Search

Microscopy and optical imaging are drastically limited by the inhomogeneities encountered by the light while propagating from the object of interest to the detection system. In this context, adaptive optics and wavefront manipulation are able to improve the contrast (visibility) of systems embedded in turbid and noisy environments. By wavefront shaping, the fluence of the light propagating through complex systems can be controlled, thus, confining the light in a defined microscopic region in the volume or at the back of scattering structures. We can imagine to counterintuitively exploit the optical barriers, turning them into scattering lenses. In this Thesis we consider these new generation of lenses: we demonstrate them to be configurable optical devices able to produce tailored light structures, hence resulting extremely advantageous if integrated in optical systems.We have initially studied their focusing spatial resolution at different scattering strengths, thereafter we have developed a method that allows to overcome the present limitations and to scale down their resolution of a factor 3 if compared with the previous standard methods. This result has been obtained selecting transmitting optical modes that are directly responsible for the resolution improvement. Therefore, we show that properly filtering the light transmitted at the back of a scatterer we can also obtain numerous novel optical features. Specifically, we demonstrate the full control of non-diffractive light structures named amorphous speckle patterns. Under wavefrontshaping manipulation those complex interference patterns can be converted to Bessel-like beams. The method discloses the possibility to produce Bessel beams at the selected location at the back of scattering barriers. The results presented have potential for providing arranged structured illumination and for enhancing the penetration depth in microscopy providing optimal imaging resolution of thick samples. Thence, complex scattering systems have been recently used as programmable stand-alone optical devices. We have foreseen their operation in optical circuits, for this reason we designed micro-fabricated scattering systems in the bulk of cover-slip glasses. Our photonic structures are proved to be unreproducible and resistant to the deterioration, hence resulting suitable for authentication control operations. Moreover, fashioning the position and thegeometry of the single scatterers via laser ablation we can engineer the process of multiple scattering in order to produce structured illuminations. In particular, we test an anisotropic geometry for the formation of light sheets. When we compare our system with conventionalcylindrical lenses we observe a significant spatial resolution enhancement in the light sheet generated. In addition, our opaque lenses result fully tunable, a property that can allow forbypassing mechanical adjustment for chromatic correction or axial scanning in selected planeillumination microscopy. Ultimately, we present a lab-on-a-chip: a programmable platformthat can support and inspect biological systems of interest in a controlled environment.

Optical imaging techniques are widely used in biological research, but their penetration depth is limited by tissue scattering. Wavefront shaping techniques are able to overcome this problem in principle, but are often slow and their performance depends on the sample. This greatly reduces their practicability for biological applications. Here we present a scattering compensation technique based on three-photon (3P) excitation, which converges faster than comparable two-photon (2P) techniques and works reliably even on densely labeled samples, where 2P approaches fail. To demonstrate its usability and advantages for biomedical imaging we apply it to the imaging of dendritic spines on GFP-labeled layer 5 neurons in an anesthetized mouse.

Optical imaging techniques are widely used in biological research, but their penetration depth is limited by tissue scattering. Wavefront shaping techniques are able to overcome this problem in principle, but are often slow, and their performance depends on the sample. This greatly reduces their practicability for biological applications. Here we present a scattering compensation technique based on three-photon excitation, which converges faster than comparable two-photon (2P) techniques and works reliably even on densely labeled samples, where 2P approaches fail. To demonstrate its usability and advantages for biomedical imaging, we apply it to the imaging of dendritic spines on layer 5 neurons labeled with green fluorescent protein in an anesthetized mouse.

The reconstruction of an object hidden behind a scattering curtain is a modern topic in the field of imaging, which has stimulated an active scientific production over the past few years. However, most of the work done in the field was in addressing the reconstruction of a bi-dimensional object. Here, instead, we tackle the reconstruction of a three-dimensional fluorescent sample hidden behind an opaque layer. To do so, we show that the auto-correlation operation well behave in projection tomography, letting us to reconstruct a three-dimensional auto-correlation of the object. By having access to such information, it is possible to implement a phase retrieval algorithm to roll back to the actual reconstruction of the specimen.

Fluorescence tomography is a well-established methodology able to provide structural and functional information on the measured object. At optical wavelengths, the unpredictable scattering of light is often considered a problem to overcome, rather than a feature to exploit. Advances in disordered photonics have shed new light on possibilities offered by opaque materials, treating them as autocorrelation lenses able to create images and focus light. In this Letter, we propose tomography through disorder, introducing a modified Fourier-slice theorem, the cornerstone of the computed tomography, aiming to reconstruct a three-dimensional fluorescent sample hidden behind an opaque curtain.

We discuss the problem of tomographic reconstruction of fluorescent objects hidden behind random media. To accomplish this, we focus on the properties of the autocorrelation, relying on phase retrieval algorithms to perform 3D reconstruction.

Light sheet fluorescence microscopy techniques have revolutionized biological microscopy enabling low-phototoxic long-term 3D imaging of living samples. Although there exist many light sheet microscopy (LSM) implementations relying on fluorescence, just a few works have paid attention to the laser elastic scattering source of contrast available in every light sheet microscope. Interestingly, elastic scattering can potentially disclose valuable information from the structure and composition of the sample at different spatial scales. However, when coherent scattered light is detected with a camera sensor, a speckled intensity is generated on top of the native imaged features, compromising their visibility. In this work, we propose a novel light sheet based optical setup which implements three strategies for dealing with speckles of elastic scattering images: (i) polarization filtering; (ii) reducing the temporal coherence of the excitation laser light; and, (iii) reducing the spatial coherence of the light sheet. Finally, we show how these strategies enable pristine light-sheet elastic-scattering imaging of structural features in challenging biological samples avoiding the deleterious effects of speckle, and without relying on, but complementing, fluorescent labelling.

Hyperuniform structures possess the ability to confine and drive light, although their fabrication is extremely challenging. Here we demonstrate that speckle patters obtained by a superposition of randomly arranged sources of Bessel beams can be used to generate hyperunifrom scalar fields. By exploiting laser light tailored with a spatial filter, we experimentally produce (without requiring any computational power) a speckle pattern possessing maxima at locations corresponding to a hyperuniform distribution. By properly filtering out intensity fluctuation from the same speckle pattern, it is possible to retrieve an intensity profile satisfying the hyperuniformity requirements. Our findings are supported by extensive numerical simulations., Comment: 13 pages, 7 figures

Optical tomographic imaging of biological specimen bases its reliability on the combination of both accurate experimental measures and advanced computational techniques. In general, due to high scattering and absorption in most of the tissues, multi view geometries are required to reduce diffuse halo and blurring in the reconstructions. Scanning processes are used to acquire the data but they inevitably introduces perturbation, negating the assumption of aligned measures. Here we propose an innovative, registration free, imaging protocol implemented to image a human tumor spheroid at mesoscopic regime. The technique relies on the calculation of autocorrelation sinogram and object autocorrelation, finalizing the tomographic reconstruction via a three dimensional Gerchberg Saxton algorithm that retrieves the missing phase information. Our method is conceptually simple and focuses on single image acquisition, regardless of the specimen position in the camera plane. We demonstrate increased deep resolution abilities, not achievable with the current approaches, rendering the data alignment process obsolete.

We describe a computational method for accurate, quantitative tomographic reconstructions in Optical Projection Tomography, based on phase retrieval algorithms. Our method overcomes limitations imposed by light scattering in opaque tissue samples under the memory effect regime, as well as reduces artifacts due to mechanical movements, misalignments or vibrations. We make use of Gerchberg-Saxton algorithms, calculating first the autocorrelation of the object and then retrieving the associated phase under four numerically simulated measurement conditions. By approaching the task in such a way, we avoid the projection alignment procedure, exploiting the fact that the autocorrelation sinogram is always aligned and centered. We thus propose two new, projection-based, tomographic imaging flowcharts that allow registration-free imaging of opaque biological specimens and unlock three-dimensional tomographic imaging of hidden objects. Two main reconstruction approaches are discussed in the text, focusing on their efficiency in the tomographic retrieval and discussing their applicability under four different numerical experiments.

Optical tomography in biomedical imaging is a highly dynamic field in which non-invasive optical and computational techniques are combined to obtain a three dimensional representation of the specimen we are interested to image. Although at optical wavelengths scattering is the main obstacle to reach diffraction limited resolution, recently several studies have shown the possibility to image even objects fully hidden behind a turbid layer exploiting the information contained in the speckle autocorrelation via an iterative phase retrieval algorithm. In this work we explore the possibility of blind three dimensional reconstruction approach based on the Optical Projection Tomography principles, a widely used tool to image almost transparent model organism such as C. Elegans and D. Rerio. By using autocorrelation information rather than projections at each angle we prove, both numerically and experimentally, the possibility to perform exact three dimensional reconstructions via a specifically designed phase retrieval algorithm, extending the capability of the projection-based tomographic methods to image behind scattering curtains. The reconstruction scheme we propose is simple to implement, does not require post-processing data alignment and moreover can be trivially implemented in parallel to fully exploit the computing power offered by modern GPUs, further reducing the need for costly computational resources.

The combined use of a wavefront modulator and a scattering medium forms an "opaque lens" which forces the light to focus tightly. The adaptive focus has the same shape as the correlation function of the original speckle pattern and it can be generated at defined positions with resolution up to hundreds of nanometers. We have demonstrated that manipulating the speckle pattern spatial components can structure the shape of the focus. Exploiting selectively spatial-frequencies from the speckle components we realized opaque lenses able to produce sub-correlation foci and Bessel beams.

Recently great progress has been made in biological and biomedical imaging by combining non-invasive optical methods, novel adaptive light manipulation and computational techniques for intensity-based phase recovery and three dimensionalimage reconstruction. In particular and in relation to the work presented here, Optical Projection Tomography (OPT) is awell-established technique for imaging mostly transparent absorbing biological models such as C. Elegans and Danio Rerio. On the contrary, scattering layers like the cocoon surrounding the Drosophila during the pupae stage constitutes achallenge for three dimensional imaging through such a complex structure. However, recent studies enabled image reconstruction through scattering curtains up to few transport mean free paths via phase retrieval iterative algorithms allowing to uncover objects hidden behind complex layers. By combining these two techniques we explore the possibility to perform a three dimensional image reconstruction of fluorescent objects embedded between scattering layers without compromising its structural integrity. Dynamical cross correlation registration was implemented for the registration process due to translational and flipping ambiguity of the phase retrieval problem, in order to provide the correct aligned set of data to perform the back-projection reconstruction. We have thus managed to reconstruct a hidden complex object between static scattering curtains and compared with the effective reconstruction to fully understand the process before the in-vivo biological implementation. Keywords: Phase Retrieval , Scattering, Optical Projection Tomography, Three Dimensional Imaging, BiomedicalImaging, Image Reconstruction , Turbid Lens , Memory Effect.

Bessel beams are non-diffracting light structures, which maintain their spatial features after meters of propagation and are realized with simple optical elements such as axicon lenses, spatial filters, and lasers. In this paper, we demonstrate a method for generating non diffractive Bessel-like beams through a heavily scattering system, exploiting wavefronts shaped by a spatial light modulator. With the proposed method starting from amorphous speckle patterns, it is possible to produce at user defined positions configurable and non-diffracting light distributions which can improve depth-of-field in speckled illumination microscopy.

Adaptive optics can focus light through opaque media by compensating the random phase delay acquired while crossing a scattering curtain. The technique is commonly exploited in many fields, including astrophysics, microscopy, biomedicine and biology. A turbid lens has the capability of producing foci with a resolution higher than conventional optics, however it has a fundamental limit: to obtain a sharp focus one has to introduce a strongly scattering medium in the optical path. Indeed a tight focusing needs strong scattering and, as a consequence, high resolution focusing is obtained only for weakly transmitting samples. Here we describe a novel method allowing to obtain highly concentrated optical spots even by introducing a minimum amount of scattering in the beam path with semi-transparent materials. By filtering the pseudo-ballistic components of the transmitted beam we are able to experimentally overcome the limits of the adaptive focus resolution, gathering light on a spot with a diameter which is one third of the original speckle correlation function.

By compensating the random phase delay acquired while a light beam crosses a scattering curtain, it is possible to address the light at selected target position beyond the obstacle. An opaque lens can produce foci with a resolution higher than conventional optics if a strongly scattering medium is exploited. In practice, subwavelength resolution is obtained only for weakly transmitting samples. Herein we present a method which allows obtaining tiny bright optical spots even in presence of a minimum amount of scattering (semi-transparent media) in the beam path. Using a High-Pass spatial filter we block the pseudo-ballistic components of the transmitted beam, we are able to gather light on a spot with a diameter which is one third of the typical speckle grain in absence of the filter.

We fabricated a relative humidity sensor based on a selectively infiltrated photonic crystal. We demonstrate how the reversible hydroscopic behavior of the embedded liquid into the air micropores impacts the photonic crystal waveguide cut-off frequency.

Strong scattering in the optical paths can be proactively exploited for determining light propagation and focusing through turbid media and ultimately improve optical imaging and light manipulation capabilities. The use of light shapers together with strongly scattering structures enables the production of foci confined in the nanoscale or on the other hand may provide significantly enhanced fields of view. Exploiting this concept, we introduce ad hoc engineered scattering structures, by direct femtosecond laser writing in the bulk of glasses, presenting isotropy in one dimension, and together with precisely phase-shaped illuminating light we produce submicrometer thin light sheets at user-defined positions. Our approach permits us to focus light of different wavelengths onto the same defined position without moving any optical element and correcting for chromatic aberrations. Furthermore, we demonstrate that natural biological tissues presenting isotropy in one dimension can also be used for generating light sheets directly in the biological sample.