In this research, we present a multilayer metasurface sensor design integrating graphene, MXene, black phosphorus, and gold for the ultrasensitive detection of brain tumor biomarkers in liquid biopsy samples. The hierarchical structure consists of a MXene-coated rectangular resonator, a black phosphorus-coated square resonator, a gold-coated circular ring, and a graphene-based circular substrate. This architecture was systematically optimized through comprehensive numerical simulations using COMSOL Multiphysics 6.3, integrated with machine learning frameworks. The proposed sensor demonstrates an outstanding sensitivity of 2308 GHz/RIU across a physiologically relevant refractive index range (1.3333–1.4833), significantly outperforming current state-of-the-art devices. Performance analysis identifies an optimal sensing regime at RI = 1.3425, achieving a figure of merit of 20.79 RIU−1 and a detection limit as low as 0.079 RIU. Detailed investigations of the transmission spectra under varying graphene chemical potentials (0.1–0.9 eV), incident angles (0°–80°), and geometric modifications of the resonators reveal highly tunable sensing behavior. Furthermore, Random Forest Regression models achieve predictive accuracies of 85%–100%, enabling reliable estimation of sensor performance across diverse operating conditions. Collectively, these results establish a solid foundation for employing advanced 2D material–based metasurfaces in minimally invasive and early-stage brain tumor diagnostics, thereby advancing the capabilities of next-generation liquid biopsy technologies.
Wekalao et al. (Sun,) studied this question.