Ultra-sensitive terahertz metasurface biosensor with MXene–BP–graphene architecture for AI-assisted early cancer detection

Cancer remains a leading global health threat, with early detection critical to improving survival rates. However, conventional diagnostic methods suffer from limitations in sensitivity, specificity, and accessibility, particularly in low- and middle-income countries. This study presents a next-generation metasurface biosensor integrating MXene (Ti3C2Tx), black phosphorus, graphene, and noble metals (Au and Ag) for high-precision cancer detection in the terahertz (THz) regime. Numerical simulations and experimental validations demonstrate the sensor’s exceptional performance: a maximum sensitivity of 2000 GHz/refractive index unit (RIU), a strong linear correlation between resonance frequency and refractive index (R2 = 0.952 76), and a figure of merit up to 29.851 RIU−1. Parametric optimizations, including graphene chemical potential (0.1–0.9 eV) and angle of incidence (0°–80°), further enhance spectral tunability and absorption efficiency. Compared to state-of-the-art biosensors, this design outperforms in sensitivity, detection accuracy (15), and adaptability for point-of-care applications. Integration with machine learning confirms robust agreement between experimental and theoretical data (R2 = 0.96–0.97), highlighting its potential to revolutionize early cancer diagnostics and reduce global health disparities.