Date of Graduation

7-2021

Document Type

Dissertation

Degree Name

Doctor of Philosophy in Engineering (PhD)

Degree Level

Graduate

Department

Electrical Engineering

Advisor

Magda El-Shenawee

Committee Member

Jingxian Wu

Second Committee Member

Narasimhan Rajaram

Third Committee Member

Zhong Chen

Keywords

Breast Cancer, Free-Space millimeter-wave measurements, Material Characterization, Microwave and millimeter-wave measurements, Pathology imaging, Terahertz Imaging and Spectroscopy

Abstract

The goal of this research is to characterize the electromagnetic properties of biological and non-biological materials at terahertz (THz), millimeter-wave, and microwave frequency bands. The biological specimens are measured using the THz imaging and spectroscopy system, whereas the non-biological materials are measured using the microwave and millimeter-wave free-space system. These facilities are located in the Engineering Research Center at the University of Arkansas. The THz imaging system (TPS 3000) uses a Ti-Sapphire laser directed on the photoconductive antennas to generate a THz time domain pulse. Upon using the Fourier Transform, the spectrum of the pulsed THz signal includes frequencies from 0.1 THz to 4 THz. On the other hand, the free space system uses a PNA network analyzer to produce a signal at frequencies ranging from 10 MHz to 110 GHz. For the biological specimens, the research focused on imaging the freshly excised breast tumors to detect the cancer on the margins using THz radiation. The tumor margin assessment depends on the THz contrast between cancer, collagen, and fat tissues in the tumor. Three models of breast tumors are investigated in this research: humans, mice (xenograft and transgenic), and Sprague Dawley rats. The results showed good differentiation between the cancerous and non-cancerous tissues in all three models. In addition, an excellent distinction was observed between cancer, collagen, and fat in the formalin-fixed paraffin-embedded (FFPE) block tissue with ~ 90-95% correlation with the pathology images. Furthermore, the FFPE ductal carcinoma in situ (DCIS) tumors are investigated, also using the THz imaging. The THz images of the DCIS samples are compared with those of the FFPE invasive ductal carcinoma (IDC) specimens. The results demonstrated that THz electric field reflection from the IDC were higher than that from the collagen, DCIS, and then the fat tissue region. Furthermore, a pilot study is conducted to investigate the effect of optical clearance (e.g., glycerol solution) on THz images of freshly excised tumors. The results showed that the glycerol reduced the absorption coefficients of pre-treated tissues compared with those of untreated tissues. For the non-biological materials, the research focuses on characterizing highly conductive non-magnetic radar absorbing materials (RAM) for the automotive industry. The ingredients of material components in the RAM samples are unrevealed under a non-disclosure agreement (NDA). The material characterization involves the extraction of the complex relative permittivity utilizing the transmission measurement data obtained at the K-band (18 GHz to 26.5 GHz) and the W-band (75 GHz to 110 GHz). The measurements are obtained using the free-space conical horn antenna system. A transmission line based extraction model is implemented, and the results are validated with the experimental measurements of the S-parameters. The maximum error reported between the measured and the calculated S-parameters was less than 1 dB. In conclusion, the THz imaging of breast cancer tumors presents a potential margin assessment of other solid tumors, and the microwave, millimeter-wave, and THz spectroscopy of materials demonstrate a potential application in the fifth and sixth generations of wireless communications.

Share

COinS