The use of adaptive lenses instead of deformable mirrors can simplify the implementation of an adaptive optics system. The recently introduced Multi-actuator Adaptive Lens (MAL) can be used in closed loop with a wavefront sensor to correct for time-variant wavefront aberrations. The MAL can guarantee a level of correction and a response time similar to the ones obtained with deformable mirrors. The adaptive lens is based on the use of piezoelectric actuators and, without any obstruction or electrodes in the clear aperture, can guarantee a fast response time, less than ~10ms. Our tests show that the MAL can be used both in combination with a wavefront sensor in a “classical” adaptive optics closed loop, or in a wavefront sensorless configuration. The latter has allowed us to design more compact and simple imaging systems for different microscopy platforms. We will show that the Multi-actuator Adaptive Lens has been successfully used for in-vivo OCT ophthalmic imaging in both mice and humans, as well as confocal and two photon microscopy. We tested and compared different optimization strategies such as coordinate search and the DONE algorithm. The results suggest that the MAL optimization can correct for eye aberrations with a pupil of 5mm or sample induced aberrations in microscopy.

Deformable mirrors are the standard adaptive optical elements for aberration correction in confocal microscopy. Their usage leads to increased contrast and resolution. However, these improvements are achieved at the cost of bulky optical setups. Since spherical aberrations are the dominating aberrations in confocal microscopy, it is not required to employ all degrees of freedom commonly offered by deformable mirrors. In this contribution, we present an alternative approach for aberration correction in confocal microscopy based on a novel adaptive lens with two degrees of freedom. These lenses enable both axial scanning and aberration correction, keeping the setup simple and compact. Using digital holography, we characterize the tuning range of the focal length and the spherical aberration correction ability of the adaptive lens. The operation at fixed trajectories in terms of focal length and spherical aberrations is demonstrated and investigated in terms of reproducibility. First results indicate that such adaptive lenses are a promising approach towards high-resolution, high-speed three-dimensional microscopy.

Liquid lenses are used to correct for low order wavefront aberrations. Electrowetting liquid lenses can nowadays control defocus and astigmatism effectively, so they start being used for ophthalmology applications. To increase the performance and applicability, we introduce a new driving mechanism to create, detect and correct higher order aberrations using standing waves on the liquid interface. The speed of a liquid lens is in general limited, because the liquid surface cannot follow fast voltage changes, while providing a spherical surface. Surface waves are created instead and with them undesired aberrations. We try to control those surface waves to turn them into an effective wavefront shaping tool. We introduce a model, which treats the liquid lens as a circular vibrating membrane with adjusted boundary conditions. Similar to tunable acoustic gradient (TAG) lenses, the nature of the surface modes are predicted to be Bessel functions. Since Bessel functions are a full set of orthogonal basis functions any surface can be created as a linear combination of different Bessel functions. The model was investigated experimentally in two setups. First the point spread functions were studied and compared to a simulation of the intensity distribution created by Fresnel propagated Bessel surfaces. Second the wavefronts were measured directly using a spatial light modulator. The surface resonance frequencies confirm the predictions made by the model as well as the wavefront measurements. By superposition of known surface modes, it is possible to create new surface shapes, which can be used to simulate and measure the human eye.

Light-sheet fluorescence microscopy is quickly becoming one of the cornerstone imaging techniques in biology as it provides rapid, three-dimensional sectioning of specimens at minimal levels of phototoxicity. It is very appealing to bring this unique combination of imaging properties into an endoscopic setting and be able to perform optical sectioning deep in tissues. Current endoscopic approaches for delivery of light-sheet illumination are based on single-mode optical fibre terminated by cylindrical gradient index lens. Such configuration generates a light-sheet plane that is axially fixed and a mechanical movement of either the sample or the endoscope is required to acquire three-dimensional information about the sample. Furthermore, the axial resolution of this technique is limited to 5um. The delivery of the light-sheet through the multimode fibre provides better axial resolution limited only by its numerical aperture, the light-sheet is scanned holographically without any mechanical movement, and multiple advanced light-sheet imaging modalities, such as Bessel and structured illumination Bessel beam, are intrinsically supported by the system due to the cylindrical symmetry of the fibre. We discuss the holographic techniques for generation of multiple light-sheet types and demonstrate the imaging on a sample of fluorescent beads fixed in agarose gel, as well as on a biological sample of Spirobranchus Lamarcki.

Light-sheet microscopy (LSM) is an emergent fluorescence microscopy technique showing great promise for biomedical research. LSM enables rapid, high-contrast imaging of large specimens with high spatiotemporal resolution and minimal photo-damage. When imaging large specimens, the intensity of the light-sheet reduces across the field-of-view (FOV) due to absorption. This results in an image with spatially-variant intensity, affecting quantitative measurements, and ultimately limits the penetration depth of the illumination. Some existing approaches to alleviate this issue involve illuminating the sample from multiple directions or rotating the sample. These approaches are not always practical and restrict specimen choice. Separately, propagation-invariant light modes have been used to develop high-resolution LSM techniques as they can overcome the natural divergence of a Gaussian beam, producing a thin and uniform light-sheet over long distances. Most notably, Bessel and Airy beam-based LSM techniques have been implemented. For propagation-invariant beams, there exists a mapping between the transverse coordinate in the pupil plane of a lens, and the axial propagation in the focal plane. Spatially-variant amplitude modulation therefore offers control of the intensity of the beam with propagation. In this paper, we report that such amplitude modulation in the pupil plane of an Airy LSM can yield a system which counteracts absorption of the light-sheet and gives uniform intensity across the FOV with a single acquisition and without restricting specimen choice. This technique is an alternative to, and may be complimented by, wavefront correction. We demonstrate this technique through numerical simulations and experimental validation in absorbing tissue phantoms.

Some biological experiments demand the observation of dynamics processes in 3D with high spatiotemporal resolution. The use of wavefront coding to extend the depth-of-field (DOF) of the collection arm of a light-sheet microscope is an interesting alternative for fast 3D imaging. Under this scheme, the 3D features of the sample are captured at high volumetric rates while the light sheet is swept rapidly within the extended DOF. The DOF is extended by coding the pupil function of the imaging lens by using a custom-designed phase mask. A posterior restoration step is required to decode the information of the captured images based on the applied phase mask [1]. This hybrid optical-digital approach is known as wavefront coding (WFC). Previously, we have demonstrated this method for performing fast 3D imaging of biological samples at medium resolution [2]. In this work, we present the extension of this approach for high-resolution microscopes. Under these conditions, the effective DOF of a standard high NA objective is of a few micrometers. Here we demonstrate that by the use of WFC, we can extend the DOF more than one order of magnitude keeping the high-resolution imaging. This is demonstrated for two designed phase masks using Zebrafish and C. elegans samples. [1] Olarte, O.E., Andilla, J., Artigas, D., and Loza-Alvarez, P., “Decoupled Illumination-Detection Microscopy. Selected Optics in Year 2105,” in Optics and Photonics news 26, p. 41 (2015). [2] Olarte, O.E., Andilla, J., Artigas, D., and Loza-Alvarez, P., “Decoupled illumination detection in light sheet microscopy for fast volumetric imaging,” Optica 2(8), 702 (2015).

We have recently reported on a method to design at will the spatial profile of transmitted coherent light after propagation through a strongly scattering sample, exploiting wavefront shaping in combination with a transmission matrix approach. In this paper, we explore experimentally and theoretically the ability of this approach to generate foci whose full width at half maximum are smaller than the diffraction-limited speckle grain size, using (Bessels) beam variations implemented with virtual annular filters.

Wide-field fluorescence microscopy is generally limited to either small volumes or low temporal resolution. We present a microscope add-on that provides fast, light-efficient extended depth-of-field (EDOF) using a deformable mirror of update rate 20kHz. Out-of-focus contributions in the raw EDOF images are suppressed with a deconvolution algorithm derived directly from the microscope 3D optical transfer function. Demonstrations of the benefits of EDOF microscopy are shown with GCaMP-labeled mouse brain tissue.

Optical phase conjugation based wavefront shaping techniques are being actively developed to focus light through or inside scattering media such as biological tissue, and they promise to revolutionize optical imaging, manipulation, and therapy. The speed of digital optical phase conjugation (DOPC) has been limited by the low speeds of cameras and spatial light modulators (SLMs), preventing DOPC from being applied to thick living tissue. Recently, a fast DOPC system was developed based on a single-shot wavefront measurement method, a field programmable gate array (FPGA) for data processing, and a digital micromirror device (DMD) for fast modulation. However, this system has the following limitations. First, the reported single-shot wavefront measurement method does not work when our goal is to focus light inside, instead of through, scattering media. Second, the DMD performed binary amplitude modulation, which resulted in a lower focusing contrast compared with that of phase modulations. Third, the optical fluence threshold causing DMDs to malfunction under pulsed laser illumination is lower than that of liquid crystal based SLMs, and the system alignment is significantly complicated by the oblique reflection angle of the DMD. Here, we developed a simple but high-speed DOPC system using a ferroelectric liquid crystal based SLM (512 × 512 pixels), and focused light through three diffusers within 4.7 ms. Using focused-ultrasound-guided DOPC along with a double exposure scheme, we focused light inside a scattering medium containing two diffusers within 7.7 ms, thus achieving the fastest digital time-reversed ultrasonically encoded (TRUE) optical focusing to date.

Optical phase conjugation (OPC) based wavefront shaping techniques focus light through or within scattering media, which is critically important for deep-tissue optical imaging, manipulation, and therapy. However, to date, the sample thicknesses used in wavefront shaping experiments have been limited to only a few millimeters or several transport mean free paths. Here, by using a long-coherence-length laser and an optimized digital OPC system that efficiently delivers light power, we focused 532 nm light through tissue-mimicking phantoms up to 9.6 cm thick, as well as through ex vivo chicken breast tissue up to 2.5 cm thick.

Optical scattering of biological tissue limits the working depth of conventional biomedical optics, which relies on the detection of ballistic photons. Recent developed optical phase conjugation (OPC) technique breaks through this depth limit by harnessing the scattered photons and shaping an optical wavefront that can “undo” the optical scattering. The OPC system measures the complex light field exiting the tissue and reconstructs a phase conjugated copy of the measured wavefront, which propagates in the reversed direction to the source of the light. To focus light inside a scattering medium, an embedded light source or “guidestar” is often required. Therefore, developing guidestar mechanisms plays an important role in advancing the OPC technique for deep tissue optical focusing and imaging. In addition to having strong optical modulation efficiency and compact size, a favorable guidestar for biomedical applications should also have good biocompatibility, fast response time, and be noninvasive or require only minimally invasive procedure. While a number of guidestar mechanisms have been developed and showed promising for various biomedical applications, they all have their own limitations. We have been developing new guidestars and tailoring them to meet the need for biomedical imaging and therapies. We are going to present our recent progress in novel guidestar development, compare them with established guidestar mechanisms, and discuss their potential in biomedical applications.

The first successful attempts (Abramson) at capturing light in flight relied on the holographic interference between the ``object'' beam scattered from a screen and a short reference pulse propagating at an angle, acting as an ultrafast shutter cite{egg}. This interference pattern was recorded on a photographic plate or film and allowed the visualisation of light as it propagated through complex environments with unprecedented temporal and spatial resolution. More recently, advances in ultrafast camera technology and in particular the use of picosecond resolution streak cameras allowed the direct digital recording of a light pulse propagating through a plastic bottle (Rasker at el.). This represented a remarkable step forward as it provided the first ever video recording (in the traditional sense with which one intends a video, i.e. something that can be played back directly on a screen and saved in digital format) of a pulse of light in flight. We will discuss a different technology that is based on an imaging camera with a pixel array in which each individual pixel is a single photon avalanche diode (SPAD). SPADs offer both sensitivity to single photons and picosecond temporal resolution of the photon arrival time (with respect to a trigger event). When adding imaging capability, SPAD arrays can deliver videos of light pulse propagating in free space, without the need for a scattering medium or diffuser as in all previous work (Gariepy et al). This capability can then be harnessed for a variety of applications. We will discuss the details of SPAD camera detection of moving objects (e.g. human beings) that are hidden from view and then conclude with a discussion of future perspectives in the field of bio-imaging.

In the last years, single-pixel imaging (SPI) was established as a suitable tool for non-invasive imaging of an absorbing object completely embedded in an inhomogeneous medium. One of the main characteristics of the technique is that it uses very simple sensors (bucket detectors such as photodiodes or photomultiplier tubes) combined with structured illumination and mathematical algorithms to recover the image. This reduction in complexity of the sensing device gives these systems the opportunity to obtain images at shallow depth overcoming the scattering problem. Nonetheless, some challenges, such as the need for improved signal-to-noise or the frame rate, remain to be tackled before extensive use in practical systems. Also, for intact or live optically thick tissues, epi-detection is commonly used, while present implementations of SPI are limited to transillumination geometries. In this work we present new features and some recent advances in SPI that involve either the use of computationally efficient algorithms for adaptive sensing or a balanced detection mechanism. Additionally, SPI has been adapted to handle reflected light to create a double pass optical system. Such developments represent a significant step towards the use of SPI in more realistic scenarios, especially in biophotonics applications. In particular, we show the design of a single-pixel ophtalmoscope as a novel way of imaging the retina in real time.

Optical microscopy is an indispensable tool for researchers, allowing them to closely investigate different organisms, revealing new features and phenomena in biomedical research. Although very useful, conventional imaging techniques that rely only on ballistic, unaffected photons to form images inside inhomogeneous media, like biological tissue, are eventually limited up to the diffusion regime of optical propagation where scattering becomes dominant and no ballistic light can be detected. Adaptive optics and nonlinear optimization methods that rely on so called guide stars have been employed to overcome this problem and image deeper inside biological tissue. These techniques attempt to recover the optimal wavefront that will enhance the image quality or that will render a focus spot inside the scattering biological tissue. In order to achieve that, they have to iterate through each correction mode (e.g. each pixel on a wavefront shaper) thus trading off measurement time with wavefront resolution. Here we present a new turbidity suppression approach, termed Focus Scanning Holographic Aberration Probing (F-SHARP or F♯) that allows us to directly measure the amplitude and phase of the scattered light distribution at the focal plane (scattered E-field PSF). Knowledge of the E-field enables rapid correction of both aberration and scattering with a high resolution. We demonstrate the power of F-SHARP by correcting for aberration and scattering and imaging neuronal structures through the larval zebrafish and mouse brain and through thinned mouse skull in vivo.

By controlling the many degrees of freedom in the incident wavefront, one can manipulate wave propagation in complex structures. Such wavefront-shaping methods have been used extensively for controlling light transmitted into wavelength-scale regions (speckles), a property that is insensitive to correlations in the speckle pattern. Extending coherent control to larger regions is of great interest both scientifically and for applications such as optical communications, photothermal therapy, and the imaging of large objects within or behind a diffusive medium. However, waves diffusing through a disordered medium are known to exhibit non-local intensity correlations, and their effect on coherent control has not been fully understood. Here, we demonstrate the effects of correlations with wavefront-shaping experiments on a scattering sample of zinc oxide microparticles. Long-range correlations substantially increase the dynamic range of coherent control over light transmitted onto larger target regions, far beyond what would be achievable if correlations were negligible. This and other effects of correlations emerge when the number of speckles targeted, M2, exceeds the dimensionless conductance g. Using a filtered random matrix ensemble appropriate for describing coherent diffusion and the lateral spreading in an open geometry, we show analytically that M2/g appears as the controlling parameter in universal scaling laws for several statistical properties of interest---predictions that we quantitatively confirm with experimental data. Our work elucidates the roles of speckle correlations and provides a general theoretical framework for modeling open systems in wavefront-shaping experiments.

Recent breakthroughs in optical wavefront engineering have opened the possibility of controlling light intensity distribution inside highly scattering medium, but their success is limited by the open geometry of the sample and the difficulty of covering all input modes. Here we demonstrate experimentally an efficient control of energy density distribution inside a strong scattering medium. Instead of the open slab geometry, we fabricate a silicon waveguide that contains scatterers and has reflecting sidewalls. The intensity distribution inside the 2D waveguide is probed from the third dimension. With a careful design of the on-chip coupling waveguide, we can access all the input modes. Such unprecedented control of incident wavefront leads to 10 times enhancement of the total transmission or 50 times suppression. A direct probe of light intensity distribution inside the disordered structure reveals that selective excitation of open channels leads to an energy buildup deep inside the scattering medium, while the excitation of closed channels greatly reduces the penetration depth. Compared to the linear decay for random input fields, the optimized wavefront can produce an intensity profile that is either peaked near the center of the waveguide or decay exponentially with depth. The total energy stored inside the waveguide is increased 3.7 times or decreased 2 times. Since the energy density dictates light-matter interactions inside a scattering system, our results demonstrate the possibility of tailoring optical excitations as well as linear and nonlinear optical processes inside the turbid medium in an on-chip platform.

Developing an efficient strategy for light focusing through scattering media is an important topic in the study of multiple light scattering. The enhancement factor of the light focusing, defined as the ratio between the optimized intensity and the background intensity is proportional to the number of controlling modes in a spatial light modulator (SLM). The demonstrated enhancement factors in previous studies are typically less than 1,000 due to several limiting factors, such as the slow refresh rate of a LCoS SLM, long optimization time, and lack of an efficient algorithm for high controlling modes. A digital micro-mirror device is an amplitude modulator, which is recently widely used for fast optimization through dynamic biological tissues. The fast frame rate of the DMD up to 16 kHz can also be exploited for increasing the number of controlling modes. However, the manipulation of large pattern data and efficient calculation of the optimized pattern remained as an issue. In this work, we demonstrate the enhancement factor more than 100,000 in focusing through scattering media by using 1 Mega controlling modes of a DMD. Through careful synchronization between a DMD, a photo-detector and an additional computer for parallel optimization, we achieved the unprecedented enhancement factor with 75 mins of the optimization time. We discuss the design principles of the system and the possible applications of the enhanced light focusing.

When an ultrashort pulse of light propagates in a scattering medium, its spatial and temporal properties get mixed and distorted because of the scattering process. Spatially, the output pattern is the result of the multiple interference between the scattered photons. Temporally, light gets stretched within the medium due to its characteristic confinement time, thus the output pulse is broadened in the time domain. Nonetheless, as the scattering process is linear and deterministic, the spatio-temporal profile of light at the output can be controlled by shaping the input light using a single spatial light modulator (SLM). We report the first experimental measurement of the Time-Resolved Transmission Matrix of a multiple scattering medium using a coherent time-gated detection system. This operator contains the relationship between the input field, controllable with a SLM, and the output field accessible with a CCD camera for a given arrival time of photons at the output of medium. The delay line of the time-gated detection system sets the arrival time at will within the time of flight distribution of photons of the output pulse. We exploit this time-resolved matrix to achieve spatio-temporal focusing of the output pulse at any arbitrary space and time position. The pulse is recompressed in time to its original Fourier-limited temporal width and spatially to the diffraction-limited size defined by the speckle grain size. We also generate more sophisticated spatio-temporal profiles such as pump-probe like pulse, thus opening interesting perspectives in coherent control, light-matter interaction and imaging in disordered media.

The propagation of light in biological tissues is rapidly dominated by multiple scattering: ballistic light is exponentially attenuated, which limits the penetration depth of conventional microscopy techniques. For coherent light, the recombination of the different scattered paths creates a complex interference: speckle. Recently, different wavefront shaping techniques have been developed to coherently manipulate the speckle. It opens the possibility to focus light through complex media and ultimately to image in them, provided however that the medium can be considered as stationary. We have studied the possibility to focus in and through time-varying biological tissues. Their intrinsic temporal dynamics creates a fast decorrelation of the speckle pattern. Therefore, focusing through biological tissues requires fast wavefront shaping devices, sensors and algorithms. We have investigated the use of a MEMS-based spatial light modulator (SLM) and a fast photodetector, combined with FPGA electronics to implement a closed-loop optimization. Our optimization process is just limited by the temporal dynamics of the SLM (200µs) and the computation time (45µs), thus corresponding to a rate of 4 kHz. To our knowledge, it’s the fastest closed loop optimization using phase modulators. We have studied the focusing through colloidal solutions of TiO2 particles in glycerol, allowing tunable temporal stability, and scattering properties similar to biological tissues. We have shown that our set-up fulfills the required characteristics (speed, enhancement) to focus through biological tissues. We are currently investigating the focusing through acute rat brain slices and the memory effect in dynamic scattering media.

The shower-curtain effect is a familiar phenomenon, routinely observed in our everyday life: an object placed behind a scattering layer appears blurred but if the object is attached to the scattering layer it can be clearly resolved. The optical system we developed takes advantage of the shower-curtain effect properties and generalizes them to achieve high-resolution imaging of objects placed at a nearly arbitrary distance behind the scattering medium. The imaging procedure is based on retrieving the object Fourier transform from the turbid medium (used as the shower-curtain) through a correlography technique based on speckle illumination. Illuminating the object with a speckle pattern rather than a coherent beam, we show that the correlography principles can be effectively applied in the near field. While the far-field condition is usually known as z<(2D^2)⁄λ (D, size of the object; λ wavelength); by tuning the spatial coherence of the illumination beam, as one can do with speckle illumination, the “far-field” condition can be written as z<(2DRc)⁄λ where Rc is the correlation radius of the speckle pattern. Using our method we present high-resolution imaging of objects hidden behind millimeter-thick tissue or dense lens cataracts, and demonstrate our imaging technique to be insensitive to rapid medium movements (<5 m∕s) beyond any biologically relevant motion. Furthermore, we show this method can be extended to several contrast mechanisms and imaging configurations.

Human subjects can detect infrared light at wavelengths over 1000 nm perceived as visible of the corresponding half wavelength. This is due to a two-photon process and requires the use of pulsed light sources well focused within the retina. We have developed an experimental system to measure, for the first time, the visual resolution of the eye when is stimulated with infrared (1043 nm) and compared with visible light (543 nm). Scanner mirrors were used to project letters of different sizes onto the retina in both wavelengths. Subjects performed a visual test to determine the smallest letter size that was distinguishable for each wavelength for a range of defocus values. An additional optical path was used to record the retinal images of the spot after reflection in the retina and double-pass through the optical media. The best visual acuity was obtained at different defocus locations corresponding to the chromatic difference between green and infrared. Although, there was some individual variability, visual acuity was found to be similar both in visible and infrared. The use of two-photon infrared vision may have some potential applications for vision in those cases were the optical media is opaque to visible wavelengths while keeping some transparency in the infrared.

Light propagation in a turbid medium is typically considered for flat or regular surfaces. However, such an approximation often does not reflect an experimental reality, and, in this report, we attempt to optimize the surface of a scattering medium to improve the optical coupling into the medium. By making conical microchannels in a turbid medium using short-pulsed laser micro-drilling, we show that we were able to substantially increase the photon life-time and diffusion radius in the medium.

The turbid nature of refractive index distribution within living tissues introduces severe aberrations to light propagation thereby severely compromising image reconstruction using currently available non-invasive techniques. Numerous approaches of endoscopy, based mainly on fibre bundles or GRIN-lenses, allow imaging within extended depths of turbid tissues, however their footprint causes profound mechanical damage to all overlying regions and their imaging performance is very limited. Progress in the domain of complex photonics enabled a new generation of minimally invasive, high-resolution endoscopes by substitution of the Fourier-based image relays with a holographic control of light propagating through apparently randomizing multimode optical waveguides. This form of endo-microscopy became recently a very attractive way to provide minimally invasive insight into hard-to-access locations within living objects. Here, we review our fundamental and technological progression in this domain and introduce several applications of this concept in bio-medically relevant environments. We present isotropic volumetric imaging based on advanced modes of light-sheet microscopy: by taking advantage of the cylindrical symmetry of the fibre, we facilitate the wavefront engineering methods for generation of both Bessel and structured Bessel beam plane illumination. Further, we demonstrate the first utilization of multimode fibers for imaging in living organisms. We present a new fibre-based geometry for deep tissue imaging in brain tissue of a living animal model. Lastly, we show the development and exploitation of highly specialised fiber probes for numerous advanced bio-photonics applications including high-resolution imaging and optical manipulation.

The periodic arrangement of core positions in multi-core fiber bundles introduces ‘ghost’ artifacts to endoscopic images obtained through them, whether in wide-field imaging (based on either direct imaging or speckle correlations) or in confocal scanning microscopy using wavefront shaping. Here we introduce partially disordered multi-core bundles as a means to overcome these artifacts. The benefits of their use will be discussed in the context of multiphoton scanning microscopy utilizing a spatial light modulator in the proximal end, and in the more general case of widefield imaging. We also show that both numerically and experimentally that the presence of disorder also enables to apply phase retrieval methods to characterize the phase distortion introduced due to propagation in the bundle without the need of an interferometrically stabilized reference. Thus, in addition to overcoming the challenge of ghost artifacts, disordered multi-core fibers have the potential to overcome another challenge, movement-induced phase distortions, by enabling real-time characterization of this phase distortion in reflection mode only via the proximal end.

A multimode fiber subjected to random mode and polarization mixing represents a complex photonic system with strong coupling of spatial, temporal, spectral and polarization degrees of freedom. By exploiting such coupling, we demonstrate a full control of the polarization state of light transmitted through a multimode fiber by adjusting the spatial profile of incident field. After applying stress to the fiber to induce strong mode and polarization mixing, we measure the polarization-dependent transmission matrices, and find the transmission eigenchannels. By launching light to specific eigenchannnels, we are able to preserve the polarization state despite strong polarization mixing in the multimode fiber, or to convert all transmitted light to the orthogonal polarization state. In addition, we show that the linearly polarized input light can be changed completely to circularly polarized output. Furthermore, arbitrary polarization states can be realized for individual spatial channels at the output by tailoring the incident wavefront of a single polarization. Such global control is possible only in the presence of strong mode mixing in the fiber. Namely, strong polarization mixing itself is not sufficient to generate arbitrary polarization states at the output. Therefore, the strong entanglement of spatial and polarization degrees of freedom is essential to achieve a complete control of polarization. Such global control of the polarization states of all output channels is more challenging than the local control of the polarization state of a single output channel.

Optical pulses propagating through a multimode fiber with random mode mixing experience temporal broadening and distortion. Principal modes have been proposed to overcome modal dispersion. They are the eigenstates of the time delay operator and the associated eigenvalues are the delay times. Principal modes retain the spatial profiles of output fields to the first order of frequency variation. In the weak mode coupling regime, principal modes are superpositions of fiber eigenmodes with similar propagation constants. In the strong mode coupling regime, a principal mode is composed of all fiber modes with very different propagation constant, yet it has a well-defined delay time due to multipath interference, which can be controlled by adjusting the spatial profile of incident field. The spectral bandwidth of principal modes determines the temporal width of optical pulses that can be transmitted through the multimode fiber without distortion. In the weak mode coupling regime, principal modes with short and long delay times have broader bandwidths, while in the strong mode coupling regime, the principal modes with intermediate delay times have the broadest bandwidths. The opposite behaviors reveal two distinct mechanisms that are responsible for the principal mode bandwidth in the weak and strong mode coupling regimes. We further investigate how the mode-dependent loss modifies the principal modes. Our study provides physical understanding of spatiotemporal dynamics in a multimode fiber with varying degree of mode mixing, which is important for controlling pulse propagation through a multimode fiber.

Using spatial light modulators(SLM) to control light propagation through scattering media is a critical topic for various applications in biomedical imaging, optical micromanipulation, and fibre endoscopy. Having limited switching rate, typically 10-100Hz, current liquid-crystal SLM can no longer meet the growing demands of high-speed imaging. A new way based on binary-amplitude holography implemented on digital micromirror devices(DMD) has been introduced recently, allowing to reach refreshing rates of 30kHz. Here, we summarise the advantages and limitations in speed, efficiency, scattering noise, and pixel cross-talk for each device in ballistic and diffusive regimes, paving the way for high-speed imaging through multimode fibres.

Multiphoton fluorescence microscopy is a well-established technique for deep-tissue imaging with subcellular resolution. Three-photon microscopy (3PM) when combined with long wavelength excitation was shown to allow deeper imaging than two-photon microscopy (2PM) in biological tissues, such as mouse brain, because out-of-focus background light can be further reduced due to the higher order nonlinear excitation. As was demonstrated in 2PM systems, imaging depth and resolution can be improved by aberration correction using adaptive optics (AO) techniques which are based on shaping the scanning beam using a spatial light modulator (SLM). In this way, it is possible to compensate for tissue low order aberration and to some extent, to compensate for tissue scattering. Here, we present a 3PM AO microscopy system for brain imaging. Soliton self-frequency shift is used to create a femtosecond source at 1675 nm and a microelectromechanical (MEMS) SLM serves as the wavefront shaping device. We perturb the 1020 segment SLM using a modified nonlinear version of three-point phase shifting interferometry. The nonlinearity of the fluorescence signal used for feedback ensures that the signal is increasing when the spot size decreases, allowing compensation of phase errors in an iterative optimization process without direct phase measurement. We compare the performance for different orders of nonlinear feedback, showing an exponential growth in signal improvement as the nonlinear order increases. We demonstrate the impact of the method by applying the 3PM AO system for in-vivo mouse brain imaging, showing improvement in signal at 1-mm depth inside the brain.

Optical imaging usually suffers from aberrations that are induced by various structures when imaging biological samples. Usually aberrations degrade the imaging system performances by broadening the point spread function (PSF). Unexpectedly we show that in spatially incoherent interferometry like full-filed optical coherence tomography (FFOCT), the system PSF width is almost insensitive to aberrations. Instead of considering the PSF of a classical imaging system such as a microscope, we specifically pay attention to the system PSF of interferometric imaging systems for which an undistorted wavefront from a reference beam interferes with the distorted wavefront of the object beam. By comparing the cases of scanning OCT with spatially coherent illumination, wide-field OCT with spatially coherent illumination and FFOCT with spatially incoherent illumination, we found that in FFOCT with spatially incoherent illumination the system PSF width is almost independent of the aberrations and only its amplitude varies. This is demonstrated by theoretical analysis as well as numerical calculations for different aberrations, and confirmed by experiments with a FFOCT system. It is the first time to the best of our knowledge that such specific merit of incoherent illumination in FFOCT has been demonstrated. Based on this, the signal level is used as metric in our adaptive optics FFOCT system for retinal imaging. Only the main aberrations (defocus and astigmatism) that are dominating in eye are corrected to improve the signal to noise ratio and the high order aberrations are skipped. This would increase the correction speed thus reducing the imaging time.

AO (Adaptive Optics) corrects wavefront errors to improve imaging quality in optical systems. An AO-system consist often of a SH-WFS (Shack–Hartmann wavefront sensor) and a DM (deformable mirror). The SH-WFS measures the local slopes of the wave front and iteratively calculates from these slopes the best fitting wavefront. The shape of the DM is then controlled by this information. Any error in the slope measurement (noise) will result in a residual wavefront error and hence in a reduced image quality. The wavefront error detection method is based on the fact that the wavefront slopes have to be integrable and allows to quantify the error in the wavefront slopes measurement. The integrable wavefront derived from the measured slopes is used to re-calculate the slopes. The difference between the re-calculated slopes and the measured slopes is identified as the none-integrable noise of the slopes measurement. The total noise is the sum of the integrable and the none-integrable noise. In order to derive a relation between the integrable and none-integrable noise 1000 measurements of the same wavefront have been taken. The average is assumed to be the noise free wave front. This wave front has been used to calculate the total noise of every single measurement. Using this information an approximation of the total noise was found as: Total noise = None-integrable noise * 1.265. This information can be used as an objective criterion for the quality of the wavefront measurement and to evaluate if the imagine performance is limited by the wavefront measurement or by the deformable mirror (e.g. number of actuator).

The landscape of biomedical research in neuroscience has changed dramatically in recent years as a result of spectacular progress in dynamic microscopy. In this context Adaptive Optics allows in-depth imaging by correcting aberrations induced by the biological sample, the key issue being then the ability to perform an accurate and reliable wavefront sensing (WFS). We present here a limitation of modal sensorless WFS in the case of a heterogeneous medium. We then build a new method called Axially-Locked Modal Sensorless (ALMS) that exploits these heterogeneities to overcome this limitation. The new method is simulated and compared to standard modal sensorless. The simulation results show a more accurate wavefront estimation even in the case of a strongly aberrated biological media.

We present the optimization of an adaptive optics loop for retinal imaging. Generally, the wave-front is overdetermined compared to the number of corrector elements. The sampling of the sensor can be reduced while maintaining an efficient correction, leading to higher sensitivity, faster correction and larger dynamic range. An analytical model was developed to characterize the link between number of actuators, number of micro-lenses and correction performance. The optimized correction loop was introduced into a scanning laser ophthalmoscope. In vivo images of foveal photoreceptors were recorded and the obtained image quality is equivalent to the state of the art in retinal AO-imaging.

Optical Coherence Tomography (OCT) has revolutionized modern ophthalmology, providing depth resolved images of the retinal layers in a system that is suited to a clinical environment. A limitation of the performance and utilization of the OCT systems has been the lateral resolution. Through the combination of wavefront sensorless adaptive optics with dual variable optical elements, we present a compact lens based OCT system that is capable of imaging the photoreceptor mosaic. We utilized a commercially available variable focal length lens to correct for a wide range of defocus commonly found in patient eyes, and a multi-actuator adaptive lens after linearization of the hysteresis in the piezoelectric actuators for aberration correction to obtain near diffraction limited imaging at the retina. A parallel processing computational platform permitted real-time image acquisition and display. The Data-based Online Nonlinear Extremum seeker (DONE) algorithm was used for real time optimization of the wavefront sensorless adaptive optics OCT, and the performance was compared with a coordinate search algorithm. Cross sectional images of the retinal layers and en face images of the cone photoreceptor mosaic acquired in vivo from research volunteers before and after WSAO optimization are presented. Applying the DONE algorithm in vivo for wavefront sensorless AO-OCT demonstrates that the DONE algorithm succeeds in drastically improving the signal while achieving a computational time of 1 ms per iteration, making it applicable for high speed real time applications.

Multiphoton imaging through the bone to image into the bone marrow or the brain is an emerging need in the scientific community. Due to the highly scattering nature of bone, bone thinning or removal is typically required to enhance the resolution and signal intensity at the imaging plane. The optical aberrations and scattering in the bone significantly affect the resolution and signal to noise ratio of deep tissue microscopy. Multiphoton microscopy uses long wavelength (nearinfrared and infrared) excitation light to reduce the effects of scattering. However, it is still susceptible to optical aberrations and scattering since the light propagates through several layers of media with inhomogeneous indices of refraction. Mechanical removal of bone is highly invasive, laborious, and cannot be applied in experiments where imaging inside of the bone is desired. Adaptive optics technology can compensate for these optical aberrations and potentially restore the diffraction limited point spread function of the system even in deep tissue. To design an adaptive optics system, a priori knowledge of the sample structure assists selection of the proper correction element and sensing methods. In this work we present the characterization of optical aberrations caused by mouse cranial bone, using second harmonic generation imaging of bone collagen. We simulate light propagation through the bone, calculate aberrations and determine the correction that can be achieved using a deformable mirror.

Adaptive optics (AO) is essential for achieving diffraction limited resolution in large numerical aperture (NA) in-vivo retinal imaging in small animals. Cellular-resolution in-vivo imaging of fluorescently labeled cells is highly desirable for studying pathophysiology in animal models of retina diseases in pre-clinical vision research. Currently, wavefront sensor-based (WFS-based) AO is widely used for retinal imaging and has demonstrated great success. However, the performance can be limited by several factors including common path errors, wavefront reconstruction errors and an ill-defined reference plane on the retina. Wavefront sensorless (WFS-less) AO has the advantage of avoiding these issues at the cost of algorithmic execution time. We have investigated WFS-less AO on a fluorescence scanning laser ophthalmoscopy (fSLO) system that was originally designed for WFS-based AO. The WFS-based AO uses a Shack-Hartmann WFS and a continuous surface deformable mirror in a closed-loop control system to measure and correct for aberrations induced by the mouse eye. The WFS-less AO performs an open-loop modal optimization with an image quality metric. After WFS-less AO aberration correction, the WFS was used as a control of the closed-loop WFS-less AO operation. We can easily switch between WFS-based and WFS-less control of the deformable mirror multiple times within an imaging session for the same mouse. This allows for a direct comparison between these two types of AO correction for fSLO. Our results demonstrate volumetric AO-fSLO imaging of mouse retinal cells labeled with GFP. Most significantly, we have analyzed and compared the aberration correction results for WFS-based and WFS-less AO imaging.

Two-photon (2P) excitation can be combined with phase-modulation approaches, such as computer-generated holography (CGH), to efficiently distribute light into two-dimensional, axially confined, user-defined shapes. Applications include lithography, uncaging, optogenetics and fast functional imaging. However, a linear proportionality between lateral shape area and axial extent degrades axial precision for cases demanding extended lateral patterning.To address this limitation, we previously combined CGH with temporal focusing (TF) to stretch laser pulses outside of the focal plane, which combined with 2P’s nonlinear fluorescence dependence, axially confines fluorescence regardless of lateral extent. However, this configuration restricts nonlinear excitation to a single spatiotemporal focal plane, which is the objective focal plane. Here we report a novel optical scheme enabling remote axial displacement and simultaneous generation of spatiotemporally focused pattern at multiple planes using two spatial light modulators to independently control transverse- and axial-target light distribution. This approach enabled simultaneous axial translation of single or multiple spatiotemporal focused patterns across the sample volume, while achieving the axial confinement of temporal focusing. We utilized the system's novel capability to dissect the functional connectivity between axially distinct neuronal layers in the mice retina. Finally, we demonstrated that TF enables robust light propagation trough optically and physiologically diverse neural systems including mice brain, zebrafish larva brain and mice retina.

Laser beam propagation through the scattering suspension of polystyrene microspheres in distilled water was studied. The distorted laser beam was analyzed by both Shack-Hartmann sensor and CCD camera. The measured local slopes of the Poynting vector were compensated for by means of bimorph deformable mirror with 14 electrodes in order to increase the intensity of the focal spot in the far-field. Three different techniques for laser beam focusing were implemented and compared: LSQ fit-error minimization by Shack-Hartmann sensor, Hill-climbing optimization by Shack-Hartmann sensor and Hill-climbing optimization by far-field CCD camera.