1. Surface plasmon resonance sensor based on MXene coated PCF for detecting the cancer cells with machine learning approach.
- Author
-
Kumar, Amit, Verma, Pankaj, and Jindal, Poonam
- Subjects
- *
SURFACE plasmon resonance , *CANCER cells , *MACHINE learning , *PHOTONIC crystal fibers , *FINITE element method , *DETECTORS , *THIN films - Abstract
This article presents a highly sensitive gold/Ti 3 C 2 T x coated photonic crystal fiber (PCF)-based surface plasmon resonance (SPR) sensor for cancer cells detection. The hybrid gold/Ti 3 C 2 T x layer is coated circularly over PCF for occurrence of SPR by coupled mode theory. The simulations and numerical analysis are done by the finite element method. The resonance wavelength shift is measured between normal and cancerous cell samples to estimate the wavelength sensitivity and resolution. In results, the maximum wavelength sensitivity for breast cancer MCF-7 cells is achieved as 10,714 nm/RIU in x-polarization mode and 13,071 nm/RIU in y-polarization mode with very low resolution of 10−6 RIU. The impact of the MXene (Ti 3 C 2 T x) thin film on sensitivity is also observed. In addition, the optimization of parametric variation is analyzed with machine learning technique with low mean squared error of 0.01525. <2% error is obtained in sensitivity analysis with multilayer perceptron regressor. Due to the advanced results and machine learning application, this sensor can be used as a fast, efficient and low-cost cancer disease detection device. [Display omitted] • A gold/Ti 3 C 2 T x coated PCF based SPR sensor is presented and numerically analyzed for detecting various cancer cells. • Vertical confinement path is used to transform the energy from core mode to SPR mode with external sensing approach. • The normal to cancerous cell is detected by measuring the corresponding resonance wavelength shift. • The Multilayer Perceptron regressor has been employed on ANN model for predicting the output of PCF-SPR sensor. [ABSTRACT FROM AUTHOR]
- Published
- 2023
- Full Text
- View/download PDF