Ahmet Can Sabuncu, Jeanne Shen, Sean Doyle
The field of cancer diagnostics has always been one of the most complex and challenging areas in biomedical research, and there has been an increasing demand for more advanced clinical diagnostic equipment over the past four decades. In this thesis, a custom-made coaxial bioimpedance sensor was used, in combination with computer-aided pattern recognition tools, to identify cancer in tissue samples obtained from the formalin-fixed kidney of a 60 year-old male patient. Impedance data was collected using the coaxial sensor at 401 logarithmically-spaced frequency points ranging from 10 kHz to 100 MHz. Principle Component Analysis and Naive Bayes Classification techniques were employed to test whether bioimpedance could discriminate between the cancer and non-cancer tissue within the specimen. The classification model was trained using measurement data from cancer and non-cancer sections of the sample. The models were then tested using cross validation techniques. It was concluded that the Naive Bayes classifier could discriminate cancer from normal samples using capacitance and conductance data.
Ahmet Can Sabuncu
Number of Pages
Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial 4.0 License
Shams, Shahriar, "Cancer Diagnosis by Bioimpedance Spectroscopy and Computer-Assisted Pattern Recognition" (2017). Mechanical Engineering Research Theses and Dissertations. 1.