UWB Rectangular Microstrip Patch Antenna Design in Matching Liquid and Evaluating the Classification Accuracy in Data Mining Using Random Forest Algorithm for Breast Cancer Detection with Microwave

• Published in: Journal of electrical engineering & technology Vol. 14; no. 5; pp. 2127 - 2136
• Main Authors: Avşar Aydin, Emine, Kaya Keleş, Mümine
• Format: Journal Article
• Language: English
• Published: Singapore Springer Singapore 01-09-2019; 대한전기학회
• Subjects: Electrical Engineering; Electrical Machines and Networks; Electronics and Microelectronics; Engineering; Instrumentation; Original Article; Power Electronics; 전기공학; Breast cancer detection; Random forest algorithm; Data mining; Rectangular microstrip patch antenna; Classification
• ISSN: 1975-0102; 2093-7423
• DOI: 10.1007/s42835-019-00205-x

The most common type of cancer for a female is breast cancer in the world. Regular checks and effective-timely treatment are noteworthy parameters for patients’ survival struggle . Against existing imaging methods, microwave imaging method has been considered more powerful and effective method by many researchers. In this paper, comprehensive design equations and parameters of rectangular microstrip patch antenna (RMPA) are given for microwave breast cancer detection. The layered breast model with a spherical tumor that was placed into the fibro-glandular layer was created by using CST Microwave Studio Software, and it was embedded in canola oil to decrease the distorted signals between the transmitting and receiving antennas. The RMPA has a wideband performance from 3 to 18 GHz. The simulation results show that differences in the electric field and reflection coefficients might more efficiently give a possibility to assign the tumor in the breast model. In addition, in this study, the data obtained from these experiments are classified by using the random forest algorithm from the data mining methods. According to the classification result, the random forest algorithm can diagnose breast cancer by classifying the tumor as 94% accuracy.