This study develops a compact, cost-effective AI-enhanced dynamic light scattering system, achieving 99% accuracy and 80% faster analysis through deep learning, significantly improving accessibility and efficiency in microparticle characterization.

This study introduces a machine learning-driven optimization framework for Shadow Sphere Lithography, significantly enhancing the design of chiral nanostructures by efficiently navigating vast parameter spaces, reducing computational costs, and doubling chiral response, advancing applications in nanophotonics and plasmonics.

This study optimizes plasmonic hydrogen sensors using heterogeneous multilayer designs, with the Pd/Ag/Pd configuration achieving significantly improved sensing performance, while quantitative analysis and FDTD simulations provide insights for future sensor advancements.

This study enhances plasmonic hydrogen sensors using phase space reconstruction and convolutional neural networks, achieving high accuracy, faster response, improved signal quality, and lower detection limits, enabling advanced hydrogen monitoring and intelligent sensing.