We introduce Chromatix: an easy to use, open-source, differentiable wave optics simulation library. Engineered to fully exploit parallelism, from single CPU and GPU workstations to servers with multiple GPUs, Chromatix removes the computational scaling barrier for differentiable wave optics simulations. Chromatix allows for designing and optimizing a wide range of optical systems (e.g., tomography, light field microscopy, and ptychography) as well as solving inverse problems. We expect Chromatix to democratize and power the exploration of a rich design space in computational optics.
We introduce LightFlow, an open-source software package for simulating light wave propagation through custom optical components and systems. Built upon TensorFlow and Keras, it benefits from GPU acceleration and offers a user-friendly and modular architecture. Optical components are represented as layers, simplifying the design and modification of simulation models. Our approach also streamlines the addition of new custom components. LightFlow’s automatic gradient calculation is valuable for computational imaging applications involving optimization algorithms and inverse problems. With its intuitive interface, tested building blocks, and expandable design, LightFlow is well-suited for education and research, from undergraduate to advanced graduate levels. The GPU-accelerated processing enables efficient, real-time visualization of optical simulations, making LightFlow valuable across a broad range of user expertise and applications.
Genetically encoded calcium indicators and optogenetics have revolutionized neuroscience by enabling the detection and modulation of neural activity with single-cell precision using light. To fully leverage the immense potential of these techniques, advanced optical instruments that can place a light on custom ensembles of neurons with a high level of spatial and temporal precision are required. Modern light sculpting techniques that have the capacity to shape a beam of light are preferred because they can precisely target multiple neurons simultaneously and modulate the activity of large ensembles of individual neurons at rates that match natural neuronal dynamics. The most versatile approach, computer-generated holography (CGH), relies on a computer-controlled light modulator placed in the path of a coherent laser beam to synthesize custom three-dimensional (3D) illumination patterns and illuminate neural ensembles on demand. Here, we review recent progress in the development and implementation of fast and spatiotemporally precise CGH techniques that sculpt light in 3D to optically interrogate neural circuit functions.
Gradient descent is an efficient algorithm to optimize differentiable functions with continuous variables, yet it is not suitable for computer generated holography (CGH) with binary light modulators. To address this, we replaced binary pixel values with continuous variables that are binarized with a thresholding operation, and we introduced gradients of the sigmoid function as surrogate gradients to ensure the differentiability of the binarization step. We implemented this method both to directly optimize binary holograms, and to train deep learning-based CGH models. Simulations and experimental results show that our method achieves greater speed, and higher accuracy and contrast than existing algorithms.
Sculpting light in 3D at high speeds is critical to track and manipulate biological events at cellular scales in real-time. Yet, even the current state of the art, Computer-Generated Holography (CGH), operates with slow algorithms and does not provide enough degrees of control to focus light precisely through deep biological tissue. We address these two challenges with new hardware and algorithms. First, we introduce DeepCGH, a deep learning model that synthesizes 3D holograms in milliseconds, and we demonstrate experimental benefits in multiphoton microscopy applications. We then present new optical instruments that modulate light both in space and time to render 4D light fields with much greater fidelity than coherent CGH techniques.
KEYWORDS: 3D scanning, Micromirrors, Mirrors, Monte Carlo methods, Lenses, Optical scanning, Microelectromechanical systems, Actuators, Diffraction, 3D acquisition
Rapid 3D optical scanning of points or patterned light is widely employed across applications in microscopy, material processing, adaptive optics and surveying. Despite this broadness in applicability, embodiments of 3D scanning tools may vary considerably as a result of the specific performance needs of each application. We present here a micromirror arraybased modular framework for the systemic design of such high-speed scanning tools. Our framework combines a semicustom commercial fabrication process with a comprehensive simulation pipeline in order to optimally reconfigure pixel wiring schemes across specific applications for the efficient allocation of available degrees of freedom. As a demonstration of this framework and to address existing bottlenecks in axial focusing, we produced a 32-ring concentric micromirror array capable of performing random-access focusing for wavelengths of up to 1040 nm at a response rate of 8.75 kHz. By partitioning the rings into electrostatically driven piston-mode pixels, we are able to operate the array through simple openloop 30 V drive, minimizing insertion complexity, and to ensure stable operation by preventing torsional failure and curling from stress. Furthermore, by taking advantage of phase-wrapping and the 32 degrees of freedom afforded by the number of independently addressable rings, we achieve good axial resolvability across the tool’s operating range with an axial fullwidth- half-maximum to range ratio of 3.5% as well as the ability to address focus depth-dependent aberrations resulting from the optical system or sample under study.
With signatures of high photon energy and short wavelength, ultraviolet (UV) light enables numerous applications such as high-resolution imaging, photolithography and sensing. In order to manipulate UV light, bulky optics are usually required and thereby do not meet the fast-growing requirements of integration in compact systems. Recently, metasurfaces, with subwavelength or wavelength thicknesses, have shown unprecedented control of light, enabling substantial miniaturization of photonic devices from Terahertz to visible regions. However, material limitations and fabrication challenges have hampered the realization of such functionalities at shorter wavelengths. Herein, we theoretically and experimentally demonstrate that metasurfaces, made of highly scattering silicon (Si) antennas, can be designed and fabricated to manipulate broadband UV light. The metasurface thickness is only one-tenth of the working wavelength, resulting in very small height-to-width aspect ratio (~ 1). Peak conversion efficiency reaches 15% and diffraction efficiency is up to 30%, which are comparable to plasmonic metasurface performances in infrared (IR). A double bar structure is proposed to further improve the metasurface’s diffraction efficiency to close to 100% in transmission mode over a broad UV band. Moreover, for the first time, we show photolithography enabled by metasurface-generated UV holograms. We attribute such performance enhancement to the high scattering cross-sections of Si antennas in the UV range, which is adequately modeled via a circuit. Our new platform will deepen our understanding of light-matter interactions and introduce even more material options to broadband metaphotonic applications, including those in integrated photonics and holographic lithography technologies.
We have developed a microfluidic device that enables the computation of three-dimensional (3-D) images of flowing samples. Using a microfluidic channel that is tilted along the optical axis, we record several progressively defocused images of the flowing sample as it passes across the focal plane. The resulting focal stack is then deconvolved to generate 3-D images. Experimental results on flowing yeast cells reveal both volume and surface profile information. The microfluidic channel eliminates the need for a precise translation stage to control defocusing and enables high sample throughput in an insulated, nontoxic, liquid environment. The experimental device can be implemented in all existing microscopes as a modified slide stage and is ideally suited for 3-D profiling in flow cytometers.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.