirm the optimization was adapted for epi-detection (see materials and methods) that allows for the imaging of thick biological samples. In order to establish epi-detection functionality, a number of phantom samples were evaluated before scatter correction was implemented within biologically relevant samples.Phantom Samples Correction

[0104] A thin fluorescent sample was evaluated under the conditions of one, and three layers of tape. FIG. 7A shows the measured relative enhancement of each phantom sample when 373 segments were employed in the optimization of the DMD pattern. Enhancement of the thin fluorescent sample is slightly reduced when the additional two layers of tape are applied to the sample, in both cases converging on an optimized solution after approximately 100 generations. Comparing enhancement measured using the thin fluorescent sample to the extended fluorescent sample, through three layers of tape, an increased enhancement was observed that converges at an q value of approximately 6 after approximately 150 generations. This difference can be attributed to the increased photon flux emitted by the extended sample due to a combination of sample brightness and more of the sample being excited by the focal volume. The enhancement measured in the KTP sample in FIG. 7A shows a similar trend with additional improvement from generation 150-300, before returning to an enhancement value similar to that of the extended fluorescent sample.

[0105] Resulting enhancements in FIGs. 7A-7C match the expectation set previously, with increased enhancement correlated to increased degrees of freedom. Similarly, the optimization convergence is delayed when more controllable degrees of freedom are available to the GA, which is most clear in FIG. 7C, where the enhancement has not converged in any of the phantom samples. While the gradient of each enhancement remains positive after 350 generations, the time to run 350 generations with 20 ms exposure time (and other overheads) exceeds 10 minutes, which becomes a significant translational drawback.

[0106] It was found that the role of initial intensity Io (in the present case, the first generation average intensity) or the number of photons measured in the target region of interest, is a restrictive factor in achieving consistent enhancement. Total signal in a target region of interestATTORNEY DOCKET NO. 222105-2420 consists of ballistic (unscattered) photons, sub-diffuse photons that are subjected to fewer scattering lengths, and fully scattered photons. Sufficient ballistic and sub-diffuse photons are necessary as a basis for the initial correction of the GA. May et. al. showed that in the specific case of a thin fluorescent sample, enhancement cannot be expected when Io is too low (< 10) but higher measurable enhancement is achieved when Io increases, and GA is most effective when Io reaches or exceeds 1000.

[0107] In this work, where the phantom samples are obscured by three layers of tape, photons experience approximately 4.5 scattering lengths, which has been classified as sub-diffuse. As more layers are applied, the number of ballistic photons is reduced, and thereby Io. The dependence of enhancement on l0accounts for the results such as FIG. 7C where l0is low and the segment count is higher. In this case the enhancement is unconverged after 350 generations, yet maintains a positive gradient. The effect of reduced Io can be corrected by setting it to be equal across each experimental case, or by increasing the excitation power. Integrating increased power and higher-order scatter correction could result in improved ablation techniques deeper within bone.

[0108] FIGs. 8A-8L show the average optimized DMD patterns (averaged from the population of 100 in the final generation) from each configuration of the active area (see material and methods). These optimized patterns are presented for each phantom sample evaluated.

[0109] When considering FIGs. 8C, 8F, 8I, and 8L, where 5785 segments were used in the optimization, it becomes clear why the GA converges later when using more segments. Each individual segment among the 5785 has a smaller impact on the intensity (or other evaluation metric) in the target spot than the equivalent individual segment among a total of 373. Where concentrated regions of on segments yield the optimal intensity (or other evaluation metric), lower segment counts reach the arrangement faster than in the case of more, smaller segment configurations.

[0110] In the case of KTP crystal-based phantom samples, which emit primarily forward SHG due to their phase matching properties there are few backscattered photons to collect, so the initial photon count in the optimization is lower than the fluorescent case. Therefore with 5785 segments used, even with fewer layers of translucent tape applied a much lower enhancement was achieved than in fluorescent results, as per FIG. 7C.

[0111] From results of each pattern presented in FIGs. 8A-8L, the correction patterns appear to concentrate on segments in a specific region of the DMD. This is indicative of a lower-order correction that may be the result of residual spherical aberration during the refractive index change that occurs as the excitation light enters the sample.ATTORNEY DOCKET NO. 222105-2420

[0112] The impact of slow correction times using the sCMOS camera (with correction times exceeding 10 minutes in the case of phantom samples), the low SHG intensity expected from biological samples (especially from within bone tissue), combine to reduce available options. The number of generations of the GA when using the sCMOS camera can be reduced, or a faster detector such as a PMT can be used for the evaluation metric within the GA. From the results of the correction within phantom samples, it is seen that 50 generations could yield satisfactory enhancement, which serves to reduce the correction time, even while exposure times are increased.

[0113] An alternative approach is also implemented, in which a PMT is used to evaluate each pattern and select according to the GA. By reducing the exposure time to 10 ms (potentially even 1 ms) per individual, the correction can be completed in under two minutes.High-frequency Wavefront Correction in Biological Samples

[0114] Using a label-free approach to imaging within the bone, the intrinsic SHG signal generated by the collagen in bone tissue is used. It was previously shown that the fraction of back-scattered SHG light that reaches the surface of a thick tendon tissue is approximately 21%. The measured F / B (Forward / Backward) ratio of SHG in that thick sample was approximately 4, so it can be deduced that approximately half of the emitted SHG at the surface is backward generated, and the other half is back-scattered. Similar ratios can be expected in bone samples which are comprised of similar collagen arrangements. During optimization, segments in the DMD mask that ’’switch on” photons that contribute to a stronger focus would be favored over segments that correct epi-directed photons. Latent lower-order corrections could be favored during optimization and although scattering is strong, the prevalence of lower-order aberrations, such as spherical aberration, would still require compensation.

[0115] An excised skull was next evaluated which require higher exposure times to collect sufficient photons and achieve a meaningful enhancement.

[0116] Illumination of the sample with 780 nm at approximately 300 mW and a depth of ±140 pm (well past the mean free distance for adult mouse bone tissue of ±45 pm) provides sufficient photons for the optimization to improve. The result was a correction time of approximately 15 minutes for a population size of 100 across 50 generations, 120 ms integration time, and overhead latency of approximately 2.5 minutes to record 373 segment individuals within the population for each generation. FIG. 9D shows the low signal in the average image of the initial generation, with the uncorrected mask projected onto the DMD in FIG. 9B. The enhancement over 50 generations can be seen in FIG. 8A, with the average image of the SHG spot in the bone in FIG. 9E, and the corrected DMD mask in FIG. 9C. TheATTORNEY DOCKET NO. 222105-2420 optimized mask in FIG. 9C indicates lower-order aberration correction, which could be a result of spherical aberration due to the shape of the skull. Low persistence times are associated with living tissue, and while correction in an excised skull would not suffer from such low persistence times, achieving apparent convergence within 50 generations is appealing for translation of the technique to living bone by reducing the optimization duration. The difference in the intensity profiles of FIGs. 9D-9E indicates a significant increase in the SHG spot in the bone after correction.

[0117] Optimized patterns generated by the scatter correction and presented in FIGs. 9A-9F are then projected onto the DMD for use in the point-scanning implementation, where both GFP and SHG channels can be observed simultaneously.

[0118] Resulting images at a depth of 140 pm, through the whole skull bone, using both uncorrected and corrected DMD patterns clearly illustrate the increased contrast in both GFP and SHG channels (FIGs. 10A-10I) when using the optimized pattern. At this depth, imaging the bone marrow in a 15 week old mouse, the collagen no longer resembles the condensed formation of bone with lacunae, but fibrillar features are more prevalent. The green fluorescent protein (GFP) present in the mitochondria of the murine sample (FIGs. 10B, 10D) are imaged separately from the SHG in bone and other collagen-based tissues (FIGs. 10C, 10E).

[0119] Applying the uncorrected DMD pattern results in low intensity in the GFP (FIG. 10B) channel and few visible spots of mitochondria, but application of the corrected DMD pattern returns increased intensity (FIG. 10D). A cross section of the identified line shows a 2-fold increase in intensity with new peaks indicating previously undiscerned mitochondria (FIG. 10F). From the center of the correction, the effective mean contrast improvement is reduced to 80% outside a radius of 400 pixels.

[0120] The SHG visible when the uncorrected DMD pattern is applied presents few features of interest (FIG. 10C), as opposed to the clear individual fibers revealed when applying the corrected DMD pattern (FIG. 10E). The cross section at the blue line in FIG. 10C shows increased intensity by a factor exceeding 2, but clearly more individually discernible fibers (FIG. 10G).

[0121] Measurement of image resolution was carried out using the Fourier Ring Correlation (FRC) BIOP Plugin for Imaged which requires two input images to calculate the FRC. The smoothed FRC outputs for each image were isolated and presented in FIGs. 12H-12I. It was found that the spatial frequency in the image of the GFP increased by 905 nm, while the increase in spatial frequency within the SHG images was 688 nm.

[0122] Collagen contains fine fibrillar structures that interconnect, filling the spaces between osteocytes, while the punctate nature of the mitochondria within the cells leads to theATTORNEY DOCKET NO. 222105-2420 expectation of improved spatial resolution for both channels. Scanned images using the corrected DMD mask allow clearer identification of the mitochondria and other tissues, enabling more studies of mitochondrial metabolism in deep tissue, and tissue structural organization.

[0123] By implementing a PMT as the detector for the evaluation of individuals of the DMD patterns within the GA, the scatter correction time was reduced. A relatively long exposure time of 100 ms was investigated and resulted in an intensity improvement of approximately 2.2 after 4 minutes. Noise from both the excitation source and the PMT result in noisy intensity measurements per individual, but after averaging, the steady improvement can be observed (FIG. 11 A). By reducing the exposure time to 10ms, the noise is more prevalent, and again the improvement is clear when the evaluation of each individual is averaged (FIG. 11 B). In the case of 10ms exposure time, the correction time is reduced to 80 seconds, and results in an intensity improvement of approximately 1.75.

[0124] Similar to the results presented in FIGs. 10A-10I, cells that were lost in the noise in the GFP channel are more clearly visible in FIGs. 12A-12I and the contrast is improved, confirmed by the cross section (FIGs. 12B, 12D, and 12F). The contrast is increased in the SHG channel when the corrected pattern is projected onto the DMD, while the overall intensity does not change significantly (FIGs. 12C, 12E, and 12G). Fourier ring correlation of the images show an improvement in spatial resolution of 741 nm in the GFP channel (FIG. 12H) and 515 nm in the SHG channel (FIG. 121). The most significant difference between the sCMOS and PMT approaches is the pattern that was output by the GA (sCMOS - FIG. 10A; PMT - FIG. 12A) where the optimized pattern of the PMT method outputs what appears to be a more truly random appearing pattern.

[0125] By comparing the relative enhancement of scanned image spatial resolution at various depths, it is seen that shallow imaging using the corrected DMD pattern does not result in significant enhancement. Once the imaging depth extends beyond the mean free distance of bone (circa 45 pm), the effect of the correction becomes evident, with improvements of approximately 2* (1.78) in SHG and more than 20* (23.2) in the GFP channels (FIG. 13). It is noted that these results will combine with other corrections as described herein to further enhance image resolution.Conclusion

[0126] Herein has been implemented a binary wavefront optimization technique that shows while in an epi-detection configuration, the intrinsic SHG can be used as a guide star. Optimization by means of a genetic algorithm result in masks that when applied to the DMD improve intensity and spatial resolution. A 2-fold improvement in GFP intensity allows accurateATTORNEY DOCKET NO. 222105-2420 identification of mitochondria and potentially study mitochondrial metabolism and its impact on osteogenesis in disease models. The 905 nm improvement in the spatial resolution of the GFP channel enables more accurate identification of the punctate mitochondria. Improved intensity and contrast within the collagen images allow for a more accurate location of the LCN boundaries and the 688 nm increase in spatial resolution will allow for the investigation and measurement of the fine collagen features within bone, improving the quality of LCN metrics within living samples. Future implementations will enable more accurate and compelling measurements in the drive toward clinical implementation of imaging histomorphology and histopathology.Example 5: Combined AO and Scatter Correction

[0127] The benefits of adaptive optics are applicable in several types of light microscopy and have been widely reported in modalities such as confocal microscopy, 2-photon microscopy, SHG imaging, and SIM. These reported methods target low order aberrations, limited by the electro-optical component employed to compensate for the distorted wavefront. Higher order aberrations require electro-optical devices capable of high spatial frequency manipulation such as the DMD that has been employed in typical scatter correction. Combining the low and high order aberration correction techniques will maximize the recovered signal at depth within tissue.

[0128] By executing a low order aberration correction of the sample at depth, a more effective scatter correction is enabled. To this end, the DM and DMD were combined into a custom setup, projecting the system correction onto the DMD and finding an SHG spot as deep as possible within the bone of an excised murine skull. Provided enough photons are collected, a sample correction can be executed using a maximum intensity metric, sweeping the lower 15 Zernike modes from -2V to +2V. Projecting the resulting compensatory phase map onto the DM will sharpen the PSF measured at this depth. A scatter correction at a slightly deeper position is enabled by this AO sample correction. Low order aberrations such as astigmatism or defocus, originating from the optical components can be corrected by measuring the wavefront propagating into the back aperture of the objective lens and applying a compensatory phase onto a low order electro-optical component such as a deformable mirror (DM). This system correction ensures that subsequent aberrations impacting the image quality are caused by the sample. In less thick tissue samples in the order of tens of microns, contributions of low order aberrations are minimal, but as tissue is extended, low order aberrations and the scattering of emitted photons combine to degrade the resulting PSF significantly. Beyond a single scattering length (~45 pm in bone) the emitted SHG PSF intensity declines sharply so that imaging through whole skull without correction at wavelengths in the visible range is not feasible. In order to correct both low order aberrationsATTORNEY DOCKET NO. 222105-2420 and losses associated with scattering tissue, the previously established low order aberration correction is combined with the binary modulation scatter correction using a DMD pattern optimized by a genetic algorithm. It is expected that the combination of corrective approaches will correct the large amplitude low order and low amplitude high order aberrations that restrict the depth of imaging in highly scattering tissues. The base scatter correction applied utilized ~50% of the segments, and the aberration correction was limited to 15 Zernike modes, without considering the first three modes (piston, tip, and tilt). To establish the most effective aberration correction within bone tissue, an excised murine skull sample was examined, conducting low order aberration correction at a depth of ~65pm with an uncorrected pattern projected onto the DMD. Here it was found that the aberration correction failed (FIG. 16A), characterized by the high phase values (limited to -TT to TT) measured by the correction algorithm which indicates that no significant peak was found during the sweep. When applying an corrected pattern to the DMD, it was found that the aberration correction yielded a phase map recovering the PSF (FIG. 16B). A cross section through the resulting images shows a 1.7-fold mean intensity in- crease where no peak is visible in the image resulting for the uncorrected pattern, but a sharp peak exists in the image resulting from the corrected DMD pattern (FIG. 16C). The two peaks visible in the cross section (FIGs. 16B-16C) 1.4 times above the mean intensity, and 1.8 times above the minimum intensity. As the focal plane is withdrawn through the sample in steps of 5 pm, the corrected DMD pattern enables aberration correction while the uncorrected pattern does not yield useful correction. The deepest point in the sample at which an uncorrected DMD pattern aids in a sample aberration correction is 50 pm, where a PSF is recovered, but the necessary phases projected onto the DM have outlier high values (FIG. 16D). An aberration correction using the corrected DMD pattern yields a much sharper and brighter PSF (FIG. 16E). A cross section through the resulting images shows a 1.7-fold peak intensity increase and generation of a peak (FWHM of 1 .56 pm) of the PSF, where no distinct peak was present in the uncorrected image (FIG. 16F). These results confirm that scatter correction enables an improved sample aberration correction.

[0129] The degree to which this combination of aberration and scatter correction affects scanned imaging was then investigated. Here, it was found that imaging at a depth of —110 pm in an excised murine skull sample, without either aberration or scatter correction, that the resulting image lacks contrast and is dim (FIG. 17A). When only aberration correction is applied, the intensity does not increase (FIG. 17B). After a corrected DMD pattern was projected onto the DMD, but no aberration correction applied to the DM, a significant increase in intensity in evident (FIG. 17C) and some subcellular features are visible. Finally, when the aberration correction was added to the scatter correction, the intensity was maximized and subcellular features became clear (FIG. 17D). Cross sections of the GFP (FIG. 17D) acrossATTORNEY DOCKET NO. 222105-2420 each of the configurations showed that without scatter correction the intensity of both GFP and SHG was low, but increased more than 2-fold when the scatter correction was applied (FIG. 17E). Aberration correction improved the intensity in both cases with or without scatter correction by a factor of approximately 3x and contributed to the visibility of more features and separate peaks in the GFP intensity profiles.

[0130] It should be emphasized that the above-described embodiments of the present disclosure are merely possible examples of implementations set forth for a clear understanding of the principles of the disclosure. Many variations and modifications may be made to the above-described embodiment(s) without departing substantially from the spirit and principles of the disclosure. All such modifications and variations are intended to be included herein within the scope of this disclosure and protected by the following claims.ATTORNEY DOCKET NO. 222105-2420REFERENCES Akbulut, D. et al. Focusing light through random photonic media by binary amplitude modulation. Optics Express, 19(5):4017-4029, 2011. Albert, O., et al. (2000). Smart microscope: an adaptive optics learning system for aberration correction in multiphoton confocal microscopy. Opt. Lett. 25, 52-54. Andermann, M. L., et al. (2010). Chronic cellular imaging of mouse visual cortex during operant behavior and passive viewing. Fronti. Cell. Neurosci. 4, 3. Aviles-Espinosa, R., et al. (2011). Measurement and correction of in vivo sample aberrations employing a nonlinear guide-star in two-photon excited fluorescence microscopy. Biomed. Opt. Expr. 2, 3135-3149. Booth, M. J. (2006). Wave front sensor-less adaptive optics: a model-based approach using sphere packings. Opt. Express 14, 1339-1352. Booth, M.J. Adaptive optical microscopy: the ongoing quest for a perfect image. Light: Science & Applications, 3(4):e165-e165, 2014. ISSN 2047-7538. Booth, M.J., et al., Adaptive aberration correction in a confocal microscope. Proceedings of the National Academy of Sciences, 2002. 99(9): p. 5788-5792. Borovoi, A.G. Analytical description of snake photons. In Advances in Optical Imaging and Photon Migration, Technical Digest Series, page PMST69. Optica Publishing Group. Briers, J.D. et al. Quasi real-time digital version of single-exposure speckle photography for full-field monitoring of velocity or flow fields. Optics Communications, 116(1):36-42, 1995. ISSN 0030-4018. Callis, P. R. (1997). Two-photon-induced fluorescence. Ann. Rev. Phys. Chem. 48, 271- 297. Calvo-Rodriguez,M., et al. (2019). In vivo detection of tau fibrils and amyloid b aggregates with luminescent conjugated oligothiophenes and multiphoton microscopy. Acta Neuropathol. Commun. 7, 171. Campagnola, P. Second harmonic generation imaging microscopy: Applications to diseases diagnostics. Analytical Chemistry, 83(9): 3224-3231 , 2011. ISSN 0003-2700. Chen, C., et al. (2021). High-resolution two-photon transcranial imaging of brain using direct wavefront sensing. Photon. Res. 9, 1144-1156.ATTORNEY DOCKET NO. 222105-2420 Chen, Z., et al. (2019). Intravital 2-photon imaging reveals distinct morphology and infiltrative properties of glioblastoma-associated macrophages. Proc. Nat. Acad. Sci. 116, 14254-14259. Christodoulou, C., et al. (2020). Live-animal imaging of native haematopoietic stem and progenitor cells. Nature 578, 278-283. Conkey, D.B. et al. Genetic algorithm optimization for focusing through turbid media in noisy environments. Optics Express, 20(5):4840-4849, 2012. Cui, M. Parallel wavefront optimization method for focusing light through random scattering media. Optics Letters, 36(6):870-872, 2011. Debarre, D., et al. (2009). Image-based adaptive optics for two-photon microscopy. Opt. Lett. 34, 2495-2497. Denk, W., et al. (1990). Two-photon laser scanning fluorescence microscopy. Science 248, 73-76. Diaspro, A., et al. (2006). Multiphoton excitation microscopy. Biomed. Eng. Online 5, 1-14. Durech, E. et al. Wavefront sensor-less adaptive optics using deep reinforcement learning. Biomedical Optics Express, 12(9): 5423-5438, 2021. Fang, L. et al. Binary wavefront optimization using particle swarm algorithm. Laser Physics, 28(7):076204, 2018. ISSN 1555-6611 1054-660X. Folmes, C. D., et al. (2012). Mitochondria in control of cell fate. Circul. Res. 110, 526-529. Fried, H. U., et al. (2001). An inexpensive mouse headholder suitable for optical recordings. Physiol. Behav. 74, 253-255. Goorden, S.A. et al. Superpixel-based spatial amplitude and phase modulation using a digital micromirror device. Optics Express, 22(15): 17999-18009, 2014. Gropman, A. L. (2004). The neurological presentations of childhood and adult mitochondrial disease: established syndromes and phenotypic variations. Mitochondrion. 4, 503-520. Hassanat, H. et al. Choosing mutation and crossover ratios for genetic algorithms — a review with a new dynamic approach. Information, 10(12):390, 2019. ISSN 2078-2489. Hell, S. (1992). Fundamental improvement of resolution with a 4Pi-confocal fluorescence microscope using two-photon excitation. Opt. Commun. 93, 277-282. Helmchen, F. et al. Deep tissue two-photon microscopy. Nat Methods, 2(12):932-40, 2005. ISSN 1548-7091 (Print) 1548-7091.ATTORNEY DOCKET NO. 222105-2420 Hu, Y., et al. (2021). Longitudinal characterization of blood-brain barrier permeability after experimental traumatic brain injury by in vivo 2-photon microscopy. J Neurotrauma. 38, 399-410. Itkin, T., et al. (2016). Distinct bonemarrow blood vessels differentially regulate haematopoiesis. Nature 532, 323-328. Jeong, D. C., et al. (2013). All-optical osteotomy to create windows for transcranial imaging in mice. Opt. Expr. 21 , 23160-23168. Jesacher, A., et al., Adaptive harmonic generation microscopy of mammalian embryos. Optics Letters, 2009. 34(20): p. 3154. Kobat, D. et al. Deep tissue multiphoton microscopy using longer wavelength excitation. Optics Express, 17(16): 13354-13364, 2009. Kong, L. (2015). In vivo deep tissue imaging via iterative multiphoton adaptive compensation technique. IEEE J. Sei. Top. Quantum Electron. 22, 40-49. Kong, L. (2015). In vivo neuroimaging through the highly scattering tissue via iterative multi-photon adaptive compensation technique. Opt. Expr. 23, 6145-6150. Kong, L., et al. (2016). Multicolor multiphoton in vivo imaging flow cytometry. Opt. Expr. 24, 6126-6135. Koza, J. et al. Genetic Programming, pages 127-164. Springer, Boston, MA, 2005. ISBN 9780387283562. Legare, F. et al. The role of backscattering in shg tissue imaging. Biophys J, 93(4):1312- 20, 2007. ISSN 0006-3495 (Print) 0006-3495. Leuschner, F., et al. (2015). Silencing of CCR2 in myocarditis. Eur. Heart J. 36, 1478- 1488. Li, Q., et al., Woofer-tweeter adaptive optical structured illumination microscopy. Photonics Research, 2017. 5(4): p. 329-334. Li, Y., et al. (2014). Application of thinned-skull cranial window to mouse cerebral blood flow imaging using optical microangiography. PLoS ONE 9, e113658. Lin, R., et al., Subcellular three-dimensional imaging deep through multicellular thick samples by structured illumination microscopy and adaptive optics. Nature Communications, 2021. 12(1): p. 3148. Lo Celso, C., et al. (2009). Live-animal tracking of individual haematopoietic stem / progenitor cells in their niche. Nature 457, 92-96.ATTORNEY DOCKET NO. 222105-2420 Marsh, P. N., et al. (2003). Practical implementation of adaptive optics in multiphoton microscopy. Opt. Expr. 11 , 1123-1130. May, M.A. et al. Fast holographic scattering compensation for deep tissue biological imaging. Nature Communications, 12(1): 4340, 2021. ISSN 2041-1723. Miao, F., et al. (2014). Novel fluorescent probes for highly selective two-photon imaging of mitochondria in living cells. FEBS Lett. 396, 7-13. Mortensen, L.J. et al. Femtosecond laser bone ablation with a high repetition rate fiber laser source. Biomed Opt Express, 6(1):32-42, 2015. ISSN 2156-7085 (Print) 2156-7085. Murphy, M. P. (2009). How mitochondria produce reactive oxygen species. Biochem. J. 417, 1-13. Neil, M. A., et al. (2000). Closed-loop aberration correction by use of a modal Zernike wave-front sensor. Opt. Lett. 25, 1083-1085. Nieuwenhuizen, R.P.J. et al. Measuring image resolution in optical nanoscopy. Nature Methods, 10(6): 557-562, 2013. ISSN 1548-7105. Noll, R. J. (1976). Zernike polynomials and atmospheric turbulence. JOsA 66, 207-211. Park, J. et al. Focusing through turbid media by polarization modulation. Opt Lett, 40 (8):1667-70, 2015. ISSN 0146-9592. Park, J. H., et al. (2015). High-resolution in vivo imaging of mouse brain through the intact skull. Proc. Nat. Acad. Sci. 112, 9236-9241. Pendleton, E.G. et al. Second harmonic generation characterization of collagen in whole bone. Biomedical Optics Express, 11(8): 4379-4396, 2020. Pham, A. H., et al. (2012). Mouse lines with photo-activatable mitochondria to study mitochondrial dynamics. Genesis 50, 833-843. Pologruto, T. A., et al. (2003). Scanimage: flexible software for operating laser scanning microscopes. Biomed. Eng. 2, 1-9. Ricard, C. (2014). Six-color intravital two-photon imaging of brain tumors and their dynamic microenvironment. Front. Cell. Neurosci. 8, 57. Rueckel, M., et al. (2006). Adaptive wavefront correction in two-photon microscopy using coherence-gated wavefront sensing. Proc. Natl. Acad. Sci. U. S. A 103, 17137-17142. Schneider, C.A. et al. Nih image to imagej: 25 years of image analysis. Nature Methods, 9(7):671-675, 2012. ISSN 1548-7105.ATTORNEY DOCKET NO. 222105-2420 Shi, L., et al. (2016). Transmission in near-infrared optical windows for deep brain imaging. J Biophotonics. 9, 38-43. Shih, A. Y., et al. (2012). Two-Photon Microscopy as a Tool to Study Blood Flow and Neurovascular Coupling in the Rodent Brain. J. Cerebral Blood Flow Metab. 32, 1277- 1309. Siegel, G. J., et al. (1981). Basic Neurochemistry. Boston, MA, USA: Little Brown. Siesjo, B. K. (1978). Brain Energy Metabolism. New York, NY, USA: John Wiley and Sons. Sinefeld, D., et al. (2015). Adaptive optics in multiphoton microscopy: comparison of two, three and four photon fluorescence. Opt. Express 23, 31472-31483. Southern, W. M., et al. (2019). PGC-1a overexpression partially rescues impaired oxidative and contractile pathophysiology following volumetric muscle loss injury. Sci. Rep. 9, 4079. Spencer, J. A., et al. (2014). Direct measurement of local oxygen concentration in the bone marrow of live animals. Nature. 508, 269-273. Streich, L. et al. High-resolution structural and functional deep brain imaging using adaptive optics three-photon microscopy. Nature Methods, 18(10): 1253-1258, 2021. ISSN 1548-7105. Tao, X. et al. High-speed scanning interferometric focusing by fast measurement of binary transmission matrix for channel demixing. Optics Express, 23(11): 14168— 14187, 2015. Tao, X., et al. (2013). Adaptive optical two-photon microscopy using autofluorescent guide stars. Opt. Lett. 38, 5075-5078. Tao, X., et al. (2013). Shack-Hartmann wavefront sensing using interferometric focusing of light onto guide-stars. Opt. Expr. 21 , 31282-31292. Tehrani, K. F., et al. (2017). Characterization of wavefront errors in mouse cranial bone using second-harmonic generation. J. Biomed. Opt. 22, 36012. Tehrani, K. F., et al. (2017). Two-photon deep-tissue spatially resolved mitochondrial imaging using membrane potential fluorescence fluctuations. Biomed. Opt. Express. 9, 254-259. Tehrani, K. F., et al. (2019). “Binary wavefront optimization for multi-photon intravital microscopy through intact cranial bone,” in Adaptive Optics and Wavefront Control for Biological Systems V, Vol. 10886 (SPIE), 1088605.ATTORNEY DOCKET NO. 222105-2420 Tehrani, K. F., et al. (2019). Five-dimensional two-photon volumetric microscopy of in-vivo dynamic activities using liquid lens remote focusing. Biomed. Opt. Expr. 10, 3591-3604. Tehrani, K. F., et al. (2021). In situ measurement of the isoplanatic patch for imaging through intact bone. J Biophotonics. 14, e202000160. Tehrani, K.F. et al. Adaptive optics stochastic optical reconstruction microscopy (ao-storm) by particle swarm optimization. Biomedical Optics Express, 8(11):5087-5097, 2017. Tehrani, K.F. et al. Nonlinear imaging histopathology: A pipeline to correlate gold-standard hematoxylin and eosin staining with modern nonlinear microscopy. IEEE Journal of Selected Topics in Quantum Electronics, 29(4: Biophotonics): 1-8, 2023. ISSN 1558-4542. Turcotte, R., et al. (2017). Image-guided transplantation of single cells in the bone marrow of live animals. Sci. Rep. 7, 1-9. Vellekoop I.M. et al. Focusing coherent light through opaque strongly scattering media. Optics Letters, 32(16):2309-2311 , 2007. Vellekoop, I.M. et al. Phase control algorithms for focusing light through turbid media. Optics Communications, 281 (11):3071-3080, 2008. ISSN 0030-4018. Wang, K., et al. (2015). Direct wavefront sensing for high-resolution in vivo imaging in scattering tissue. Nat. Commun. 6, 7276. Wang, M., et al. (2018). Comparing the effective attenuation lengths for long wavelength in vivo imaging of the mouse brain. Biomed. Opt. Expr. 9, 3534-3543. Wilk, K., et al. (2017). Postnatal calvarial skeletal stem cells expressing PRX1 reside exclusively in the calvarial sutures and are required for bone regeneration. Stem Cell Rep. 8, 933-946. Wright, A. J., et al. (2005). Exploration of the optimisation algorithms used in the implementation of adaptive optics in confocal and multiphoton microscopy. Microsc. Res. Tech. 67, 36-44. Wu, D. et al. A thorough study on genetic algorithms in feedback- based wavefront shaping. Journal of Innovative Optical Health Sciences, 12(04): 1942004, 2019. Xu, C. (1996). Measurement of two-photon excitation cross sections of molecular fluorophores with data from 690 to 1050 nm. J. Opt. Soc. Am. B 13, 481-491. Xu, Z., et al. (2016). Transport of Calcium Ions into Mitochondria. Curr. Genom. 17, 215- 219.ATTORNEY DOCKET NO. 222105-2420 Yang, G., et al. (2013). Transcranial two-photon imaging of synaptic structures in the cortex of awake head-restrained mice. Methods Mol. Biol. 1010, 35-43. Yang, J. et al. Anti-scattering light focusing by fast wavefront shaping based on multi-pixel encoded digital-micromirror device. Light: Science & Applications, 10(1):149, 2021. ISSN 2047-7538. Yang, W., et al. (2018). Simultaneous two-photon imaging and two-photon optogenetics of cortical circuits in three dimensions, elife 7, e32671. Yeh, S.A. et al. In vivo 3d histomorphometry quantifies bone apposition and skeletal progenitor cell differentiation. Sci Rep, 8(1):5580, 2018. ISSN 2045-2322. Yeminy, T. et al. Guidestar-free image-guided wavefront shaping. Science Advances, 7(21):eabf5364, 2021. Zhang, B. et al. Focusing light through strongly scattering media using genetic algorithm with sbr discriminant. Journal of Optics, 20(2):025601 , 2018. ISSN 2040-89862040-8978. Zhang, X. et al. Binary wavefront optimization using a genetic algorithm. Journal of Optics, 16 (12):125704, 2014. ISSN 2040-8978 2040-8986. Zhao, C. (2007). Orthonormal vector polynomials in a unit circle, Part I: basis set derived from gradients of Zernike polynomials. Opt. Expr. 15, 18014-18024. Zheng, T. et al. Imaging mitochondria through bone in live mice using two-photon fluorescence microscopy with adaptive optics. Frontiers in Neuroimaging, 2:4, 2023. ISSN 2813-1193. Zheng, W. et al. Adaptive optics improves multiphoton super-resolution imaging. Nature Methods, 14(9): 869-872, 2017. ISSN 1548-7105.

Claims

ATTORNEY DOCKET NO. 222105-2420CLAIMSWhat is claimed is:

1. A method for correcting distortions or aberrations in a multi-photon microscopy image of a sample, the method comprising:(a) illuminating the sample;(b) producing an image of the sample with an imaging device having one or more detection elements that receive a signal from the sample; and(c) processing the image of the sample using a scattering correction and a phase correction.

2. The method of claim 1 , wherein the multi-photon microscopy image comprises a two-photon microscopy image or a three-photon microscopy image.

3. The method of claim 1 , wherein the phase correction comprises a low order deformable mirror (DM) adaptive optics (AO) correction.

4. The method of claim 1 , wherein the scattering correction comprises a binary scattering correction, a phase scattering correction, or any combination thereof5. The method of claim 1 , wherein the scattering correction comprises a high-order digital micromirror device (DMD) correction.

6. The method of claim 1 , further comprising a spatial light modulator (SLM) correction.

7. The method of claim 1 , wherein illuminating the sample comprises using a beam from a light source to induce fluorescence in the sample8. The method of claim 7, wherein the light source comprises a Thsapphire laser.

9. The method of claim 7, wherein the signal from the sample comprises a fluorescence signal.

10. The method of claim 7, wherein the light source induces fluorescence using a wavelength of from about 680 nm to about 1080 nm.11 . The method of claim 7, wherein the light source produces a plurality of pulses of energy of from about 75 fs to about 400 fs.

12. The method of claim 7, wherein the light source produces a plurality of 137 fs pulses of energy.

13. The method of claim 7, wherein power of the light source is modulated using a halfwaveplate, a polarizing beam splitter, or both.ATTORNEY DOCKET NO. 222105-242014. The method of claim 13, wherein after splitting, beam is recollimated and passed through a Pockels cell for rapid intensity modulation during imaging.

15. The method of claim 14, wherein after the beam exits the Pockels cell, the beam passes through a pinhole, wherein passing through the pinhole improves the Gaussian profile of the beam.

16. The method of claim 14, wherein the Pockels cell is further used to block the beam during flyback and to correct beam intensity for a resonant scanner.

17. The method of claim 3, wherein AO correction comprises an instrumental component and a software component.

18. The method of claim 3, wherein the AO correction corrects for system aberrations.

19. The method of claim 3, wherein the AO correction is conducted prior to correcting for sample aberrations.

20. The method of claim 3, wherein the instrumental component of the AO correction comprises a deformable mirror (DM).

21. The method of claim 20, wherein the DM is placed conjugate to a back pupil plane and wherein the DM is used for low-order AO wavefront correction.

22. The method of claim 20, wherein the DM has a continuous face sheet.

23. The method of claim 20, wherein the DM comprises from about 50 to about 200 actuators.

24. The method of claim 20, wherein the DM comprises 97 actuators.

25. The method of claim 20, wherein the DM is inserted into a path of the beam with a first optical relay system and a second optical relay system.

26. The method of claim 5, further comprising using a digital micromirror device (DMD) to expand the beam to utilize from about 10% to about 50% of a DMD active array surface.

27. The method of claim 5, further comprising using a digital micromirror device (DMD) to expand the beam to utilize about 30% of a DMD active array surface.

28. The method of claim 26, wherein the DMD device comprises a plurality of micromirrors, wherein each micromirror has a diameter of from about 5 pm to about 20 pm.

29. The method of claim 26, wherein each micromirror has a diameter of about 13.68 pm.

30. The method of claim 28, wherein each of the plurality of micromirrors tilts at from about ±5° to about ±20 °.

31. The method of claim 28, wherein each of the plurality of micromirrors tilts at +12° or -12°.ATTORNEY DOCKET NO. 222105-242032. The method of claim 26, wherein a diffraction grating (DG) compensates for dispersion generated by a surface of the DMD.

33. The method of claim 5, further comprising using a Genetic Algorithm to process data obtained from the DMD.

34. The method of claim 1 , wherein beam scanning is performed by a resonant scanning box housing a fast resonant galvo scanner for horizontal sweeping and a slow galvo scanner for vertical sweeping, thereby producing a scanned beam.

35. The method of claim 34, wherein the fast resonant galvo scanner and the slow galvo scanner are placed close together to reduce astigmatism.

36. The method of claim 34, wherein the scanned beam is relayed using a first achromatic doublet serving as a scanning lens and a second achromatic doublet serving as a tube lens.

37. The method of claim 34, wherein the scanned beam is further relayed to a water immersion objective lens.

38. The method of claim 1 , wherein back-propagated emission light from the sample is separated from excitation light using a first dichroic mirror and sent to a signal capture means.

39. The method of claim 38, wherein the signal capture means comprises a camera or sensor.

40. The method of claim 39, wherein the camera comprises a charge-coupled device (CCD) camera, an electron multiplying CCD camera (EMCCD), a complementary metal-oxide semiconductor (CMOS) camera or sensor, a scientific CMOS (sCMOS) camera or sensor, or any combination thereof.

41. The method of claim 39, wherein the camera is an sCMOS camera.

42. The method of claim 41 , wherein the sCMOS camera is used to look at a point spread function (PSF) shape or separated with a second dichroic mirror and sent to one or more photomultiplier tubes.

43. The method of claim 41 , wherein light emitted by the sample is captured by the sCOMS camera through a filter for a second harmonic generation (SHG) guidestar.

44. The method of claim 38, wherein a third dichroic mirror and a fourth dichroic mirror and one or more filters are used to separate spectral channels to capture signals from the illuminated sample.

45. The method of claim 44, wherein the one or more filters comprise a 571 / 72 nm filter, a 509 / 22 nm filter, a 390 / 18 nm filter, or any combination thereof.ATTORNEY DOCKET NO. 222105-242046. The method of claim 1 , wherein the beam or components thereof pass through a Shack- Hartmann wavefront (SHWF) sensor comprising a lenslet array to measure total system aberration prior to the beam encountering an objective lens.

47. The method of claim 46, wherein the lenslet array creates spots on the image, wherein displacement of the spots is compared to an internal reference guidestar that allows calculation of wavefront distortions.

48. The method of claim 20, wherein one correction is performed at each focal plane of the DM.

49. The method of claim 47, further comprising decomposing wavefront distortions into Zernike modes.

50. The method of claim 49, further comprising using the DM to correct for Zernike modes 5 to 37.

51. The method of claim 49, wherein Zernike decomposition of the wavefront identifies contributions to distortion from tip, tilt, and defocus, and wherein Shack-Hartmann wavefront correction is applied to correct the distortions caused by tip, tilt, and defocus.

52. The method of claim 1 , wherein the distortions comprise aberrations generated by non- homogeneous wave propagation through complex tissue.

53. The method of claim 52, wherein the distortions are non-uniform in the complex tissue.

54. The method of claim 1 , wherein the method can be completed in one minute or less.

55. The method of claim 1 , wherein the method is performed in a histological tissue section.

56. The method of claim 1 , wherein the method is performed in a living subject.

57. The method of claim 56, wherein the living subject is a mammal.

58. The method of claim 57, wherein the mammal is a human, non-human primate, cat, dog, horse, cattle, swine, sheep, goat, mouse, rat, rabbit, guinea pig, or hamster.

59. The method of claim 56, wherein performing the method does not induce inflammation or another immune response in the subject.

60. The method of claim 52, wherein the complex tissue comprises bone, bone marrow, or a combination thereof.

61. The method of claim 60, wherein the method enables visualization of osteocytes, mitochondria, or both osteocytes and mitochondria.

62. The method of claim 61 , wherein the method is capable of internally visualizing the complex tissue at a depth of up to 250 pm.ATTORNEY DOCKET NO. 222105-242063. A method for detecting a disease or monitoring the progress of a disease in a subject, the method comprising performing multi-photon imaging on an area of interest in the subject, wherein the area of interest comprises complex tissue, and performing the method of any one of claims 1-62 to correct distortions.

64. The method of claim 63, wherein the method can be completed in one minute or less.

65. The method of claim 63, wherein the method is performed in a histological tissue section.

66. The method of claim 63, wherein the method is performed in a living subject.

67. The method of claim 66, wherein the living subject is a mammal.

68. The method of claim 67, wherein the mammal is a human, non-human primate, cat, dog, horse, cattle, swine, sheep, goat, mouse, rat, rabbit, guinea pig, or hamster.

69. The method of claim 66, wherein performing the method does not induce inflammation or another immune response in the subject.

70. The method of claim 63, wherein the complex tissue comprises bone, bone marrow, or a combination thereof.

71. The method of claim 70, wherein the method enables visualization of osteocytes, mitochondria, or both osteocytes and mitochondria.

72. The method of claim 63, wherein the method is capable of internally visualizing the complex tissue at a depth of up to 250 pm.

73. The method of claim 63, wherein the disease comprises osteoporosis, osteomalacia, Paget’s disease, hypophosphatasia, aging-related bone changes, X-linked hypophosphatemia, bone repair after injury, or osteogenesis imperfecta.

74. The method of claim 63, wherein the area of interest comprises a fluorophore.

75. The method of claim 74, further comprising using a bandpass filter to capture images of the fluorophore in the area of interest.

76. The method of claim 74, wherein the fluorophore is green fluorescent protein (GFP).

77. The method of claim 75, wherein the bandpass filter comprises a 457 / 50 nm bandpass filter.

Images