Optimization of graphene-based biosensor design for haemoglobin detection using the gradient boosting algorithm for behaviour prediction

This study presents an advanced biosensor based on metasurfaces for hemoglobin detection. The proposed sensor design integrates graphene with gold and silver, leveraging their exceptional optical properties and ability to support surface plasmon resonances. The metasurface-based architecture enhances interactions between the sensor and hemoglobin biomolecules, resulting in improved sensitivity and other performance parameters. Extensive optimization of the design parameters, including resonator dimensions and graphene chemical potential, was conducted to achieve an optimized sensor design. The sensor exhibits exceptional characteristics, including a peak sensitivity of 267 GHzRIU −1, a quality factor of 10.457, and a sensor resolution of 0.094, among other remarkable performance metrics. To streamline the optimization process and reduce computational complexity, the Gradient Boosting Algorithm (GBoost) is integrated into this study for behaviour prediction. The GBoost model demonstrates impressive performance, including an optimal coefficient of determination (R2) score of 1.0 for all cases considered, indicating perfect predictive accuracy within the model’s scope. These outstanding results suggest the significant potential of the proposed biosensor for rapid and precise blood testing, as well as monitoring medical conditions such as anaemia, by enabling early and accurate detection of hemoglobin levels. The sensor’s high-performance metrics, coupled with its simple design, represent a substantial advancement in the field of biosensing technology.