Design and Evaluation of Tunable Terahertz Metasurface Biosensor for Malaria Detection with Machine learning Optimization Using Artificial Intelligence

Abstract

Malaria continues to be a major global health issue, impacting millions each year and leading to hundreds of thousands of deaths, especially in less developed areas. Timely and precise diagnosis is essential for effective treatment and management of this parasitic illness. This study presents the design and evaluation of a tunable terahertz (THz) metasurface biosensor tailored for malaria detection, integrating plasmonic materials with artificial intelligence. The biosensor employs a multi-layer structure comprising graphene, gold, and silver to leverage surface plasmon resonance effects. Comprehensive electromagnetic simulations and parameter optimization demonstrate the sensor's ability to detect minute changes in malaria parasite concentrations, achieving a peak sensitivity of 429 GHzRIU⁻¹, detection accuracy of 25.6 and a figure of merit of 10.989 RIU_-1. The sensor features tunable elements that allow dynamic performance adjustments. Additionally, the XGBoost machine learning algorithm is harnessed to predict sensor performance across various design parameters, consistently demonstrating maximum R² ranging up to 100%. This fusion of advanced materials, precise engineering, and predictive analytics represents a significant advancement in biosensing technology for malaria detection, offering substantial potential for early and accurate diagnosis.