Square-slotted metasurface optical sensor based on graphene material for efficient detection of brain tumor using machine learning

Abstract

This study presents a non-invasive sensor design for identifying brain tumors, leveraging on advanced machine learning techniques to enhance diagnostic effectiveness. Key achievements of the sensor design include an optimal sensitivity of 3076 GHzRIU⁻¹. The sensor features a simple design and exemplifies an impressive figure of merit (FOM) of 42.137 RIU⁻¹, indicating its high responsiveness. Additionally, the sensor’s performance is characterized by a quality factor (Q) ranging from 12.139 to 12.611, along with a notable detection limit of 0.032, making it highly effective for early detection applications. By integrating the Random Forest machine learning algorithm, the diagnostic accuracy of the sensor is significantly enhanced, ensuring precise and reliable results. This proposed sensor represents a major advancement in non-invasive diagnostic technologies, offering a promising approach for early detection of neurological diseases.