Design of terahertz antenna to detect lung cancer and classify its stages using machine learning

Abstract The proposed work focuses on the detection of lung cancer using a microstrip patch antenna. A compact microstrip patch antenna operating at ISM band has been designed and placed on healthy as well as cancer affected lung (different stages) phantom. Scatter and other parameters are observed, and data set is created for healthy as well as for each stage of cancerous lung. Variation because of cancer and its stages in the antenna parameters such as electric field, magnetic field, surface current, power flow, current density, power loss density, electrical energy density, and magnetic energy density, return loss, antenna gain, and tumor radius is observed. The created dataset has been further classified to differentiate a healthy lung from a cancerous one and its stages using Random Forest machine learning algorithm with an accuracy of 93.75%.