High-sensitivity surface plasmon resonance biosensor with gold-based metasurfaces and polynomial regression optimization for early breast cancer detection

Early-stage breast cancer detection is critical for improving diagnostic accuracy and treatment outcomes. This study presents a graphene-enhanced metasurface biosensor designed to provide high sensitivity, predictive precision, and secure data transmission. The sensor architecture consists of a cruciform gold resonator encompassed by an annular silver-coated ring structure deposited on a silicon dioxide substrate, with graphene integration to enhance plasmonic response. Finite element modeling using COMSOL Multiphysics was employed to assess electromagnetic and optical properties, and polynomial regression–based machine learning algorithms were applied to predict operational performance parameters. The biosensor achieved a refractive index sensitivity of 929 GHz·RIU −1 , a figure of merit of 18.571 RIU −1 , and a minimum detectable refractive index change of 0.05 RIU. Quality factors exceeding 17 were maintained across three frequency bands: 0.7–1.0 THz, 1.4–1.5 THz, and 1.62–1.8 THz. The machine learning model delivered complete predictive accuracy for parameter estimation. Additionally, modulation of graphene’s electrochemical potential enabled binary data encoding, supporting encrypted biosensing functionality. Overall, the proposed graphene–metallic metasurface biosensor combines excellent sensitivity, robust predictive capabilities, and secure data handling, positioning it as a promising platform for early-stage breast malignancy detection and as a foundation for next-generation encrypted biosensing technologies.