Multiscale diffractive U-Net: a robust all-optical deep learning framework modeled with sampling and skip connections

Fan-in/fan-out for heterogeneous 19-core fibers based on metasurfaces with nonuniform phase plates

In space-division-multiplexed transmission systems, it is essential to realize fan-in/fan-out devices that connect the cores between multicore fibers and single-mode fibers. In this Letter, we propose a metasurface-based fan-in/fan-out device with nonuniform phase plates for heterogeneous 19-core fibers across the full C band. Our results show that an average insertion loss of 0.85 dB and a maximum crosstalk of -25.5 dB can be achieved at 1550 nm. Across the C band, the insertion loss and crosstalk are better than 2.78 dB and -19.96 dB, respectively. The proposed concept can flexibly handle various fiber configurations without additional complexity.

Pulmonary Nodule Detection and Classification Using All-Optical Deep Diffractive Neural Network

A deep diffractive neural network (D2NN) is a fast optical computing structure that has been widely used in image classification, logical operations, and other fields. Computed tomography (CT) imaging is a reliable method for detecting and analyzing pulmonary nodules. In this paper, we propose using an all-optical D2NN for pulmonary nodule detection and classification based on CT imaging for lung cancer. The network was trained based on the LIDC-IDRI dataset, and the performance was evaluated on a test set. For pulmonary nodule detection, the existence of nodules scanned from CT images were estimated with two-class classification based on the network, achieving a recall rate of 91.08% from the test set. For pulmonary nodule classification, benign and malignant nodules were also classified with two-class classification with an accuracy of 76.77% and an area under the curve (AUC) value of 0.8292. Our numerical simulations show the possibility of using optical neural networks for fast medical image processing and aided diagnosis.