2021 Vol. 10, No. 9

Light Peoples
Light People: Tarik Bourouina
Hui Wang, Xiaoyi Liu
Published. 2021, 10(9) : 1560-1564 doi: 10.1038/s41377-021-00602-w
In 2015, 195 countries of the United Nations proposed Sustainable Development Goals so as to alleviate the problems of climate change and global pollution. In France, there is a scientist dedicated to contribute providing solutions for above issues by virtue of MEMS, Lab-On-Chip and metamaterials. This expert is Prof. Tarik Bourouina, a Professor of Physics at ESIEE Paris, Université Gustave Eiffel. He devoted himself to the investigations on micro sensors and metamaterials, and kept seeking their applications in the future blueprint of "Sustainable" and "Smart" cities. On the other hand, he has formed an indissoluble bond with Light: Science and Applications (LIGHT) since the very beginning of the journal. He also set up the LIGHT's Paris office, which is the first LIGHT's overseas office in Europe. We are much honored to have an opportunity to exclusively communicate with Prof. Tarik Bourouina, who will share his research experience and stories with LIGHT in this interview.
Light People: Professor James C. Wyant
Hui Wang, Cun Yu
Published. 2021, 10(9) : 1565-1570 doi: 10.1038/s41377-021-00582-x
He is the founding dean of the University of Arizona's College of Optical Sciences. He was elected president of the OSA and the SPIE, two of the most respected international academic organizations in the world of optics. He is the founder or co-founder of multiple hi-tech companies. He is Prof. James C. Wyant. Professor Wyant is many things: master of optics, entrepreneur, and philanthropist, but above all, he is one of those rare lucky people who see their childhood dreams come true. He is the hardworking scientist who pushes optical science to the limit. He is the successful entrepreneur whose Midas touch turns research achievements into lucrative commercial projects. He is the generous philanthropist who has given a large fortune to help promote the development of optical science. He is the avid athlete who enjoys the outdoors and takes pleasure in hiking and running. From tender memories to cutting-edge research and hard-nosed business advice, Prof. Wyant will share with us how he became fascinated with optics and what he thinks about academic life and organizations.
Uncovering recent progress in nanostructured light-emitters for information and communication technologies
Frédéric Grillot, Jianan Duan, Bozhang Dong, Heming Huang
Published. 2021, 10(9) : 1571-1587 doi: 10.1038/s41377-021-00598-3
Semiconductor nanostructures with low dimensionality like quantum dots and quantum dashes are one of the best attractive and heuristic solutions for achieving high performance photonic devices. When one or more spatial dimensions of the nanocrystal approach the de Broglie wavelength, nanoscale size effects create a spatial quantization of carriers leading to a complete discretization of energy levels along with additional quantum phenomena like entangled-photon generation or squeezed states of light among others. This article reviews our recent findings and prospects on nanostructure based light emitters where active region is made with quantum-dot and quantum-dash nanostructures. Many applications ranging from silicon-based integrated technologies to quantum information systems rely on the utilization of such laser sources. Here, we link the material and fundamental properties with the device physics. For this purpose, spectral linewidth, polarization anisotropy, optical nonlinearities as well as microwave, dynamic and nonlinear properties are closely examined. The paper focuses on photonic devices grown on native substrates (InP and GaAs) as well as those heterogeneously and epitaxially grown on silicon substrate. This research pipelines the most exciting recent innovation developed around light emitters using nanostructures as gain media and highlights the importance of nanotechnologies on industry and society especially for shaping the future information and communication society.
Advances of surface-enhanced Raman and IR spectroscopies: from nano/microstructures to macro-optical design
Hai-Long Wang, En-Ming You, Rajapandiyan Panneerselvam, Song-Yuan Ding, Zhong-Qun Tian
Published. 2021, 10(9) : 1588-1606 doi: 10.1038/s41377-021-00599-2
Raman and infrared (IR) spectroscopy are powerful analytical techniques, but have intrinsically low detection sensitivity. There have been three major steps (i) to advance the optical system of the light excitation, collection, and detection since 1920s, (ii) to utilize nanostructure-based surface-enhanced Raman scattering (SERS) and surface-enhanced infrared absorption (SEIRA) since 1990s, and (iii) to rationally couple (i) and (ii) for maximizing the total detection sensitivity since 2010s. After surveying the history of SERS and SEIRA, we outline the principle of plasmonics and the different mechanisms of SERS and SEIRA. We describe various interactions of light with nano/microstructures, localized surface plasmon, surface plasmon polariton, and lightning-rod effect. Their coupling effects can significantly increase the surface sensitivity by designing nanoparticle–nanoparticle and nanoparticle–substrate configuration. As the nano/microstructures have specific optical near-field and far-field behaviors, we focus on how to systematically design the macro-optical systems to maximize the excitation efficiency and detection sensitivity. We enumerate the key optical designs in particular ATR-based operation modes of directional excitation and emission from visible to IR spectral region. We also present some latest advancements on scanning-probe microscopy-based nanoscale spectroscopy. Finally, prospects and further developments of this field are given with emphasis on emerging techniques and methodologies.
Hybrid laser precision engineering of transparent hard materials: challenges, solutions and applications
Huagang Liu, Wenxiong Lin, Minghui Hong
Published. 2021, 10(9) : 1607-1629 doi: 10.1038/s41377-021-00596-5
Laser has been demonstrated to be a mature and versatile tool that presents great flexibility and applicability for the precision engineering of a wide range of materials over other established micromachining techniques. Past decades have witnessed its rapid development and extensive applications ranging from scientific researches to industrial manufacturing. Transparent hard materials remain several major technical challenges for conventional laser processing techniques due to their high hardness, great brittleness, and low optical absorption. A variety of hybrid laser processing technologies, such as laser-induced plasma-assisted ablation, laser-induced backside wet etching, and etching assisted laser micromachining, have been developed to overcome these barriers by introducing additional medium assistance or combining different process steps. This article reviews the basic principles and characteristics of these hybrid technologies. How these technologies are used to precisely process transparent hard materials and their recent advancements are introduced. These hybrid technologies show remarkable benefits in terms of efficiency, accuracy, and quality for the fabrication of microstructures and functional devices on the surface of or inside the transparent hard substrates, thus enabling widespread applications in the fields of microelectronics, bio-medicine, photonics, and microfluidics. A summary and outlook of the hybrid laser technologies are also highlighted.
There is plenty of room at the top: generation of hot charge carriers and their applications in perovskite and other semiconductor-based optoelectronic devices
Irfan Ahmed, Lei Shi, Hannu Pasanen, Paola Vivo, Partha Maity, et al.
Published. 2021, 10(9) : 1630-1657 doi: 10.1038/s41377-021-00609-3
Hot charge carriers (HC) are photoexcited electrons and holes that exist in nonequilibrium high-energy states of photoactive materials. Prolonged cooling time and rapid extraction are the current challenges for the development of future innovative HC-based optoelectronic devices, such as HC solar cells (HCSCs), hot energy transistors (HETs), HC photocatalytic reactors, and lasing devices. Based on a thorough analysis of the basic mechanisms of HC generation, thermalization, and cooling dynamics, this review outlines the various possible strategies to delay the HC cooling as well as to speed up their extraction. Various materials with slow cooling behavior, including perovskites and other semiconductors, are thoroughly presented. In addition, the opportunities for the generation of plasmon-induced HC through surface plasmon resonance and their technological applications in hybrid nanostructures are discussed in detail. By judiciously designing the plasmonic nanostructures, the light coupling into the photoactive layer and its optical absorption can be greatly enhanced as well as the successful conversion of incident photons to HC with tunable energies can also be realized. Finally, the future outlook of HC in optoelectronics is highlighted which will provide great insight to the research community.
Neural network-based image reconstruction in swept-source optical coherence tomography using undersampled spectral data
Yijie Zhang, Tairan Liu, Manmohan Singh, Ege Çetintaş, Yilin Luo, et al.
Published. 2021, 10(9) : 1658-1671 doi: 10.1038/s41377-021-00594-7
Optical coherence tomography (OCT) is a widely used non-invasive biomedical imaging modality that can rapidly provide volumetric images of samples. Here, we present a deep learning-based image reconstruction framework that can generate swept-source OCT (SS-OCT) images using undersampled spectral data, without any spatial aliasing artifacts. This neural network-based image reconstruction does not require any hardware changes to the optical setup and can be easily integrated with existing swept-source or spectral-domain OCT systems to reduce the amount of raw spectral data to be acquired. To show the efficacy of this framework, we trained and blindly tested a deep neural network using mouse embryo samples imaged by an SS-OCT system. Using 2-fold undersampled spectral data (i.e., 640 spectral points per A-line), the trained neural network can blindly reconstruct 512 A-lines in 0.59 ms using multiple graphics-processing units (GPUs), removing spatial aliasing artifacts due to spectral undersampling, also presenting a very good match to the images of the same samples, reconstructed using the full spectral OCT data (i.e., 1280 spectral points per A-line). We also successfully demonstrate that this framework can be further extended to process 3× undersampled spectral data per A-line, with some performance degradation in the reconstructed image quality compared to 2× spectral undersampling. Furthermore, an A-line-optimized undersampling method is presented by jointly optimizing the spectral sampling locations and the corresponding image reconstruction network, which improved the overall imaging performance using less spectral data points per A-line compared to 2× or 3× spectral undersampling results. This deep learning-enabled image reconstruction approach can be broadly used in various forms of spectral-domain OCT systems, helping to increase their imaging speed without sacrificing image resolution and signal-to-noise ratio.
Modeling the optical properties of twisted bilayer photonic crystals
Haoning Tang, Fan Du, Stephen Carr, Clayton DeVault, Olivia Mello, et al.
Published. 2021, 10(9) : 1672-1679 doi: 10.1038/s41377-021-00601-x
We demonstrate a photonic analog of twisted bilayer graphene that has ultra-flat photonic bands and exhibits extreme slow-light behavior. Our twisted bilayer photonic device, which has an operating wavelength in the C-band of the telecom window, uses two crystalline silicon photonic crystal slabs separated by a methyl methacrylate tunneling layer. We numerically determine the magic angle using a finite-element method and the corresponding photonic band structure, which exhibits a flat band over the entire Brillouin zone. This flat band causes the group velocity to approach zero and introduces light localization, which enhances the electromagnetic field at the expense of bandwidth. Using our original plane-wave continuum model, we find that the photonic system has a larger band asymmetry. The band structure can easily be engineered by adjusting the device geometry, giving significant freedom in the design of devices. Our work provides a fundamental understanding of the photonic properties of twisted bilayer photonic crystals and opens the door to the nanoscale-based enhancement of nonlinear effects.
Dual-resonance enhanced quantum light-matter interactions in deterministically coupled quantum-dot-micropillars
Shunfa Liu, Yuming Wei, Xueshi Li, Ying Yu, Jin Liu, et al.
Published. 2021, 10(9) : 1680-1686 doi: 10.1038/s41377-021-00604-8
Optical microcavities have widely been employed to enhance either the optical excitation or the photon emission processes for boosting light-matter interactions at the nanoscale. When both the excitation and emission processes are simultaneously facilitated by the optical resonances provided by the microcavities, as referred to the dual-resonance condition in this article, the performances of many nanophotonic devices approach to the optima. In this work, we present versatile accessing of dual-resonance conditions in deterministically coupled quantum-dot (QD)-micropillars, which enables emission from neutral exciton (X)—charged exciton (CX) transition with improved single-photon purity. In addition, the rarely observed up-converted single-photon emission process is achieved under dual-resonance conditions. We further exploit the vectorial nature of the high-order cavity modes to significantly improve the excitation efficiency under the dual-resonance condition. The dual-resonance enhanced light-matter interactions in the quantum regime provide a viable path for developing integrated quantum photonic devices based on cavity quantum electrodynamics (QED) effect, e.g., highly efficient quantum light sources and quantum logical gates.
Structured 3D linear space–time light bullets by nonlocal nanophotonics
Cheng Guo, Meng Xiao, Meir Orenstein, Shanhui Fan
Published. 2021, 10(9) : 1687-1701 doi: 10.1038/s41377-021-00595-6
We propose the generation of 3D linear light bullets propagating in free space using a single passive nonlocal optical surface. The nonlocal nanophotonics can generate space–time coupling without any need for bulky pulse-shaping and spatial modulation techniques. Our approach provides simultaneous control of various properties of the light bullets, including the external properties such as the group velocity and the propagation distance, and internal degrees of freedom such as the spin angular momentum and the orbital angular momentum.
Focusing light into scattering media with ultrasound-induced field perturbation
Zhongtao Cheng, Lihong V. Wang
Published. 2021, 10(9) : 1702-1713 doi: 10.1038/s41377-021-00605-7
Focusing light into scattering media, although challenging, is highly desirable in many realms. With the invention of time-reversed ultrasonically encoded (TRUE) optical focusing, acousto-optic modulation was demonstrated as a promising guidestar mechanism for achieving noninvasive and addressable optical focusing into scattering media. Here, we report a new ultrasound-assisted technique, ultrasound-induced field perturbation optical focusing, abbreviated as UFP. Unlike in conventional TRUE optical focusing, where only the weak frequency-shifted first-order diffracted photons due to acousto-optic modulation are useful, here UFP leverages the brighter zeroth-order photons diffracted by an ultrasonic guidestar as information carriers to guide optical focusing. We find that the zeroth-order diffracted photons, although not frequency-shifted, do have a field perturbation caused by the existence of the ultrasonic guidestar. By detecting and time-reversing the differential field of the frequency-unshifted photons when the ultrasound is alternately ON and OFF, we can focus light to the position where the field perturbation occurs inside the scattering medium. We demonstrate here that UFP optical focusing has superior performance to conventional TRUE optical focusing, which benefits from the more intense zeroth-order photons. We further show that UFP optical focusing can be easily and flexibly developed into double-shot realization or even single-shot realization, which is desirable for high-speed wavefront shaping. This new method upsets conventional thinking on the utility of an ultrasonic guidestar and broadens the horizon of light control in scattering media. We hope that it provides a more efficient and flexible mechanism for implementing ultrasound-guided wavefront shaping.
High-accuracy source-independent radiometric calibration with low complexity for infrared photonic sensors
Qiang Guo, Fuchun Chen, Xiangyang Li, Boyang Chen, Xin Wang, et al.
Published. 2021, 10(9) : 1714-1725 doi: 10.1038/s41377-021-00597-4
Radiometric calibration (RC) is an essential solution to guarantee measurements from infrared photonic sensors with certain accuracy, the main task of which is to determine the radiometric responsivity of sensor and usually be solved by comparing with some radiation source (i.e., blackbody), called source-based RC (SBRC). In addition to the complexity in manufacture, the nonideal characteristics of an available source will inevitably introduce unexpected uncertainties to reduce the final calibration accuracy by around 0.2–0.5 K in SBRC. Therefore, we propose an original source-independent RC (SIRC) principle based on modeling instead of comparing for SBRC, where the incident background radiation to detector, as a dominated factor influencing the responsivity characteristics of a photonic sensor, is modeled to implement RC for both two fundamental types (photoconductive and photovoltaic) of HgCdTe photonic detectors. The SIRC merely requires the temperature information of main components of a sensor other than some complex source and its assembly, and provides a traceable way at lower uncertainty costs relative to the traditional SBRC. The SIRC is being implemented in Fengyun-2 satellites since 2019, which ensures a long-term stable service of Chinese geostationary meteorological satellites for the global observation system under the framework of World Meteorological Organization. Moreover, a 20-year-period traceable Fengyun-2 dataset to be recalibrated with SIRC will benefit the further climate applications.
Nonlinear control of photonic higher-order topological bound states in the continuum
Zhichan Hu, Domenico Bongiovanni, Dario Jukić, Ema Jajtić, Shiqi Xia, et al.
Published. 2021, 10(9) : 1726-1735 doi: 10.1038/s41377-021-00607-5
Higher-order topological insulators (HOTIs) are recently discovered topological phases, possessing symmetry-protected corner states with fractional charges. An unexpected connection between these states and the seemingly unrelated phenomenon of bound states in the continuum (BICs) was recently unveiled. When nonlinearity is added to the HOTI system, a number of fundamentally important questions arise. For example, how does nonlinearity couple higher-order topological BICs with the rest of the system, including continuum states? In fact, thus far BICs in nonlinear HOTIs have remained unexplored. Here we unveil the interplay of nonlinearity, higher-order topology, and BICs in a photonic platform. We observe topological corner states that are also BICs in a laser-written second-order topological lattice and further demonstrate their nonlinear coupling with edge (but not bulk) modes under the proper action of both self-focusing and defocusing nonlinearities. Theoretically, we calculate the eigenvalue spectrum and analog of the Zak phase in the nonlinear regime, illustrating that a topological BIC can be actively tuned by nonlinearity in such a photonic HOTI. Our studies are applicable to other nonlinear HOTI systems, with promising applications in emerging topology-driven devices.
Laser solid-phase synthesis of single-atom catalysts
Yudong Peng, Jianyun Cao, Yang Sha, Wenji Yang, Lin Li, et al.
Published. 2021, 10(9) : 1736-1749 doi: 10.1038/s41377-021-00603-9
Single-atom catalysts (SACs) with atomically dispersed catalytic sites have shown outstanding catalytic performance in a variety of reactions. However, the development of facile and high-yield techniques for the fabrication of SACs remains challenging. In this paper, we report a laser-induced solid-phase strategy for the synthesis of Pt SACs on graphene support. Simply by rapid laser scanning/irradiation of a freeze-dried electrochemical graphene oxide (EGO) film loaded with chloroplatinic acid (H2PtCl6), we enabled simultaneous pyrolysis of H2PtCl6 into SACs and reduction/graphitization of EGO into graphene. The rapid freezing of EGO hydrogel film infused with H2PtCl6 solution in liquid nitrogen and the subsequent ice sublimation by freeze-drying were essential to achieve the atomically dispersed Pt. Nanosecond pulsed infrared (IR; 1064 nm) and picosecond pulsed ultraviolet (UV; 355 nm) lasers were used to investigate the effects of laser wavelength and pulse duration on the SACs formation mechanism. The atomically dispersed Pt on graphene support exhibited a small overpotential of −42.3 mV at −10 mA cm−2 for hydrogen evolution reaction and a mass activity tenfold higher than that of the commercial Pt/C catalyst. This method is simple, fast and potentially versatile, and scalable for the mass production of SACs.
Quantum verification of NP problems with single photons and linear optics
Aonan Zhang, Hao Zhan, Junjie Liao, Kaimin Zheng, Tao Jiang, et al.
Published. 2021, 10(9) : 1750-1760 doi: 10.1038/s41377-021-00608-4
Quantum computing is seeking to realize hardware-optimized algorithms for application-related computational tasks. NP (nondeterministic-polynomial-time) is a complexity class containing many important but intractable problems like the satisfiability of potentially conflict constraints (SAT). According to the well-founded exponential time hypothesis, verifying an SAT instance of size n requires generally the complete solution in an O(n)-bit proof. In contrast, quantum verification algorithms, which encode the solution into quantum bits rather than classical bit strings, can perform the verification task with quadratically reduced information about the solution in \begin{document}$\tilde O(\sqrt n)$\end{document} qubits. Here we realize the quantum verification machine of SAT with single photons and linear optics. By using tunable optical setups, we efficiently verify satisfiable and unsatisfiable SAT instances and achieve a clear completeness-soundness gap even in the presence of experimental imperfections. The protocol requires only unentangled photons, linear operations on multiple modes and at most two-photon joint measurements. These features make the protocol suitable for photonic realization and scalable to large problem sizes with the advances in high-dimensional quantum information manipulation and large scale linear-optical systems. Our results open an essentially new route toward quantum advantages and extend the computational capability of optical quantum computing.
Double-bowl state in photonic Dirac nodal line semimetal
Mengying Hu, Ye Zhang, Xi Jiang, Tong Qiao, Qiang Wang, et al.
Published. 2021, 10(9) : 1761-1767 doi: 10.1038/s41377-021-00614-6
The past decade has seen a proliferation of topological materials for both insulators and semimetals in electronic systems and classical waves. Topological semimetals exhibit topologically protected band degeneracies, such as nodal points and nodal lines. Dirac nodal line semimetals (DNLS), which own four-fold line degeneracy, have drawn particular attention. DNLSs have been studied in electronic systems but there is no photonic DNLS. Here in this work, we provide a new mechanism, which is unique for photonic systems to investigate a stringent photonic DNLS. When truncated, the photonic DNLS exhibits double-bowl states (DBS), which comprise two sets of perpendicularly polarized surface states. In sharp contrast to nondegenerate surface states in other photonic systems, here the two sets of surface states are almost degenerate over the whole-spectrum range. The DBS and the bulk Dirac nodal ring (DNR) dispersion along the relevant directions, are experimentally resolved.
Fiber-tip polymer clamped-beam probe for high-sensitivity nanoforce measurements
Mengqiang Zou, Changrui Liao, Shen Liu, Cong Xiong, Cong Zhao, et al.
Published. 2021, 10(9) : 1768-1779 doi: 10.1038/s41377-021-00611-9
Micromanipulation and biological, material science, and medical applications often require to control or measure the forces asserted on small objects. Here, we demonstrate for the first time the microprinting of a novel fiber-tip-polymer clamped-beam probe micro-force sensor for the examination of biological samples. The proposed sensor consists of two bases, a clamped beam, and a force-sensing probe, which were developed using a femtosecond-laser-induced two-photon polymerization (TPP) technique. Based on the finite element method (FEM), the static performance of the structure was simulated to provide the basis for the structural design. A miniature all-fiber micro-force sensor of this type exhibited an ultrahigh force sensitivity of 1.51 nm μN−1, a detection limit of 54.9 nN, and an unambiguous sensor measurement range of ~2.9 mN. The Young's modulus of polydimethylsiloxane, a butterfly feeler, and human hair were successfully measured with the proposed sensor. To the best of our knowledge, this fiber sensor has the smallest force-detection limit in direct contact mode reported to date, comparable to that of an atomic force microscope (AFM). This approach opens new avenues towards the realization of small-footprint AFMs that could be easily adapted for use in outside specialized laboratories. As such, we believe that this device will be beneficial for high-precision biomedical and material science examination, and the proposed fabrication method provides a new route for the next generation of research on complex fiber-integrated polymer devices.
0.75 Gbit/s high-speed classical key distribution with mode-shift keying chaos synchronization of Fabry–Perot lasers
Hua Gao, Anbang Wang, Longsheng Wang, Zhiwei Jia, Yuanyuan Guo, et al.
Published. 2021, 10(9) : 1780-1788 doi: 10.1038/s41377-021-00610-w
High-speed physical key distribution is diligently pursued for secure communication. In this paper, we propose and experimentally demonstrate a scheme of high-speed key distribution using mode-shift keying chaos synchronization between two multi-longitudinal-mode Fabry–Perot lasers commonly driven by a super-luminescent diode. Legitimate users dynamically select one of the longitudinal modes according to private control codes to achieve mode-shift keying chaos synchronization. The two remote chaotic light waveforms are quantized to generate two raw random bit streams, and then those bits corresponding to chaos synchronization are sifted as shared keys by comparing the control codes. In this method, the transition time, i.e., the chaos synchronization recovery time is determined by the rising time of the control codes rather than the laser transition response time, so the key distribution rate is improved greatly. Our experiment achieved a 0.75-Gbit/s key distribution rate with a bit error rate of 3.8 × 10−3 over 160-km fiber transmission with dispersion compensation. The entropy rate of the laser chaos is evaluated as 16 Gbit/s, which determines the ultimate final key rate together with the key generation ratio. It is therefore believed that the method pays a way for Gbit/s physical key distribution.
Quantum superposition demonstrated higher-order topological bound states in the continuum
Yao Wang, Bi-Ye Xie, Yong-Heng Lu, Yi-Jun Chang, Hong-Fei Wang, et al.
Published. 2021, 10(9) : 1789-1796 doi: 10.1038/s41377-021-00612-8
Higher-order topological insulators, as newly found non-trivial materials and structures, possess topological phases beyond the conventional bulk-boundary correspondence. In previous studies, in-gap boundary states such as the corner states were regarded as conclusive evidence for the emergence of higher-order topological insulators. Here, we present an experimental observation of a photonic higher-order topological insulator with corner states embedded into the bulk spectrum, denoted as the higher-order topological bound states in the continuum. Especially, we propose and experimentally demonstrate a new way to identify topological corner states by exciting them separately from the bulk states with photonic quantum superposition states. Our results extend the topological bound states in the continuum into higher-order cases, providing an unprecedented mechanism to achieve robust and localized states in a bulk spectrum. More importantly, our experiments exhibit the advantage of using the time evolution of quantum superposition states to identify topological corner modes, which may shed light on future exploration between quantum dynamics and higher-order topological photonics.
Label-free SARS-CoV-2 detection and classification using phase imaging with computational specificity
Neha Goswami, Yuchen R. He, Yu-Heng Deng, Chamteut Oh, Nahil Sobh, et al.
Published. 2021, 10(9) : 1797-1808 doi: 10.1038/s41377-021-00620-8
Efforts to mitigate the COVID-19 crisis revealed that fast, accurate, and scalable testing is crucial for curbing the current impact and that of future pandemics. We propose an optical method for directly imaging unlabeled viral particles and using deep learning for detection and classification. An ultrasensitive interferometric method was used to image four virus types with nanoscale optical path-length sensitivity. Pairing these data with fluorescence images for ground truth, we trained semantic segmentation models based on U-Net, a particular type of convolutional neural network. The trained network was applied to classify the viruses from the interferometric images only, containing simultaneously SARS-CoV-2, H1N1 (influenza-A virus), HAdV (adenovirus), and ZIKV (Zika virus). Remarkably, due to the nanoscale sensitivity in the input data, the neural network was able to identify SARS-CoV-2 vs. the other viruses with 96% accuracy. The inference time for each image is 60 ms, on a common graphic-processing unit. This approach of directly imaging unlabeled viral particles may provide an extremely fast test, of less than a minute per patient. As the imaging instrument operates on regular glass slides, we envision this method as potentially testing on patient breath condensates. The necessary high throughput can be achieved by translating concepts from digital pathology, where a microscope can scan hundreds of slides automatically.