Advancing microwave ablation applicators: integrating computational modeling of a graphene-based applicator with machine learning for ablation zone prediction

This research focuses on the design and optimization of a graphene-based microwave ablation applicator (MWA) for tumor treatment. The unique properties of graphene, such as high impedance at microwave frequency and tunable Fermi energy, make it an ideal candidate for inducing charge carriers through an electric field or chemical doping. The study further employs machine learning techniques, including support vector regression (SVR) and artificial neural networks (ANN) to predict the ablation zone. The applicator design incorporates a helix antenna element connected to a coaxial cable working at 2.45 GHz with a graphene sheet and T-ring attached to the outer conductor. The challenge of achieving a zero-Fermi energy is addressed by introducing defects to increase surface resistivity, resulting in an impedance close to the required value. The results show that the graphene-based applicator enhances the ablation zone, leading to efficient and controlled tumor treatment. To predict the ablation zone accurately, the study employs machine learning techniques, including SVR and ANN utilizing Taguchi method to reduce computational complexity. Large and round ablation zone is achieved using novel applicator. Further, the performance metrics, including root-mean-squared error and coefficient of determination (R²), are utilized to evaluate the predictive capabilities of the model and found to be optimum. The research demonstrates the potential of graphene in improving MWA treatment and highlights the importance of machine learning in optimizing MWA and predicting treatment outcomes.