Deep learning has been transforming our ability to execute advanced inference tasks using computers. Here we introduce a physical mechanism to perform machine learning by demonstrating an all-optical diffractive deep neural network (D2NN) architecture that can implement various functions following the deep learning-based design of passive diffractive layers that work collectively. We created 3D-printed D2NNs that implement classification of images of handwritten digits and fashion products, as well as the function of an imaging lens at a terahertz spectrum. Our all-optical deep learning framework can perform, at the speed of light, various complex functions that computer-based neural networks can execute; will find applications in all-optical image analysis, feature detection, and object classification; and will also enable new camera designs and optical components that perform distinctive tasks using D2NNs.

/pdf/all-optical-machine-learning-using-diffractive-deep-neural-1o1xdg7mzl.pdf

All-optical machine learning using diffractive deep neural networks

Quantitative phase imaging (QPI) has emerged as a valuable method for investigating cells and tissues. QPI operates on unlabelled specimens and, as such, is complementary to established fluorescence microscopy, exhibiting lower phototoxicity and no photobleaching. As the images represent quantitative maps of optical path length delays introduced by the specimen, QPI provides an objective measure of morphology and dynamics, free of variability due to contrast agents. Owing to the tremendous progress witnessed especially in the past 10–15 years, a number of technologies have become sufficiently reliable and translated to biomedical laboratories. Commercialization efforts are under way and, as a result, the QPI field is now transitioning from a technology-development-driven to an application-focused field. In this Review, we aim to provide a critical and objective overview of this dynamic research field by presenting the scientific context, main principles of operation and current biomedical applications. Over the past 10–15 years, quantitative phase imaging has moved from a research-driven to an application-focused field. This Review presents the main principles of operation and representative basic and clinical science applications.

Quantitative phase imaging in biomedicine

Random walk is a fundamental concept with applications ranging from quantum physics to econometrics. Remarkably, one specific model of random walks appears to be ubiquitous across many fields as a tool to analyze transport phenomena in which the dispersal process is faster than dictated by Brownian diffusion. The Levy walk model combines two key features, the ability to generate anomalously fast diffusion and a finite velocity of a random walker. Recent results in optics, Hamiltonian chaos, cold atom dynamics, bio-physics, and behavioral science demonstrate that this particular type of random walks provides significant insight into complex transport phenomena. This review provides a self-consistent introduction to Levy walks, surveys their existing applications, including latest advances, and outlines further perspectives.

L\'evy walks

Optical imaging of nanoscale objects, whether it is based on scattering or fluorescence, is a challenging task due to reduced detection signal-to-noise ratio and contrast at subwavelength dimensions. Here, we report a field-portable fluorescence microscopy platform installed on a smart phone for imaging of individual nanoparticles as well as viruses using a lightweight and compact opto-mechanical attachment to the existing camera module of the cell phone. This hand-held fluorescent imaging device utilizes (i) a compact 450 nm laser diode that creates oblique excitation on the sample plane with an incidence angle of ∼75°, (ii) a long-pass thin-film interference filter to reject the scattered excitation light, (iii) an external lens creating 2× optical magnification, and (iv) a translation stage for focus adjustment. We tested the imaging performance of this smart-phone-enabled microscopy platform by detecting isolated 100 nm fluorescent particles as well as individual human cytomegaloviruses that are fluor...

/pdf/fluorescent-imaging-of-single-nanoparticles-and-viruses-on-a-11fx06bejn.pdf

Fluorescent Imaging of Single Nanoparticles and Viruses on a Smart Phone

Computer-assisted sperm analysis (CASA): capabilities and potential developments.

In this perspective, the authors present the basic features of lens-free computational imaging tools and report performance comparisons with conventional microscopy methods. They also discuss the challenges that these computational on-chip microscopes face for their wide-scale biomedical application.

/pdf/imaging-without-lenses-achievements-and-remaining-challenges-1tjjxwd1cx.pdf

Imaging without lenses: achievements and remaining challenges of wide-field on-chip microscopy

Dynamic tracking of human sperms across a large volume is a challenging task. To provide a high-throughput solution to this important need, here we describe a lensfree on-chip imaging technique that can track the three-dimensional (3D) trajectories of > 1,500 individual human sperms within an observation volume of approximately 8–17 mm3. This computational imaging platform relies on holographic lensfree shadows of sperms that are simultaneously acquired at two different wavelengths, emanating from two partially-coherent sources that are placed at 45° with respect to each other. This multiangle and multicolor illumination scheme permits us to dynamically track the 3D motion of human sperms across a field-of-view of > 17 mm2 and depth-of-field of approximately 0.5–1 mm with submicron positioning accuracy. The large statistics provided by this lensfree imaging platform revealed that only approximately 4–5% of the motile human sperms swim along well-defined helices and that this percentage can be significantly suppressed under seminal plasma. Furthermore, among these observed helical human sperms, a significant majority (approximately 90%) preferred right-handed helices over left-handed ones, with a helix radius of approximately 0.5–3 μm, a helical rotation speed of approximately 3–20 rotations/s and a linear speed of approximately 20–100 μm/s. This high-throughput 3D imaging platform could in general be quite valuable for observing the statistical swimming patterns of various other microorganisms, leading to new insights in their 3D motion and the underlying biophysics.

https://www.pnas.org/content/pnas/early/2012/09/10/1212506109.full.pdf

High-throughput lensfree 3D tracking of human sperms reveals rare statistics of helical trajectories

A lensfree holographic imaging platform which can track the 3D motion of >1,500 human sperms within an observation volume of 8-17 mm^3 is developed to reveal the statistics of the rare (~4.4%) and mostly right-handed (~90%) helical trajectories.

Liang Xue

Papers

Imaging without lenses: achievements and remaining challenges of wide-field on-chip microscopy

High-throughput lensfree 3D tracking of human sperms reveals rare statistics of helical trajectories

Lensfree 3D Tracking of Sperms at High-throughput