PhD defence of Jasmine Boparai - Dielectric Characterization of Skin-Mimicking Phantoms in the Microwave Range: Towards a Diagnostic Tool for Skin Cancer
Abstract
Skin cancer is the most common form of cancer in the world. A report by the skin cancer foundation estimates that one in five Americans will develop skin cancer by the age of 70. Melanoma, which is considered to be the deadliest type of skin cancer, causes the majority of fatalities. The other type is non-melanoma, which is further classified as basal cell carcinoma (BCC) and squamous cell carcinoma (SCC), affects millions of human lives every year.
Early-stage diagnosis of all skin cancers, including melanoma, results in a better prognosis and improved survival rate for the patient. The current method of diagnosis relies on the visual inspection and therefore, implicitly, on the skill and experience of the dermatologist. They inspect the color, size, shape, irregularities in borders, bleeding in lesions, or if lesion is raised or hard to touch. In case of a suspicious lesion, a biopsy is required, so that the sample can be tested in a cell pathology lab. These methods of visual inspection are vulnerable to human error and vary in interpretation according to the dermatologist’s experience.
As a result, a significant research effort is underway to develop diagnostic tools for skin cancer. Due to the reported inherent difference in electrical properties between healthy and cancerous tissues, microwave-based techniques hold promise for skin cancer detection.
Therefore, the main focus of this research is systematic design and development of experimental tissue-mimicking phantoms which are extensively required to check the viability and performance of the designed tools in a controlled laboratory environment and prior to in vivo trials. The thesis describes the mathematical modeling and construction of series of phantom models and comprehensive dielectric property characterization of skin and skin cancer tissue-mimicking phantoms by measuring their reflectance properties in the microwave region. In addition, the work contributes with the estimation of the Cole-Cole function parameters for experimental values that gives the best fit with the computed results. Sensing depth of the probe is quantified by realizing realistic phantoms of different thicknesses. Lastly, the dielectric probe was numerically simulated to estimate the specific absorption rate (SAR) in order to confirm the compliance with the safety standards.