Search

Optical coherence tomography (OCT) is a cross-sectional, micrometre-scale imaging modality with widespread clinical application. Typical OCT systems sacrifice lateral resolution to achieve long depths of focus for bulk tissue imaging, and therefore tend to have better axial than lateral resolution. Such anisotropic resolution can obscure fine ultrastructural features. Furthermore, conventional OCT suffers from refraction-induced image distortions. Here, we introduce optical coherence refraction tomography (OCRT), which extends the superior axial resolution to the lateral dimension, synthesizing undistorted cross-sectional image reconstructions from multiple conventional images acquired with angular diversity. In correcting refraction-induced distortions to register the OCT images, OCRT also achieves spatially resolved refractive index imaging. We demonstrate greater than threefold improvement in lateral resolution as well as speckle reduction in imaging the tissue ultrastructure, consistent with histology. With further optimization in optical designs to incorporate angular diversity into clinical instruments, OCRT could be widely applied as an enhancement over conventional OCT. By synthesizing undistorted cross-sectional image reconstructions from multiple conventional images acquired with angular diversity, optical coherence refraction tomography offers greater than threefold improvement in lateral resolution and speckle reduction in imaging tissue ultrastructure, and reconstructs the tissue’s internal refractive index distribution.

This work demonstrates a multi-lens microscopic imaging system that overlaps multiple independent fields of view on a single sensor for high-efficiency automated specimen analysis. Automatic detection, classification and counting of various morphological features of interest is now a crucial component of both biomedical research and disease diagnosis. While convolutional neural networks (CNNs) have dramatically improved the accuracy of counting cells and sub-cellular features from acquired digital image data, the overall throughput is still typically hindered by the limited space-bandwidth product (SBP) of conventional microscopes. Here, we show both in simulation and experiment that overlapped imaging and co-designed analysis software can achieve accurate detection of diagnostically-relevant features for several applications, including counting of white blood cells and the malaria parasite, leading to multi-fold increase in detection and processing throughput with minimal reduction in accuracy.

Microscopic imaging of anisotropic samples has many important applications in cytopathology. The endogenous contrast from the polarization properties of a specimen, such as its birefringence, provides valuable diagnostic information for several deadly diseases, including cardiac amyloidosis and squamous cell carcinoma, for example. In the past, polarized light microscopy (PLM) has been widely used as a diagnostic tool during the clinical review. However, in analogy with the standard microscope, the PLM typically has a restricted spatial-bandwidth product (SBP). As a consequence, one can either image a large area with low resolution or see the details of a very small area of the sample at the resolutions required for accurate analysis. To address the SBP issue of the PLM, we propose a computational microscopy method, termed vectorial Fourier ptychography, to illuminate the specimen with polarized light from different angles and detects different polarization states of the diffracted light. By illuminating a specimen with plane waves from different angles, our vectorial Fourier ptychography method effectively modulates the high-spatial-frequency components of the specimen into lower frequencies that can be detected by the optical system. With a Jones calculus-based forward model and a second-order phase retrieval method, we can reconstruct high-resolution, wide field-of-view(FOV) amplitude, phase, birefringence, retardance, and diattenuation of the specimen. To assess the reconstruction accuracy of our method, we imaged polystyrene beads submerged in immersion oils of different refractive index, as well as monosodium urate crystals. Further, To validate the diattenuation reconstruction accuracy, we reconstruct a USAF resolution test chart with a half blocked by a linear polarizer. These experiments confirm quantitatively accurate reconstruction results with a 1.25 um full-pitch resolution over a FOV of 6.6 x 4.4 mm^2, which is 5 times higher than the native (brightfield) resolution of the non-computational optical system. Finally, we demonstrate our technique by producing high SBP polarization images of several anisotropic biologic samples, includes collagen tissue, congo red stained cardiac tissue, and a bean root sample.

We present a feature-free photogrammetric technique that enables quantitative 3D mesoscopic (mm-scale height variation) imaging with tens-of-micron accuracy from sequences of images acquired by a smartphone at close range (several cm) under freehand motion without additional hardware. Our end-to-end, pixel-intensity-based approach jointly registers and stitches all the images by estimating a coaligned height map, which acts as a pixel-wise radial deformation field that orthorectifies each camera image to allow plane-plus-parallax registration. The height maps themselves are reparameterized as the output of an untrained encoder-decoder convolutional neural network (CNN) with the raw camera images as the input, which effectively removes many reconstruction artifacts. Our method also jointly estimates both the camera’s dynamic 6D pose and its distortion using a nonparametric model, the latter of which is especially important in mesoscopic applications when using cameras not designed for imaging at short working distances, such as smartphone cameras. We also propose strategies for reducing computation time and memory, applicable to other multi-frame registration problems. Finally, we demonstrate our method using sequences of multi-megapixel images captured by an un-stabilized smartphone on a variety of samples (e.g., painting brushstrokes, circuit board, seeds).

We present a gigapixel-scale multi-aperture microscope capable of measuring a sample’s 3D height profile over multi-centimeter-scale fields of view with a series of single synchronized camera snapshots. Exploiting the overlap redundancy in our multi-aperture camera array microscope, we developed a novel, end-to-end photogrammetric reconstruction algorithm that simultaneously calibrates the cameras’ 3D positions and poses, stitches the acquired images, and generates a coregistered, pixel-wise 3D height map of the sample. Our work opens the door to video-rate 3D monitoring of dynamic scenes at micrometer-scale resolutions and centimeter-scale fields of view.

We present 3D optical coherence refraction tomography (OCRT), which incoherently synthesizes an extended 3D transfer function using multi-view OCT volumes spanning two angular dimensions. Our setup features a novel use of parabolic mirrors, achieving millimetric fields of view over ±75 ◦ without rotation stages.

We present a general theory of optical coherence tomography (OCT), which synthesizes the fundamental concepts and implementations of OCT under a common 3D k -space framework. At the heart of this analysis is the Fourier diffraction theorem, which relates the coherent interaction between a sample and plane wave to the Ewald sphere in the 3D k -space representation of the sample. While only the axial dimension of OCT is typically analyzed in k -space, we show that embracing a fully 3D k -space formalism allows explanation of nearly every fundamental physical phenomenon or property of OCT, including contrast mechanism, resolution, dispersion, aberration, limited depth of focus, and speckle. The theory also unifies diffraction tomography, confocal microscopy, point-scanning OCT, line-field OCT, full-field OCT, Bessel beam OCT, transillumination OCT, interferometric synthetic aperture microscopy (ISAM), and optical coherence refraction tomography (OCRT), among others. Our unified theory not only enables clear understanding of existing techniques but also suggests new research directions to continue advancing the field of OCT.

We present a tomographic imaging technique, termed Deep Prior Diffraction Tomography (DP-DT), to reconstruct the 3D refractive index (RI) of thick biological samples at high resolution from a sequence of low-resolution images collected under angularly varying illumination. DP-DT processes the multi-angle data using a phase retrieval algorithm that is extended by a deep image prior (DIP), which reparameterizes the 3D sample reconstruction with an untrained, deep generative 3D convolutional neural network (CNN). We show that DP-DT effectively addresses the missing cone problem, which otherwise degrades the resolution and quality of standard 3D reconstruction algorithms. As DP-DT does not require pre-captured data or pre-training, it is not biased towards any particular dataset. Hence, it is a general technique that can be applied to a wide variety of 3D samples, including scenarios in which large datasets for supervised training would be infeasible or expensive. We applied DP-DT to obtain 3D RI maps of bead phantoms and complex biological specimens, both in simulation and experiment, and show that DP-DT produces higher-quality results than standard regularization techniques. We further demonstrate the generality of DP-DT, using two different scattering models, the first Born and multi-slice models. Our results point to the potential benefits of DP-DT for other 3D imaging modalities, including X-ray computed tomography, magnetic resonance imaging, and electron microscopy.

Traditional imaging systems exhibit a well-known trade-off between the resolution and the field of view of their captured images. Typical cameras and microscopes can either “zoom in” and image at high-resolution, or they can “zoom out” to see a larger area at lower resolution, but can rarely achieve both effects simultaneously. In this review, we present details about a relatively new procedure termed Fourier ptychography (FP), which addresses the above trade-off to produce gigapixel-scale images without requiring any moving parts. To accomplish this, FP captures multiple low-resolution, large field-of-view images and computationally combines them in the Fourier domain into a high-resolution, large field-of-view result. Here, we present details about the various implementations of FP and highlight its demonstrated advantages to date, such as aberration recovery, phase imaging, and 3D tomographic reconstruction, to name a few. After providing some basics about FP, we list important details for successful experimental implementation, discuss its relationship with other computational imaging techniques, and point to the latest advances in the field while highlighting persisting challenges.

We present a preliminary k-space-based analysis of OCT and our recently proposed extension, optical coherence refraction tomography (OCRT). Our theory provides insight into how angular compounding enhances resolution and reduces speckle in coherence imaging techniques.

Illuminating or imaging samples from a broad angular range is essential in a wide variety of computational 3D imaging and resolution-enhancement techniques, such as optical projection tomography, optical diffraction tomography, synthetic aperture microscopy, Fourier ptychographic microscopy, structured illumination microscopy, photogrammetry, and optical coherence refraction tomography. The wider the angular coverage, the better the resolution enhancement or 3D-resolving capabilities. However, achieving such angular ranges is a practical challenge, especially when approaching ± 90 ∘ or beyond. Often, researchers resort to expensive, proprietary high numerical aperture (NA) objectives or to rotating the sample or source-detector pair, which sacrifices temporal resolution or perturbs the sample. Here, we propose several new strategies for multiangle imaging approaching 4pi steradians using concave parabolic or ellipsoidal mirrors and fast, low rotational inertia scanners, such as galvanometers. We derive theoretically and empirically relations between a variety of system parameters (e.g., NA, wavelength, focal length, telecentricity) and achievable fields of view (FOVs) and importantly show that intrinsic tilt aberrations do not restrict FOV for many multiview imaging applications, contrary to conventional wisdom. Finally, we present strategies for avoiding spherical aberrations at obliquely illuminated flat boundaries. Our simple designs allow for high-speed multiangle imaging for microscopic, mesoscopic, and macroscopic applications.

Combining principles of computed tomography with modern machine-learning tools significantly improves OCT’s resolution while extending imaging depth, reducing noise and reconstructing refractive-index maps of biological samples.

Since its invention, the microscope has been optimized for interpretation by a human observer. With the recent development of deep learning algorithms for automated image analysis, there is now a clear need to re-design the microscope's hardware for specific interpretation tasks. To increase the speed and accuracy of automated image classification, this work presents a method to co-optimize how a sample is illuminated in a microscope, along with a pipeline to automatically classify the resulting image, using a deep neural network. By adding a ""physical layer"" to a deep classification network, we are able to jointly optimize for specific illumination patterns that highlight the most important sample features for the particular learning task at hand, which may not be obvious under standard illumination. We demonstrate how our learned sensing approach for illumination design can automatically identify malaria-infected cells with up to 5-10% greater accuracy than standard and alternative microscope lighting designs. We show that this joint hardware-software design procedure generalizes to offer accurate diagnoses for two different blood smear types, and experimentally show how our new procedure can translate across different experimental setups while maintaining high accuracy.

In optical coherence tomography (OCT), the axial resolution is often superior to the lateral resolution, which is sacrificed for long imaging depths. To address this anisotropy, we previously developed optical coherence refraction tomography (OCRT), which uses images from multiple angles to computationally reconstruct an image with isotropic resolution, given by the OCT axial resolution. On the other hand, spectroscopic OCT (SOCT), an extension of OCT, trades axial resolution for spectral resolution and hence often has superior lateral resolution. Here, we present spectroscopic OCRT (SOCRT), which uses SOCT images from multiple angles to reconstruct a spectroscopic image with isotropic spatial resolution limited by the OCT lateral resolution. We experimentally show that SOCRT can estimate bead size based on Mie theory at simultaneously high spectral and isotropic spatial resolution. We also applied SOCRT to a biological sample, achieving axial resolution enhancement limited by the lateral resolution.

We integrate spectroscopic optical coherence tomography (SOCT) with stimulated Raman scattering (SRS) to enable simultaneously multiplexed spatial and spectral imaging with sensitivity to many endogenous biochemical species that play an important role in biology and medicine. The combined approach, termed SRS-SOCT, overcomes the limitations of each individual method. Ultimately, SRS-SOCT has the potential to achieve fast, volumetric, and highly sensitive label-free molecular imaging. We demonstrate the approach by imaging excised human adipose tissue and detecting the lipids' Raman signatures in the high-wavenumber region.

Optical coherence tomography (OCT) enables non-invasive, high-resolution, tomographic imaging of biological tissues by leveraging principles of low coherence interferometry; however, OCT lacks molecular specificity. Spectroscopic OCT (SOCT) overcomes this limitation by providing depth-resolved spectroscopic signatures of chromophores, but SOCT has been limited to a couple of endogenous molecules, namely hemoglobin and melanin. Stimulated Raman scattering, on the other hand, can provide highly specific molecular information of many endogenous species, but lacks the spatial and spectral multiplexing capabilities of SOCT. In this work we integrate the two methods, SRS and SOCT, to enable simultaneously multiplexed spatial and spectral imaging with sensitivity to many endogenous biochemical species that play an important role in biology and medicine. The method, termed SRS-SOCT, has the potential to achieve fast, volumetric, and highly sensitive label-free molecular imaging, which would be valuable for many applications. We demonstrate the approach by imaging excised human adipose tissue and detecting the lipids’ Raman signatures in the high-wavenumber region. Details of this method along with validations and results will be presented.

One critical barrier to the robust study of cilia-driven fluid flow in developmental biology is a lack of methods for acquiring three-dimensional (3D) images of three vector component (3C) measurements of flow velocities. A 3D3C map of cilia-driven fluid flow quantifies the flow speed along three axes (e.g. three Cartesian vector components v_x, v_y, v_z) at each point in 3D space. 3D3C quantification is important because cilia-driven fluid flow is not amenable to simplifying assumptions (e.g. parabolic flow profile. Such quantification may enable systematically detailed characterization of performance using shear force and power dissipation metrics derived from 3D3C flow velocity fields. We report our OCT-based results in developing methods for the 3D3C quantification of cilia-driven flow fields. First, we used custom scan protocols and reconstruction algorithms to synthesize 3D3C flow velocity fields from 2D2C fields generated using correlation-based methods (directional dynamic light scattering and digital particle image velocimetry). Xenopus results include flow driven by ciliated embryo skin and flow driven by ciliated ependymal cells in developing brain ventricles. Second, we developed a new approach to particle tracking velocimetry that generates 2D2.5C (2.5C: v_x,|v_y|,v_z) velocity fields from single-plane 2D image acquisitions. We demonstrated this particle streak velocimetry method in calibrated flow phantoms and in flow driven by ciliated Xenopus embryo skin. Additionally, we have preliminary results extending particle streak velocimetry to 3D3C in calibrated flow phantoms with ongoing work in Xenopus embryos.

We present a new method for 2.5 and 3 vector component velocimetry. We call this method particle streak velocimetry OCT (PSV-OCT). PSV-OCT generates two-dimensional, 2.5 vector component (v_x,|v_y|,v_z) cross-sectional maps of microscale flow velocity (e.g. biological cilia-driven fluid flow). The enabling insight is that a tracer particle in sparsely-seeded fluid flow traces out streaks in (x,z,t)-space. The streak orientations in x-t and z-t yield v_x and v_z, respectively. The in-plane (x-z plane) residence time yields the out-of-plane speed |v_y|. Vector component values are generated by fitting streaks to a model of image formation. We demonstrate cross-sectional estimation of (v_x,|v_y|,v_z) in two important animal models in ciliary biology: Xenopus embryos (tadpoles) and mouse trachea. Further, by incorporation the assumption of incompressible flow into the estimation process, we are able to generate 3 vector component (v_x,v_y,v_z) estimates in three spatial dimensions from 2.5 vector component measurements taken in parallel OCT planes in 3D space.

We present a new OCT method for flow speed quantification and directional velocimetry: particle streak velocimetry-OCT (PSV-OCT). PSV-OCT generates two-dimensional, 2.5-vector component (vx ,|vy |,vz ) maps of microscale flow velocity fields. Knowledge of 2.5-vector components also enables the estimation of total flow speed. The enabling insight behind PSV-OCT is that tracer particles in sparsely-seeded fluid flow trace out streaks in (x,z,t)-space. The streak orientations in x-t and z-t yield vx and vz , respectively. The in-plane (x-z plane) residence time yields the out-of-plane speed |vy |. Vector component values are generated by fitting streaks to a model of image formation that incorporates equations of motion in 3D space. We demonstrate cross-sectional estimation of (vx ,|vy |,vz ) in two important animal models in ciliary biology: Xenopus embryos (tadpoles) and mouse trachea.

OCT is a popular cross-sectional microscale imaging modality in medicine and biology. While structural imaging using OCT is a mature technology in many respects, flow and motion estimation using OCT remains an intense area of research. In particular, there is keen interest in maximizing information extraction from the complex-valued OCT signal. Here, we introduce a Bayesian framework into the data workflow in OCT-based velocimetry. We demonstrate that using prior information in this Bayesian framework can significantly improve velocity estimate precision in a correlation-based, model-based framework for Doppler and transverse velocimetry. We show results in calibrated flow phantoms as well as in vivo in a Drosophila melanogaster (fruit fly) heart. Thus, our work improves upon the current approaches in terms of improved information extraction from the complex-valued OCT signal.

Researchers have long been interested in the evolution of culture and the ways in which change in cultural systems can be reconstructed and tracked. Within the realm of language, these questions are increasingly investigated with Bayesian phylogenetic methods. However, such work in cultural phylogenetics could be improved by more explicit quantification of reconstruction and transition probabilities. We apply such methods to numerals in the languages of Australia. As a large phylogeny with almost universal ‘low-limit' systems, Australian languages are ideal for investigating numeral change over time. We reconstruct the most likely extent of the system at the root and use that information to explore the ways numerals evolve. We show that these systems do not increment serially, but most commonly vary their upper limits between 3 and 5. While there is evidence for rapid system elaboration beyond the lower limits, languages lose numerals as well as gain them. We investigate the ways larger numerals build on smaller bases, and show that there is a general tendency to both gain and replace 4 by combining 2 + 2 (rather than inventing a new unanalysable word ‘four'). We develop a series of methods for quantifying and visualizing the results.