Advanced FEM–NN–GA-based inverse design of a high-sensitivity plasmonic sensor for lung cancer diagnostics

View at publisher Summarise with VIRAT
Authors: Ahmed S. Hatem, Aqilah Baseri Huddin, Norhana Arsad
Cited by: 0
References: 42

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

This study presents an optimized plasmonic refractive-index sensor based on a metal–insulator–metal (MIM) waveguide coupled with a truncated-elliptical resonator. The device is optimized using a hybrid inverse-design framework combining finite element method (FEM) simulations, neural networks (NN), and a genetic algorithm (GA). A dataset generated from COMSOL simulations was used to train the NN to predict the nonlinear relationship between geometric parameters and optical responses, including resonance wavelength, transmission efficiency ( Te ), and FWHM. The NN-optimized design improves performance compared with the conventional structure, achieving a sensitivity of 909 nm/RIU, FWHM of 13 nm, and figure of merit (FoM) of 70. The hybrid NN–GA approach further enhances sensitivity to 1136 nm/RIU. For lung cancer detection based on refractive-index differences between healthy lung cells ( n =1.37) and NCI-H1299 cancer cells ( n =1.39), the optimized NN–GA sensor achieves a sensitivity of approximately 1550 nm/RIU and FoM of 97, suggesting potential applicability in label-free biomedical sensing, pending further experimental validation.

Fields of Study & Subjects

Plasmonic and Surface Plasmon Research Metamaterials and Metasurfaces Applications Gold and Silver Nanoparticles Synthesis and Applications

Publication details

Journal / source
Journal of the Optical Society of America B
Volume
43
Issue
7
Pages
1421–1421
Publication date
23 March 2026
Language
EN