Advanced brain tumour detection using a hybrid 2D material metasurface sensor with machine learning-enhanced performance
Jacob Wekalao, Hussein A. Elsayed, May Bin‐Jumah, Mostafa R. Abukhadra, Stefano Bellucci, Amuthakkannan Rajakannu, Ahmed Mehaney
Abstract
This study reports a metasurface sensor architecture for brain tumour detection, leveraging the synergistic integration of multiple two-dimensional materials. In this regard, the numerical simulations have been investigated based on the finite element method via COMSOL Multiphysics. The proposed design demonstrates an exceptional sensitivity of 1538 GHz/RIU within the 0.1–0.3 THz spectral window, exhibiting robust response across refractive indices ranging from 1.3333 to 1.4833 RIU. Additionally, further enhancement was achieved through machine learning-driven optimisation utilising one-dimensional convolutional neural networks, attaining an ideal coefficient of determination (R2 = 1.0) for critical parameters including graphene chemical potential, incident angle and resonator dimensions.