Inverse design of plasmonic nanoantenna using generative adversarial network.

The local surface plasmon resonance (LSPR) effect has been widely used in various nanophotonic applications. However, because the LSPR effect is highly sensitive to the structure and geometry, it is desirable to efficiently search viable geometries for predefined local field enhancement spectrum. Herein we present a generative adversarial network-based LSPR nanoantenna design scheme. By encoding the antenna structure information into an red-green-blue (RGB) color image, the corresponding nanoantenna structure can be inverse-designed to achieve the required enhancement spectrum of the local field. The proposed scheme can accurately offer the multiple geometry layout for the customized specific spectrum in seconds, which could be beneficial for fast design and fabrication of plasmonic nanoantenna.

Nearly dispersionless multicolor metasurface beam deflector for near eye display designed by a physics-driven deep neural network.

Dispersion is one of the most important issues in see-through near eye displays with waveguide technology. In particular, the nanophotonics design is challenging but demanding. In this paper, we propose a design method for a multilayer achromatic metasurface structure for near eye display application by a physics-driven generative neural network. Two in-coupling metagratings under different projector illuminations are presented and numerically verified with the absolute diffraction efficiency over 89%. A beam splitter, which provides a balance between compactness and visual comfort in a single-projector-binocular display, is also designed. Finally, we apply this method to an out-coupling metasurface with the capability of enlarging the visible region by threefold.

Prediction of cancer-specific survival and overall survival in middle-aged and older patients with rectal adenocarcinoma using a nomogram model.

OBJECTIVE: To develop a new nomogram tool for predicting survival in middle-aged and elderly patients with rectal adenocarcinoma. METHODS: A total of 6,116 patients were randomly assigned in a 7:3 ratio to training and validation cohorts. Univariate and multivariate Cox proportional hazards regression analyses were used to identify independent prognostic factors associated with overall survival (OS) and cancer-specific survival (CSS) in the training set, and two nomogram prognostic models were constructed. The validity, accuracy, discrimination, predictive ability, and clinical utility of the models were assessed based on the concordance index (C-index), area under the receiver operating characteristics (ROC) curve, time-dependent area under the ROC curve (AUC), Kaplan-Meier survival curve, and decision curve analyses. RESULTS: Predictors of OS and CSS were identified, and nomograms were successfully constructed. The calibration discrimination for both the OS and CSS nomogram prediction models was good (C-index: 0.763 and 0.787, respectively). The AUC showed excellent predictive performance, and the calibration curve exhibited significant predictive power for both nomograms. The time-dependent AUC showed that the predictive ability of the predictor-based nomogram was better than that of the TNM stage. The nomograms successfully discriminated high-, medium-, and low-risk patients for all-cause and cancer-specific mortality. The decision curve demonstrated that the nomograms are useful with respect to good decision power. CONCLUSION: Our nomogram survival prediction models may aid in evaluating the prognosis of middle-aged and older patients with rectal adenocarcinoma and guiding the selection of the clinical treatment measures.

Multiple Dirac points by high-order photonic bands in plasmonic-dielectric superlattices.

The emergence of Dirac points (DPs) characterizes the topological phase transition and the gapless interface states in composite metal-dielectric metamaterials. In this work, we study a kind of compound plasmonic-dielectric periodic structure (PDPS) which sustains both plasmonic modes and multiple photonic modes. The structure has primitive cell consisting of four layers made from triple constituent components. Due to the generalized Su-Schrieffer-Heeger, DPs can emerge at the Brillouin zone center. More specifically, in weak plasmonic-photonic mode interaction regime, multiple DPs would emerge at the Brillouin zone center and edge due to the band folding, from the perspective of general effective medium. From the rigorous field analysis, the origin of these DPs is clearly demonstrated. These interleaved DPs behave as the intermediate transitions of the surface impedance for the PDPS and raise fully spanned topological interface states originated from 0 to 2nd-order photonic bands in the PDPS. The cases of combining our PDPS with either a plasmonic or dielectric homogenous medium are presented.

Frequency selective topological edge wave routing in meta-structures made of cylinders.

The propagation direction of edge states is essentially related to the band topology invariant of the constituent structures and the momentum of the excitation source. However, it is difficult to control the propagation path when the chirality of the excitation source and the boundary structures are determined. Here, we study a frequency selective waveguide structure based on photonic crystals with different topological invariant characterized by bulk polarization. By designing different types of interface made from spatially arranged dielectric rods, distinct topological edge states could be realized at different frequencies in the band gap. Therefore, we can construct a meta-structure in which the wave guiding path can be switched by the excitation frequency. Our study provides an alternative approach to designing topological devices such as frequency dependent optical waveguides and frequency division devices.

Designing metal-dielectric nanoantenna for unidirectional scattering via Bayesian optimization.

Recently, using various intelligent approaches to achieve the efficient inverse design of photonic nanostructures with predefined and appropriate functionalities has attracted considerable attention. We propose a method to design subwavelength metal-dielectric nanoantennas and optimize the scattering directionality using a Bayesian optimization approach. The nanoantennas consisted of three gold disks separated by two dielectric layers. The geometrical parameters were optimized in an intelligent and fully automatic process. We showed that with the aid of the machine learning method, strong forward scattering or backward scattering at a specific wavelength could be efficiently achieved. We further showed that unidirectional scattering in opposite directions at two separate wavelengths can be designed. Moreover, it is possible to exchange the forward and backward directionality at two target wavelengths. The multipole decomposition approach was applied to analyze the multipole moments of the scattering field up to the third order. In the optimized unidirectional nanoantennas the electric and magnetic dipole moments satisfied the Kerker or anti-Kerker conditions at the wavelengths of interest. Our results demonstrated the possibility of automatically designing nanoantennas for specific applications via a machine learning scheme.

Exploring optical resonances of nanoparticles excited by optical Skyrmion lattices.

Recently, optical Skyrmion lattices (OSLs) have been realized in evanescent electromagnetic fields. OSLs possess topologically stable field configurations, which promise many optics and photonics applications. Here, we demonstrate that OSLs can serve as versatile structured optical near-fields to assist with studies of a variety of photonic modes in nanoparticles. We firstly show that OSL is capable of selectively exciting electric and magnetic multipole modes by placing a nanoparticle at different positions in the lattice. We then disclose that OSLs can efficiently excite some intriguing resonant modes, including toroidal and plasmonic dark modes, in dielectric or metal nanoparticles. Our results may enhance understanding of the interaction between OSLs and nanoparticles and find applications associated with precise control over resonant modes in nanostructures.

The X-like shaped spatiotemporal structure of the biphoton entangled state in a cold two-level atomic ensemble.

We study the spatiotemporal structure of the biphoton entangled state generated by the four-wave mixing (FWM) process in a cold two-level atomic ensemble. We analyze, for the first time, the X-like shaped structure of the biphoton entangled state and the geometry of the biphoton correlation for different lengths and densities of the cold atomic ensemble. The propagation equations of the photon pairs generated from FWM process are derived in a spatiotemporal framework. By means of the input-output relations of the propagation equations, the biphoton amplitude function is obtained in a spatiotemporal domain. In the given frequency range, the biphoton amplitude displays an X-like shaped geometry, nonfactorizable in the space-time domain. Such an X-like shaped spatiotemporal structure is caused by the phase matching and the FWM gain. The former leads to the X-like shaped envelope of the biphoton correlation, while the latter gives rise to the oscillations around the X-like shaped envelope.