Effective tumor feature extraction for smart phone based microwave tomography breast cancer screening

Mobile Microwave Tomography (MMT) is a new alternative technique to detect breast cancer using smart phone based electronic healthcare system. In this paper, we propose a new solution to extract tumor information from MMT raw data for early breast cancer screening. MMT reflects water contents of breast tissue by measuring their electrical properties and sends permittivity and conductivity raw data to processing servers in hospital via WiFi or 3G/4G networks. In this approach we investigate three different sets of MMT tumor features and perform a comparative study to investigate their set of accuracy measurements for each classification. Through extensive empirical study of the classification results, we have identified the following six parameters as useful to extract tumor information: average permittivity of healthy tissue (APHT), average permittivity of probable tumor area (APPTA), maximum and minimum values of permittivity of probable tumor area (MaxPPTA, and MinPPTA), and energy values of healthy tissue (EVHT) and probable tumor area (EVPA).