An optimized support vector machine classifier to extract abnormal features from breast microwave tomography data

Microwave Tomography (MT) as a new electronic healthcare system tries to measure dielectric properties of tissues inside the breast and helps early breast cancer detection. In this paper, we propose a new classifier to extract tumor information from Microwave Tomography raw data to determine whether the breast needs further diagnosis or not. The proposed method uses grid search algorithm to optimize support vector machine classifier. The results show that optimized SVM can improve measure of performances such as MCC, specificity and sensitivity. The new classifier can be a promising tool to provide preliminary decision support information to physicians for further diagnosis.