Cancer is diagnosed in over 1 million new patients every year (Frangioni A. M., 2003). It is often treated with surgical resection of tumors, but cancer is an invasive disease marked by the mutations of cells that continually reproduce and invade, or metastasize to, other organs (National Cancer Institute, U.S. National Institutes of Health, 2009). For a curative surgery to be successful the margins must be clear without cancer cells present in the unresected tissue. Currently, this analysis is done by pathological analysis of small portions of the tissue under a microscope and well after the surgery. Such information, if available to the surgeon while the patient is still in the OR would provide valuable real-time information and impact decision making such as whether or not more tissue needs to be removed.
This thesis describes the use of an electromagnetic (EM) probe to distinguish between cancerous tissue and healthy tissue and to enable imaging surgically excised tissue for quantification of margins. The electromagnetic properties of cancerous and healthy tissues are shown to be different and distinguishable by monitoring changes in mutual inductance between a pair or coils caused by the formation of eddy currents in the tissues. The output voltage and phase of the receiver are monitored using a dual channel lock-in amplifier with the driver coil excited by a 99 kHz, 7 VPP, sawtooth input.
Point-wise voltage measurements are made with the EM probe on surgically excised tissue samples showing that the EM properties of cancer and healthy tissue are indeed different and differentiable. Also, supporting experiments conducted on copper wire loops to show that the voltage and phase of the EM probe signal is affected by eddy current domain size. The EM probe is most sensitive when the eddy current loops are of the same size as the probe’s diameter. It is also observed that the phase response is considerably more sensitive than the voltage response. A numerical model is developed to predict the probe’s response to different excitations as well as to different eddy current loop domain sizes. The model agrees well with experiments for sinusoidal excitations, however, the responses for a sawtooth excitation are severely under-predicted. This is likely due to either frequency dependent impedances or the excitation of the coil near resonance by a harmonic of the sawtooth excitation.
This thesis culminates in using the EM probe to develop a new method for imaging surgically excised tissue. A passive, non-electromagnetically interacting device is developed to allow the probe to traverse convex shaped and flat surfaces. The new imaging technique is used to image paraffin phantoms. The positions of embedded copper artifacts within the phantoms are successfully imaged, with all objects smaller than the effective probe diameter appearing to be of a size on the order of the probe diameter. The new imaging technique shows great promise for quantifying surgical margins in real-time, impacting the decisions of surgical oncologists in the OR and enhancing the curative outcomes of surgical removal of cancer.