This paper proposes a deep-learning-assisted design method for 2-bit coding metasurfaces. This method uses a skip connection module and the idea of an attention mechanism in squeeze-and-excitation networks based on a fully connected network and a convolutional neural network. The accuracy limit of the basic model is further improved. The convergence ability of the model increased nearly 10 times, and the mean-square error loss function converges to 0.000168. The forward prediction accuracy of the deep-learning-assisted model is 98%, and the accuracy of inverse design results is 97%. This approach offers the advantages of an automatic design process, high efficiency, and low computational cost. It can serve users who lack metasurface design experience.
Download full-text PDF |
Source |
|---|---|
| http://dx.doi.org/10.1364/AO.487867 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Deep Learning-Assisted Design of Novel Donor-Acceptor Combinations for Organic Photovoltaic Materials with Enhanced Efficiency.
Adv Mater
December 2024
Shanghai Key Laboratory of Advanced Polymeric Materials, Key Laboratory for Ultrafine Materials of Ministry of Education, Frontiers Science Center for Materiobiology and Dynamic Chemistry, School of Materials Science and Engineering, East China University of Science and Technology, Shanghai, 200237, China.
Designing donor (D) and acceptor (A) structures and discovering promising D-A combinations can effectively improve organic photovoltaic (OPV) device performance. However, to obtain excellent power conversion efficiency (PCE), the trial-and-error structural design in the infinite chemical space is time-consuming and costly. Herein, a deep learning (DL)-assisted design framework for OPV materials is proposed.
View Article and Find Full Text PDFAntigen-independent single-cell circulating tumor cell detection using deep-learning-assisted biolasers.
Biosens Bioelectron
March 2025
Department of Biomedical Engineering, University of Michigan, Ann Arbor, MI, 48109, USA; Center for Wireless Integrated MicroSensing and Systems (WIMS(2)), University of Michigan, Ann Arbor, MI, 48109, USA; Max Harry Weil Institute for Critical Care Research and Innovation, University of Michigan, Ann Arbor, MI, 48109, USA. Electronic address:
Circulating tumor cells (CTCs) in the bloodstream are important biomarkers for clinical prognosis of cancers. Current CTC identification methods are based on immuno-labeling, which depends on the differential expression of specific antigens between the cancer cells and white blood cells. Here we present an antigen-independent CTC detection method utilizing a deep-learning-assisted single-cell biolaser.
View Article and Find Full Text PDFParaAntiProt provides paratope prediction using antibody and protein language models.
Sci Rep
November 2024
Department of Computer Engineering, Sharif University of Technology, Tehran, Iran.
Efficiently predicting the paratope holds immense potential for enhancing antibody design, treating cancers and other serious diseases, and advancing personalized medicine. Although traditional methods are highly accurate, they are often time-consuming, labor-intensive, and reliant on 3D structures, restricting their broader use. On the other hand, machine learning-based methods, besides relying on structural data, entail descriptor computation, consideration of diverse physicochemical properties, and feature engineering.
View Article and Find Full Text PDFDeep Learning Assisted Plasmonic Dark-Field Microscopy for Super-Resolution Label-Free Imaging.
Nano Lett
December 2024
Department of Electrical and Computer Engineering, University of California-San Diego, 9500 Gilman Drive, La Jolla, California 92093, United States.
Article Synopsis
- Dark-field microscopy (DFM) is popular for high-contrast imaging of specimens without labels, but it faces limitations in optical alignment and resolution.
- The new technique, deep-learning assisted plasmonic dark-field microscopy (DAPD), combines plasmonic dark-field microscopy with a deep-learning approach to achieve super-resolution imaging.
- DAPD has shown a resolution improvement of 2.8 times for label-free specimens and offers a compact and enhanced alternative to traditional DFM methods.
Highly air-permeable and dust-holding protective membranes by hierarchical structuring of electroactive poly(lactic acid) micro- and nanofibers.
J Hazard Mater
December 2024
School of Materials Science and Physics, China University of Mining and Technology, Xuzhou 221116, Jiangsu, China; School of Architecture & Design, China University of Mining and Technology, Xuzhou 221116, China; College of Polymer Science and Engineering, State Key Laboratory of Polymer Materials Engineering, Chengdu 610065, China. Electronic address:
The application of biodegradable electrospun poly(lactic acid) (PLA) fibrous membranes (FMs) toward respiratory protection has long been dwarfed by the poor electret effect and short service life. Herein, a micro-on-nano (MON) approach was proposed to fabricate highly electroactive dual-scale poly(lactic acid) (DS-PLA) FMs consisting of inner-layer nanofibers (667 nm) and outer-layer microfibers (1.22 µm).
View Article and Find Full Text PDFWant AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!