Super-resolution microscopy allows the observation of sub-cellular structures with a resolution beyond diffraction limit of conventional fluorescence microscopy. However, most super-resolution microscopes have a limited imaging depth due to the inhomogeneous refractive index of the sample that leads to optical aberrations. Adaptive optics has been successfully adopted by many imaging techniques, including 3D Structured illumination microscopy (SIM). We use a fast deformable mirror to modulate the wavefront of fluorescence to compensate for optical aberration and changing focus position at the same time. Adaptive optics successfully extends the depth, the range and the speed of 3D-SIM imaging.

This paper proposes a fast whole-organ histological imaging method with real-time staining and mechanical sectioning. Time-consuming and laborious sample processing procedures are not needed. The imaged tissue block will be labeled along with the serial sectioning and optical scanning to improve the overall speed and the uniformity of staining. A super-resolution network (ESRGAN) and an optical-sectioning imaging technique (HiLo microscopy) have been applied to optimize the imaging speed and resolution. The proposed system can realize whole-organ histological imaging within hours to days, depending on the volume of the imaged sample.

Polarimetric microscopy has become a useful tool in multidimensional microscopy, especially for single molecule imaging where the orientation of a molecule is related to its polarization. Using an engineered scattering element in a photonic integrated circuit, we have produced a synthetic dipole source whose orientation is controllable through amplitude and phase of each waveguide.

Open-Top Light-Sheet (OTLS) microscopy is an emerging technique for cleared-tissue imaging that alleviates many of the physical constraints on sample size and geometry associated with conventional light-sheet microscopes. Existing OTLS designs generally use either two orthogonal objectives or a single shared objective for illumination and collection, however these architectures have various limitations for moderate NA cleared-tissue imaging. We previously developed an alternative Nonorthogonal Dual-Objective (NODO) OTLS configuration with a number of advantages for moderate-NA imaging of cleared tissues. Here we describe our latest NODO system, which is optimized specifically for high-throughput imaging of clinical samples for 3D pathology.

Light-Sheet Fluorescence Microscopy (LSFM) has proven to be an excellent tool for imaging the nervous system zebrafish and other model organisms. However, these LSFMs typically have low resolution and suffer from striping artifacts. Here, we present a light sheet system using a novel single-objective light-sheet geometry to allow for multi-direction light-sheet illumination with structured light. This geometry is designed to mitigate striping artifacts and achieve lateral resolutions of better than 200 nm over a 277 × 277 × 100 µm3 volume. We demonstrate the system by imaging the central nervous system of dissected Drosophila embryos as well as cerebellum organoids.

Single-Objective Light Sheet (SOLS) microscopes have in recent years emerged as a viable option for rapid and gentle three-dimensional imaging of biological samples. Though ingenious tricks have pushed the optical resolution of such systems to its physical limits, diffraction still forces a bothersome trade-off between light sheet thickness and propagation length. In this work we show that Reversibly Switchable Fluorescent Proteins (RSFPs) can be used to overcome this limit on light sheet thickness allowing fast volumetric imaging with approximately five to ten thinner effective light sheets compared to diffraction limited approaches. This volumetrically parallelized acquisition scheme opens up new possibilities for super-resolution imaging.

Anisotropic molecular alignment occurs ubiquitously and often heterogeneously at different size scales. However, conventional optical imaging approaches can only provide incomplete molecular orientation maps on the 2D polarization plane. Here, we present two new approaches based on polarization-controlled IR microscopy and broadband CARS microscopy to determine the 3D angles of molecular orientations independently at each image pixel. These new approaches are based on concurrent polarization analysis of multiple vibrational modes to map the 3D orientation angles and the order parameter of the local orientational distribution of polymer chains. We demonstrate the new technique using a non-banded poly(e-caprolactone) film and a ring-banded polyethylene film.

We propose an end-to-end image analysis pipeline based on the Adaptive Particle Representation (APR) for analyzing large 3D cleared tissue samples such as whole mouse brains or large human brain sections, achieving 100+ times faster computation. Our pipeline is compatible with real-time use, i.e. analysis can be done during acquisition. In addition to faster processing, using APR yields memory and storage compression ratios ranging from dozens to thousands depending on the labeling sparsity, saving costs on the storage and computing infrastructures.

Non-destructive 3D microscopy enables the accurate characterization of diagnostically and prognostically significant microstructures in clinical specimens with significantly increased volumetric coverage than traditional 2D histology. We are using open-top light-sheet microscopy to image prostate cancer biopsies and investigating the prognostic significance of 3D spatial features of nuclei within prostate cancer microstructures. Using a previously published 3D nuclear segmentation workflow, we identify a preliminary set of 3D graph-based nuclear features to quantify the 3D spatial arrangement of nuclei in prostate cancer biopsies. Using a machine classifier, we identify the features which prognosticate prostate cancer risk and demonstrate agreement with patient outcomes.

Optical coherence tomography (OCT) has been applied in analyzing the micron-level structure of human tissue, including collagen fiber orientation mapping in the depth of human tissue, which is essential for understanding human uterus mechanics. Previous research, due to the imaging depth limitation of OCT, offers multiple fiber orientation information of separate layers of uterus sample, which could not provide the full-thickness information for analyzing the change of fiber orientation along the depth. To solve this problem, this research provides a method to combine the separate fiber orientation information and give an analysis based on the depth of fiber location.

This conference presentation was prepared for SPIE BiOS, 2023.

The Microscopy Innovation Centre (MIC) at King’s College London, aims to provide advanced and home-built imaging solutions to its users. Such microscopy systems are often not provided by core facilities due to issues including the need for specific expertise, specific sample requirements, and continued maintenance. The MIC provides a home to such imaging platforms and has dedicated staff members to maintain as well as provide them to a wide user base. We present a general overview of the MIC as well as showing some of the work done using our home-built microscopes.

We present differential structured illumination microscopy (dSIM) for high-resolution, large field-of-view 3D computational phase imaging of biological samples. Using modified structured illuminations in a differential imaging scheme, we show our modality provides 4x resolution enhancements over the coherent imaging bandwidth while maintaining an almost 1mm2 field-of-view. This approach provides efficient reconstructions of 3D objects using a closed-form inverse scattering model. We illustrate this technique in simulation and experiment on biological samples.

Angular light scattering provides a promising method for sizing organelle distributions in single cells, but most methods of validating size estimates are ill-suited for single cells. We demonstrate tomography-based size estimates as a simultaneous, independent, non-destructive comparison. Optical Diffraction Tomography (ODT) builds a 3D refractive index map which is segmented to provide an estimate of organelle sizes. The normal incidence complex field measurement obtained as part of ODT is Fourier transformed to compute the angular scattering. Comparing both approaches will provide important insight into the capabilities of using angular light scattering to estimate organelle size distributions in single cells.

Holographically Informed Fluorescent Imaging (HIFI) is used to generate 3D fluorescent images. A hologram is acquired via Phase Shifted Digital Holography (PSDH), followed by a fluorescent image. The hologram is numerically propagated, generating a 3D distribution. The optimal depth of focus for localized regions are determined using localized variance in the holographic 3D space. The fluorescent image is correlated with the hologram brought into a single plane and is propagated back to the localized depth of focus, generating a 3D fluorescence image. As a proof of principle, fluorescent labeled polystyrene microspheres are used to generate 3D fluorescent images via HIFI.

We demonstrate a miniaturized mask-based microscope for implantable imaging applications with computational 3D reconstruction. The miniaturized microscope is composed of a custom-designed doublet microlens array which was fabricated by two-photon polymerization. The device is integrated with excitation light sources, fluorescent filter sets, and a board-level image sensor. The 3D object can be reconstructed by trained deep learning models with high efficiency and fast speed. We validate the device’s performance with both resolution targets and fluorescent samples. Our device shows great promise in imaging brain activity in freely moving mice with real-time reconstruction.

We introduce the GRIN-axicon, a new low-cost optical component that is easy to manufacture and could replace the axicon in various setups such as a two-photon microscope. In neuroscience, the imaging of in vivo samples requires high temporal resolution in order to capture the interactions between neurons located at different depths in the tissue. To achieve this, the use of an axicon lens increases the depth of field of the microscope and reduces the number of scans to be performed. However, the axicon is difficult to manufacture and generally has defects on the tip of the cone, thus degrading the quality of the resultant Bessel-Gauss beam.

We present a novel optical microscope technique to detect the chromatin organization and DNA damage foci in the unlabeled live cell nuclei. The dynamic scattering signal of chromatin is recorded by a highly sensitive interference microscopy, contrast-enhanced coherent brightfield microscopy (COBRI), at a high speed. By analyzing the temporal fluctuations of the scattering signal, the density and the compaction level of chromatin are accurately estimated with the sub-micrometer spatial resolutions. This imaging strategy is referred to as “DYNAMICS imaging”. Using DYNAMICS imaging, we monitor the dynamic chromatin remodeling initiated by DNA damage.

Comprehensive evaluation of microvascular function under normal and pathological conditions requires high-resolution three-dimensional microangiography capable of providing both morphological and functional information. Herein, we propose the stereovision Diffuse Optical Localization imaging (sDOLI) approach to attain transcranial volumetric brain microangiography through triangulation and stereo-matching of images collected with two short-wave infrared cameras. The spatio-temporal sparsity of flowing microparticles allows their precise localization while minimizing structural overlaps occurring in the dual-view projections. sDOLI is shown to preserve high spatial resolution which enables transcranial mapping of murine cortical microcirculation at capillary resolution while retrieving quantitative functional information across the entire mouse cortex.

This conference presentation was prepared for the Three-Dimensional and Multidimensional Microscopy: Image Acquisition and Processing XXX conference at SPIE BiOS, 2023.

This conference presentation was prepared for the Three-Dimensional and Multidimensional Microscopy: Image Acquisition and Processing XXX conference at SPIE BiOS, 2023.