A machine learning approach for enhanced early detection of cancerous cells using optimized metamaterial graphene biosensors

In this article the design of a metamaterial graphene biosensor having a double L-shaped structure to detect cancerous cells is investigated and presented. The structural parameters of the proposed biosensor such as the thickness of the double L-shape, the thickness of the graphene layer, and the gap between the double L-shape were optimized to enhance the sensitivity and figure of merit and allow accurate detection of cancerous cells. To obtain best performance, accurate prediction, and low errors, machine learning algorithms for function approximation were used. Machine learning methods accelerate the bio sensing operation and help in predicting the early stage detection of cancerous cells in the body. This approach minimizes human intervention and optimizes the output from the sensing platform, making it a potential biosensor for early detection of cancerous cells. The obtained results demonstrated that the proposed biosensor achieves large resonant wavelength shifts leading to a significant increase in its characteristic parameters such as sensitivity, figure of merit, and limit of detection.